diff options
author | Lennart Weller <lhw@ring0.de> | 2017-07-27 09:55:55 +0000 |
---|---|---|
committer | Lennart Weller <lhw@ring0.de> | 2017-07-27 09:55:55 +0000 |
commit | 1413c8953bf9ab447967fedb6246c03afa3f788e (patch) | |
tree | 9bde7ef5a8010d08c0723badb22a24bca8926834 /python.d | |
parent | Release v. 1.6.0+dfsg-3 to Unstable (diff) | |
parent | New upstream version 1.7.0+dfsg (diff) | |
download | netdata-1413c8953bf9ab447967fedb6246c03afa3f788e.tar.xz netdata-1413c8953bf9ab447967fedb6246c03afa3f788e.zip |
Updated version 1.7.0+dfsg from 'upstream/1.7.0+dfsg'
with Debian dir e101ee6549c6786aab4b3f5c4f9bbb7638f17e34
Diffstat (limited to 'python.d')
-rw-r--r-- | python.d/Makefile.am | 4 | ||||
-rw-r--r-- | python.d/Makefile.in | 144 | ||||
-rw-r--r-- | python.d/README.md | 170 | ||||
-rw-r--r-- | python.d/apache.chart.py | 72 | ||||
-rw-r--r-- | python.d/bind_rndc.chart.py | 319 | ||||
-rw-r--r-- | python.d/cpufreq.chart.py | 2 | ||||
-rw-r--r-- | python.d/dns_query_time.chart.py | 135 | ||||
-rw-r--r-- | python.d/elasticsearch.chart.py | 198 | ||||
-rw-r--r-- | python.d/fail2ban.chart.py | 267 | ||||
-rw-r--r-- | python.d/go_expvar.chart.py | 228 | ||||
-rw-r--r-- | python.d/haproxy.chart.py | 254 | ||||
-rw-r--r-- | python.d/isc_dhcpd.chart.py | 190 | ||||
-rw-r--r-- | python.d/mdstat.chart.py | 167 | ||||
-rw-r--r-- | python.d/mongodb.chart.py | 53 | ||||
-rw-r--r-- | python.d/mysql.chart.py | 10 | ||||
-rw-r--r-- | python.d/ovpn_status_log.chart.py | 94 | ||||
-rw-r--r-- | python.d/postgres.chart.py | 12 | ||||
-rw-r--r-- | python.d/python_modules/base.py | 211 | ||||
-rw-r--r-- | python.d/rabbitmq.chart.py | 187 | ||||
-rw-r--r-- | python.d/redis.chart.py | 5 | ||||
-rw-r--r-- | python.d/samba.chart.py | 124 | ||||
-rw-r--r-- | python.d/smartd_log.chart.py | 13 | ||||
-rw-r--r-- | python.d/web_log.chart.py | 1179 |
23 files changed, 2844 insertions, 1194 deletions
diff --git a/python.d/Makefile.am b/python.d/Makefile.am index bfe28ff2..43f25cff 100644 --- a/python.d/Makefile.am +++ b/python.d/Makefile.am @@ -18,12 +18,14 @@ dist_python_DATA = \ bind_rndc.chart.py \ cpufreq.chart.py \ cpuidle.chart.py \ + dns_query_time.chart.py \ dovecot.chart.py \ elasticsearch.chart.py \ example.chart.py \ exim.chart.py \ fail2ban.chart.py \ freeradius.chart.py \ + go_expvar.chart.py \ haproxy.chart.py \ hddtemp.chart.py \ ipfs.chart.py \ @@ -38,8 +40,10 @@ dist_python_DATA = \ phpfpm.chart.py \ postfix.chart.py \ postgres.chart.py \ + rabbitmq.chart.py \ redis.chart.py \ retroshare.chart.py \ + samba.chart.py \ sensors.chart.py \ squid.chart.py \ smartd_log.chart.py \ diff --git a/python.d/Makefile.in b/python.d/Makefile.in index 9b784668..33efd42d 100644 --- a/python.d/Makefile.in +++ b/python.d/Makefile.in @@ -1,9 +1,8 @@ -# Makefile.in generated by automake 1.11.3 from Makefile.am. +# Makefile.in generated by automake 1.14.1 from Makefile.am. # @configure_input@ -# Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, -# 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011 Free Software -# Foundation, Inc. +# Copyright (C) 1994-2013 Free Software Foundation, Inc. + # This Makefile.in is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. @@ -17,6 +16,51 @@ VPATH = @srcdir@ +am__is_gnu_make = test -n '$(MAKEFILE_LIST)' && test -n '$(MAKELEVEL)' +am__make_running_with_option = \ + case $${target_option-} in \ + ?) ;; \ + *) echo "am__make_running_with_option: internal error: invalid" \ + "target option '$${target_option-}' specified" >&2; \ + exit 1;; \ + esac; \ + has_opt=no; \ + sane_makeflags=$$MAKEFLAGS; \ + if $(am__is_gnu_make); then \ + sane_makeflags=$$MFLAGS; \ + else \ + case $$MAKEFLAGS in \ + *\\[\ \ ]*) \ + bs=\\; \ + sane_makeflags=`printf '%s\n' "$$MAKEFLAGS" \ + | sed "s/$$bs$$bs[$$bs $$bs ]*//g"`;; \ + esac; \ + fi; \ + skip_next=no; \ + strip_trailopt () \ + { \ + flg=`printf '%s\n' "$$flg" | sed "s/$$1.*$$//"`; \ + }; \ + for flg in $$sane_makeflags; do \ + test $$skip_next = yes && { skip_next=no; continue; }; \ + case $$flg in \ + *=*|--*) continue;; \ + -*I) strip_trailopt 'I'; skip_next=yes;; \ + -*I?*) strip_trailopt 'I';; \ + -*O) strip_trailopt 'O'; skip_next=yes;; \ + -*O?*) strip_trailopt 'O';; \ + -*l) strip_trailopt 'l'; skip_next=yes;; \ + -*l?*) strip_trailopt 'l';; \ + -[dEDm]) skip_next=yes;; \ + -[JT]) skip_next=yes;; \ + esac; \ + case $$flg in \ + *$$target_option*) has_opt=yes; break;; \ + esac; \ + done; \ + test $$has_opt = yes +am__make_dryrun = (target_option=n; $(am__make_running_with_option)) +am__make_keepgoing = (target_option=k; $(am__make_running_with_option)) pkgdatadir = $(datadir)/@PACKAGE@ pkgincludedir = $(includedir)/@PACKAGE@ pkglibdir = $(libdir)/@PACKAGE@ @@ -35,10 +79,10 @@ PRE_UNINSTALL = : POST_UNINSTALL = : build_triplet = @build@ host_triplet = @host@ -DIST_COMMON = $(dist_python_DATA) $(dist_python_SCRIPTS) \ - $(dist_pythonmodules_DATA) $(dist_pythonyaml2_DATA) \ - $(dist_pythonyaml3_DATA) $(srcdir)/Makefile.am \ - $(srcdir)/Makefile.in $(top_srcdir)/build/subst.inc +DIST_COMMON = $(top_srcdir)/build/subst.inc $(srcdir)/Makefile.in \ + $(srcdir)/Makefile.am $(dist_python_SCRIPTS) \ + $(dist_python_DATA) $(dist_pythonmodules_DATA) \ + $(dist_pythonyaml2_DATA) $(dist_pythonyaml3_DATA) subdir = python.d ACLOCAL_M4 = $(top_srcdir)/aclocal.m4 am__aclocal_m4_deps = $(top_srcdir)/m4/ax_c___atomic.m4 \ @@ -86,13 +130,32 @@ am__installdirs = "$(DESTDIR)$(pythondir)" "$(DESTDIR)$(pythondir)" \ "$(DESTDIR)$(pythonmodulesdir)" "$(DESTDIR)$(pythonyaml2dir)" \ "$(DESTDIR)$(pythonyaml3dir)" SCRIPTS = $(dist_python_SCRIPTS) +AM_V_P = $(am__v_P_@AM_V@) +am__v_P_ = $(am__v_P_@AM_DEFAULT_V@) +am__v_P_0 = false +am__v_P_1 = : +AM_V_GEN = $(am__v_GEN_@AM_V@) +am__v_GEN_ = $(am__v_GEN_@AM_DEFAULT_V@) +am__v_GEN_0 = @echo " GEN " $@; +am__v_GEN_1 = +AM_V_at = $(am__v_at_@AM_V@) +am__v_at_ = $(am__v_at_@AM_DEFAULT_V@) +am__v_at_0 = @ +am__v_at_1 = SOURCES = DIST_SOURCES = +am__can_run_installinfo = \ + case $$AM_UPDATE_INFO_DIR in \ + n|no|NO) false;; \ + *) (install-info --version) >/dev/null 2>&1;; \ + esac DATA = $(dist_python_DATA) $(dist_pythonmodules_DATA) \ $(dist_pythonyaml2_DATA) $(dist_pythonyaml3_DATA) +am__tagged_files = $(HEADERS) $(SOURCES) $(TAGS_FILES) $(LISP) DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST) ACLOCAL = @ACLOCAL@ AMTAR = @AMTAR@ +AM_DEFAULT_VERBOSITY = @AM_DEFAULT_VERBOSITY@ AUTOCONF = @AUTOCONF@ AUTOHEADER = @AUTOHEADER@ AUTOMAKE = @AUTOMAKE@ @@ -251,12 +314,14 @@ dist_python_DATA = \ bind_rndc.chart.py \ cpufreq.chart.py \ cpuidle.chart.py \ + dns_query_time.chart.py \ dovecot.chart.py \ elasticsearch.chart.py \ example.chart.py \ exim.chart.py \ fail2ban.chart.py \ freeradius.chart.py \ + go_expvar.chart.py \ haproxy.chart.py \ hddtemp.chart.py \ ipfs.chart.py \ @@ -271,8 +336,10 @@ dist_python_DATA = \ phpfpm.chart.py \ postfix.chart.py \ postgres.chart.py \ + rabbitmq.chart.py \ redis.chart.py \ retroshare.chart.py \ + samba.chart.py \ sensors.chart.py \ squid.chart.py \ smartd_log.chart.py \ @@ -368,8 +435,11 @@ $(ACLOCAL_M4): @MAINTAINER_MODE_TRUE@ $(am__aclocal_m4_deps) $(am__aclocal_m4_deps): install-dist_pythonSCRIPTS: $(dist_python_SCRIPTS) @$(NORMAL_INSTALL) - test -z "$(pythondir)" || $(MKDIR_P) "$(DESTDIR)$(pythondir)" @list='$(dist_python_SCRIPTS)'; test -n "$(pythondir)" || list=; \ + if test -n "$$list"; then \ + echo " $(MKDIR_P) '$(DESTDIR)$(pythondir)'"; \ + $(MKDIR_P) "$(DESTDIR)$(pythondir)" || exit 1; \ + fi; \ for p in $$list; do \ if test -f "$$p"; then d=; else d="$(srcdir)/"; fi; \ if test -f "$$d$$p"; then echo "$$d$$p"; echo "$$p"; else :; fi; \ @@ -400,8 +470,11 @@ uninstall-dist_pythonSCRIPTS: dir='$(DESTDIR)$(pythondir)'; $(am__uninstall_files_from_dir) install-dist_pythonDATA: $(dist_python_DATA) @$(NORMAL_INSTALL) - test -z "$(pythondir)" || $(MKDIR_P) "$(DESTDIR)$(pythondir)" @list='$(dist_python_DATA)'; test -n "$(pythondir)" || list=; \ + if test -n "$$list"; then \ + echo " $(MKDIR_P) '$(DESTDIR)$(pythondir)'"; \ + $(MKDIR_P) "$(DESTDIR)$(pythondir)" || exit 1; \ + fi; \ for p in $$list; do \ if test -f "$$p"; then d=; else d="$(srcdir)/"; fi; \ echo "$$d$$p"; \ @@ -418,8 +491,11 @@ uninstall-dist_pythonDATA: dir='$(DESTDIR)$(pythondir)'; $(am__uninstall_files_from_dir) install-dist_pythonmodulesDATA: $(dist_pythonmodules_DATA) @$(NORMAL_INSTALL) - test -z "$(pythonmodulesdir)" || $(MKDIR_P) "$(DESTDIR)$(pythonmodulesdir)" @list='$(dist_pythonmodules_DATA)'; test -n "$(pythonmodulesdir)" || list=; \ + if test -n "$$list"; then \ + echo " $(MKDIR_P) '$(DESTDIR)$(pythonmodulesdir)'"; \ + $(MKDIR_P) "$(DESTDIR)$(pythonmodulesdir)" || exit 1; \ + fi; \ for p in $$list; do \ if test -f "$$p"; then d=; else d="$(srcdir)/"; fi; \ echo "$$d$$p"; \ @@ -436,8 +512,11 @@ uninstall-dist_pythonmodulesDATA: dir='$(DESTDIR)$(pythonmodulesdir)'; $(am__uninstall_files_from_dir) install-dist_pythonyaml2DATA: $(dist_pythonyaml2_DATA) @$(NORMAL_INSTALL) - test -z "$(pythonyaml2dir)" || $(MKDIR_P) "$(DESTDIR)$(pythonyaml2dir)" @list='$(dist_pythonyaml2_DATA)'; test -n "$(pythonyaml2dir)" || list=; \ + if test -n "$$list"; then \ + echo " $(MKDIR_P) '$(DESTDIR)$(pythonyaml2dir)'"; \ + $(MKDIR_P) "$(DESTDIR)$(pythonyaml2dir)" || exit 1; \ + fi; \ for p in $$list; do \ if test -f "$$p"; then d=; else d="$(srcdir)/"; fi; \ echo "$$d$$p"; \ @@ -454,8 +533,11 @@ uninstall-dist_pythonyaml2DATA: dir='$(DESTDIR)$(pythonyaml2dir)'; $(am__uninstall_files_from_dir) install-dist_pythonyaml3DATA: $(dist_pythonyaml3_DATA) @$(NORMAL_INSTALL) - test -z "$(pythonyaml3dir)" || $(MKDIR_P) "$(DESTDIR)$(pythonyaml3dir)" @list='$(dist_pythonyaml3_DATA)'; test -n "$(pythonyaml3dir)" || list=; \ + if test -n "$$list"; then \ + echo " $(MKDIR_P) '$(DESTDIR)$(pythonyaml3dir)'"; \ + $(MKDIR_P) "$(DESTDIR)$(pythonyaml3dir)" || exit 1; \ + fi; \ for p in $$list; do \ if test -f "$$p"; then d=; else d="$(srcdir)/"; fi; \ echo "$$d$$p"; \ @@ -470,11 +552,11 @@ uninstall-dist_pythonyaml3DATA: @list='$(dist_pythonyaml3_DATA)'; test -n "$(pythonyaml3dir)" || list=; \ files=`for p in $$list; do echo $$p; done | sed -e 's|^.*/||'`; \ dir='$(DESTDIR)$(pythonyaml3dir)'; $(am__uninstall_files_from_dir) -tags: TAGS -TAGS: +tags TAGS: + +ctags CTAGS: -ctags: CTAGS -CTAGS: +cscope cscopelist: distdir: $(DISTFILES) @@ -618,20 +700,20 @@ uninstall-am: uninstall-dist_pythonDATA uninstall-dist_pythonSCRIPTS \ .MAKE: install-am install-strip -.PHONY: all all-am check check-am clean clean-generic distclean \ - distclean-generic distdir dvi dvi-am html html-am info info-am \ - install install-am install-data install-data-am \ - install-dist_pythonDATA install-dist_pythonSCRIPTS \ - install-dist_pythonmodulesDATA install-dist_pythonyaml2DATA \ - install-dist_pythonyaml3DATA install-dvi install-dvi-am \ - install-exec install-exec-am install-html install-html-am \ - install-info install-info-am install-man install-pdf \ - install-pdf-am install-ps install-ps-am install-strip \ - installcheck installcheck-am installdirs maintainer-clean \ - maintainer-clean-generic mostlyclean mostlyclean-generic pdf \ - pdf-am ps ps-am uninstall uninstall-am \ - uninstall-dist_pythonDATA uninstall-dist_pythonSCRIPTS \ - uninstall-dist_pythonmodulesDATA \ +.PHONY: all all-am check check-am clean clean-generic cscopelist-am \ + ctags-am distclean distclean-generic distdir dvi dvi-am html \ + html-am info info-am install install-am install-data \ + install-data-am install-dist_pythonDATA \ + install-dist_pythonSCRIPTS install-dist_pythonmodulesDATA \ + install-dist_pythonyaml2DATA install-dist_pythonyaml3DATA \ + install-dvi install-dvi-am install-exec install-exec-am \ + install-html install-html-am install-info install-info-am \ + install-man install-pdf install-pdf-am install-ps \ + install-ps-am install-strip installcheck installcheck-am \ + installdirs maintainer-clean maintainer-clean-generic \ + mostlyclean mostlyclean-generic pdf pdf-am ps ps-am tags-am \ + uninstall uninstall-am uninstall-dist_pythonDATA \ + uninstall-dist_pythonSCRIPTS uninstall-dist_pythonmodulesDATA \ uninstall-dist_pythonyaml2DATA uninstall-dist_pythonyaml3DATA .in: diff --git a/python.d/README.md b/python.d/README.md index 7df6e3e8..c4504a7c 100644 --- a/python.d/README.md +++ b/python.d/README.md @@ -228,6 +228,16 @@ It produces one stacked chart per CPU, showing the percentage of time spent in each state. --- +# dns_query_time + +This module provides dns query time statistics. + +**Requirement:** +* `python-dnspython` package + +It produces one aggregate chart or one chart per dns server, showing the query time. + +--- # dovecot @@ -473,6 +483,44 @@ and restart/reload your FREERADIUS server. --- +# go_expvar + +--- + +The `go_expvar` module can monitor any Go application that exposes its metrics with the use of `expvar` package from the Go standard library. + +`go_expvar` produces charts for Go runtime memory statistics and optionally any number of custom charts. Please see the [wiki page](https://github.com/firehol/netdata/wiki/Monitoring-Go-Applications) for more info. + +For the memory statistics, it produces the following charts: + +1. **Heap allocations** in kB + * alloc: size of objects allocated on the heap + * inuse: size of allocated heap spans + +2. **Stack allocations** in kB + * inuse: size of allocated stack spans + +3. **MSpan allocations** in kB + * inuse: size of allocated mspan structures + +4. **MCache allocations** in kB + * inuse: size of allocated mcache structures + +5. **Virtual memory** in kB + * sys: size of reserved virtual address space + +6. **Live objects** + * live: number of live objects in memory + +7. **GC pauses average** in ns + * avg: average duration of all GC stop-the-world pauses + +### configuration + +Please see the [wiki page](https://github.com/firehol/netdata/wiki/Monitoring-Go-Applications#using-netdata-go_expvar-module) for detailed info about module configuration. + +--- + # haproxy Module monitors frontend and backend metrics such as bytes in, bytes out, sessions current, sessions in queue current. @@ -1198,6 +1246,60 @@ When no configuration file is found, module tries to connect to TCP/IP socket: ` --- +# rabbitmq + +Module monitor rabbitmq performance and health metrics. + +Following charts are drawn: + +1. **Queued Messages** + * ready + * unacknowledged + +2. **Message Rates** + * ack + * redelivered + * deliver + * publish + +3. **Global Counts** + * channels + * consumers + * connections + * queues + * exchanges + +4. **File Descriptors** + * used descriptors + +5. **Socket Descriptors** + * used descriptors + +6. **Erlang processes** + * used processes + +7. **Memory** + * free memory in megabytes + +8. **Disk Space** + * free disk space in gigabytes + +### configuration + +```yaml +socket: + name : 'local' + host : '127.0.0.1' + port : 15672 + user : 'guest' + pass : 'guest' + +``` + +When no configuration file is found, module tries to connect to: `localhost:15672`. + +--- + # redis Get INFO data from redis instance. @@ -1241,6 +1343,68 @@ When no configuration file is found, module tries to connect to TCP/IP socket: ` --- +# samba + +Performance metrics of Samba file sharing. + +It produces the following charts: + +1. **Syscall R/Ws** in kilobytes/s + * sendfile + * recvfle + +2. **Smb2 R/Ws** in kilobytes/s + * readout + * writein + * readin + * writeout + +3. **Smb2 Create/Close** in operations/s + * create + * close + +4. **Smb2 Info** in operations/s + * getinfo + * setinfo + +5. **Smb2 Find** in operations/s + * find + +6. **Smb2 Notify** in operations/s + * notify + +7. **Smb2 Lesser Ops** as counters + * tcon + * negprot + * tdis + * cancel + * logoff + * flush + * lock + * keepalive + * break + * sessetup + +### configuration + +Requires that smbd has been compiled with profiling enabled. Also required +that `smbd` was started either with the `-P 1` option or inside `smb.conf` +using `smbd profiling level`. + +This plugin uses `smbstatus -P` which can only be executed by root. It uses +sudo and assumes that it is configured such that the `netdata` user can +execute smbstatus as root without password. + +For example: + + netdata ALL=(ALL) NOPASSWD: /usr/bin/smbstatus -P + +```yaml +update_every : 5 # update frequency +``` + +--- + # sensors System sensors information. @@ -1251,6 +1415,12 @@ Charts are created dynamically. For detailed configuration information please read [`sensors.conf`](https://github.com/firehol/netdata/blob/master/conf.d/python.d/sensors.conf) file. +### possible issues + +There have been reports from users that on certain servers, ACPI ring buffer errors are printed by the kernel (`dmesg`) when ACPI sensors are being accessed. +We are tracking such cases in issue [#827](https://github.com/firehol/netdata/issues/827). +Please join this discussion for help. + --- # squid diff --git a/python.d/apache.chart.py b/python.d/apache.chart.py index 2e4d16dd..71fe0300 100644 --- a/python.d/apache.chart.py +++ b/python.d/apache.chart.py @@ -22,7 +22,8 @@ ORDER = ['requests', 'connections', 'conns_async', 'net', 'workers', 'reqpersec' CHARTS = { 'bytesperreq': { - 'options': [None, 'apache Lifetime Avg. Response Size', 'bytes/request', 'statistics', 'apache.bytesperreq', 'area'], + 'options': [None, 'apache Lifetime Avg. Response Size', 'bytes/request', + 'statistics', 'apache.bytesperreq', 'area'], 'lines': [ ["size_req"] ]}, @@ -30,15 +31,19 @@ CHARTS = { 'options': [None, 'apache Workers', 'workers', 'workers', 'apache.workers', 'stacked'], 'lines': [ ["idle"], - ["busy"] + ["idle_servers", 'idle'], + ["busy"], + ["busy_servers", 'busy'] ]}, 'reqpersec': { - 'options': [None, 'apache Lifetime Avg. Requests/s', 'requests/s', 'statistics', 'apache.reqpersec', 'area'], + 'options': [None, 'apache Lifetime Avg. Requests/s', 'requests/s', 'statistics', + 'apache.reqpersec', 'area'], 'lines': [ ["requests_sec"] ]}, 'bytespersec': { - 'options': [None, 'apache Lifetime Avg. Bandwidth/s', 'kilobytes/s', 'statistics', 'apache.bytesperreq', 'area'], + 'options': [None, 'apache Lifetime Avg. Bandwidth/s', 'kilobytes/s', 'statistics', + 'apache.bytesperreq', 'area'], 'lines': [ ["size_sec", None, 'absolute', 1, 1000] ]}, @@ -66,43 +71,54 @@ CHARTS = { ]} } +ASSIGNMENT = {"BytesPerReq": 'size_req', + "IdleWorkers": 'idle', + "IdleServers": 'idle_servers', + "BusyWorkers": 'busy', + "BusyServers": 'busy_servers', + "ReqPerSec": 'requests_sec', + "BytesPerSec": 'size_sec', + "Total Accesses": 'requests', + "Total kBytes": 'sent', + "ConnsTotal": 'connections', + "ConnsAsyncKeepAlive": 'keepalive', + "ConnsAsyncClosing": 'closing', + "ConnsAsyncWriting": 'writing'} + class Service(UrlService): def __init__(self, configuration=None, name=None): UrlService.__init__(self, configuration=configuration, name=name) - if len(self.url) == 0: - self.url = "http://localhost/server-status?auto" self.order = ORDER self.definitions = CHARTS - self.assignment = {"BytesPerReq": 'size_req', - "IdleWorkers": 'idle', - "BusyWorkers": 'busy', - "ReqPerSec": 'requests_sec', - "BytesPerSec": 'size_sec', - "Total Accesses": 'requests', - "Total kBytes": 'sent', - "ConnsTotal": 'connections', - "ConnsAsyncKeepAlive": 'keepalive', - "ConnsAsyncClosing": 'closing', - "ConnsAsyncWriting": 'writing'} + self.url = self.configuration.get('url', 'http://localhost/server-status?auto') + + def check(self): + if UrlService.check(self): + if 'idle_servers' in self._data_from_check: + self.__module__ = 'lighttpd' + for chart in self.definitions: + opts = self.definitions[chart]['options'] + opts[1] = opts[1].replace('apache', 'lighttpd') + opts[4] = opts[4].replace('apache', 'lighttpd') + return True + return False def _get_data(self): """ Format data received from http request :return: dict """ - try: - raw = self._get_raw_data().split("\n") - except AttributeError: + raw_data = self._get_raw_data() + if not raw_data: return None - data = {} - for row in raw: + data = dict() + + for row in raw_data.split('\n'): tmp = row.split(":") - if str(tmp[0]) in self.assignment: + if tmp[0] in ASSIGNMENT: try: - data[self.assignment[tmp[0]]] = int(float(tmp[1])) + data[ASSIGNMENT[tmp[0]]] = int(float(tmp[1])) except (IndexError, ValueError): - pass - if len(data) == 0: - return None - return data + continue + return data or None diff --git a/python.d/bind_rndc.chart.py b/python.d/bind_rndc.chart.py index a4d75370..5a974928 100644 --- a/python.d/bind_rndc.chart.py +++ b/python.d/bind_rndc.chart.py @@ -2,100 +2,141 @@ # Description: bind rndc netdata python.d module # Author: l2isbad -from base import SimpleService -from re import compile, findall -from os.path import getsize, split -from os import access as is_accessible, R_OK +from os.path import getsize +from os import access, R_OK from subprocess import Popen +from collections import defaultdict +from base import SimpleService priority = 60000 retries = 60 update_every = 30 -NMS = ['requests', 'responses', 'success', 'auth_answer', 'nonauth_answer', 'nxrrset', 'failure', - 'nxdomain', 'recursion', 'duplicate', 'rejections'] -QUERIES = ['RESERVED0', 'A', 'NS', 'CNAME', 'SOA', 'PTR', 'MX', 'TXT', 'X25', 'AAAA', 'SRV', 'NAPTR', - 'A6', 'DS', 'RRSIG', 'DNSKEY', 'SPF', 'ANY', 'DLV'] +ORDER = ['name_server_statistics', 'incoming_queries', 'outgoing_queries', 'named_stats_size'] + +CHARTS = { + 'name_server_statistics': { + 'options': [None, 'Name Server Statistics', 'stats', 'name server statistics', + 'bind_rndc.name_server_statistics', 'line'], + 'lines': [ + ['nms_requests', 'requests', 'incremental'], + ['nms_rejected_queries', 'rejected_queries', 'incremental'], + ['nms_success', 'success', 'incremental'], + ['nms_failure', 'failure', 'incremental'], + ['nms_responses', 'responses', 'incremental'], + ['nms_duplicate', 'duplicate', 'incremental'], + ['nms_recursion', 'recursion', 'incremental'], + ['nms_nxrrset', 'nxrrset', 'incremental'], + ['nms_nxdomain', 'nxdomain', 'incremental'], + ['nms_non_auth_answer', 'non_auth_answer', 'incremental'], + ['nms_auth_answer', 'auth_answer', 'incremental'], + ['nms_dropped_queries', 'dropped_queries', 'incremental'], + ]}, + 'incoming_queries': { + 'options': [None, 'Incoming Queries', 'queries', 'incoming queries', + 'bind_rndc.incoming_queries', 'line'], + 'lines': [ + ]}, + 'outgoing_queries': { + 'options': [None, 'Outgoing Queries', 'queries', 'outgoing queries', + 'bind_rndc.outgoing_queries', 'line'], + 'lines': [ + ]}, + 'named_stats_size': { + 'options': [None, 'Named Stats File Size', 'MB', 'file size', + 'bind_rndc.stats_size', 'line'], + 'lines': [ + ['stats_size', None, 'absolute', 1, 1 << 20] + ]} +} + +NMS = { + 'nms_requests': + ['IPv4 requests received', + 'IPv6 requests received', + 'TCP requests received', + 'requests with EDNS(0) receive'], + 'nms_responses': + ['responses sent', + 'truncated responses sent', + 'responses with EDNS(0) sent', + 'requests with unsupported EDNS version received'], + 'nms_failure': + ['other query failures', + 'queries resulted in SERVFAIL'], + 'nms_auth_answer': + ['queries resulted in authoritative answer'], + 'nms_non_auth_answer': + ['queries resulted in non authoritative answer'], + 'nms_nxrrset': + ['queries resulted in nxrrset'], + 'nms_success': + ['queries resulted in successful answer'], + 'nms_nxdomain': + ['queries resulted in NXDOMAIN'], + 'nms_recursion': + ['queries caused recursion'], + 'nms_duplicate': + ['duplicate queries received'], + 'nms_rejected_queries': + ['auth queries rejected', + 'recursive queries rejected'], + 'nms_dropped_queries': + ['queries dropped'] +} + +STATS = ['Name Server Statistics', 'Incoming Queries', 'Outgoing Queries'] class Service(SimpleService): def __init__(self, configuration=None, name=None): SimpleService.__init__(self, configuration=configuration, name=name) + self.order = ORDER + self.definitions = CHARTS self.named_stats_path = self.configuration.get('named_stats_path', '/var/log/bind/named.stats') - self.regex_values = compile(r'([0-9]+) ([^\n]+)') - # self.options = ['Incoming Requests', 'Incoming Queries', 'Outgoing Queries', - # 'Name Server Statistics', 'Zone Maintenance Statistics', 'Resolver Statistics', - # 'Cache DB RRsets', 'Socket I/O Statistics'] - self.options = ['Name Server Statistics', 'Incoming Queries', 'Outgoing Queries'] - self.regex_options = [r'(%s(?= \+\+)) \+\+([^\+]+)' % option for option in self.options] self.rndc = self.find_binary('rndc') + self.data = dict(nms_requests=0, nms_responses=0, nms_failure=0, nms_auth=0, + nms_non_auth=0, nms_nxrrset=0, nms_success=0, nms_nxdomain=0, + nms_recursion=0, nms_duplicate=0, nms_rejected_queries=0, + nms_dropped_queries=0) def check(self): - # We cant start without 'rndc' command if not self.rndc: self.error('Can\'t locate \'rndc\' binary or binary is not executable by netdata') return False - # We cant start if stats file is not exist or not readable by netdata user - if not is_accessible(self.named_stats_path, R_OK): + if not access(self.named_stats_path, R_OK): self.error('Cannot access file %s' % self.named_stats_path) return False - size_before = getsize(self.named_stats_path) run_rndc = Popen([self.rndc, 'stats'], shell=False) run_rndc.wait() - size_after = getsize(self.named_stats_path) - # We cant start if netdata user has no permissions to run 'rndc stats' if not run_rndc.returncode: - # 'rndc' was found, stats file is exist and readable and we can run 'rndc stats'. Lets go! - self.create_charts() - - # BIND APPEND dump on every run 'rndc stats' - # that is why stats file size can be VERY large if update_interval too small - dump_size_24hr = round(86400 / self.update_every * (int(size_after) - int(size_before)) / 1048576, 3) - - # If update_every too small we should WARN user - if self.update_every < 30: - self.info('Update_every %s is NOT recommended for use. Increase the value to > 30' % self.update_every) - - self.info('With current update_interval it will be + %s MB every 24hr. ' - 'Don\'t forget to create logrotate conf file for %s' % (dump_size_24hr, self.named_stats_path)) - - self.info('Plugin was started successfully.') - return True - else: - self.error('Not enough permissions to run "%s stats"' % self.rndc) - return False + self.error('Not enough permissions to run "%s stats"' % self.rndc) + return False def _get_raw_data(self): """ Run 'rndc stats' and read last dump from named.stats - :return: tuple( - file.read() obj, - named.stats file size - ) + :return: dict """ + result = dict() try: current_size = getsize(self.named_stats_path) - except OSError: - return None, None - - run_rndc = Popen([self.rndc, 'stats'], shell=False) - run_rndc.wait() - - if run_rndc.returncode: - return None, None - - try: - with open(self.named_stats_path) as bind_rndc: - bind_rndc.seek(current_size) - result = bind_rndc.read() - except OSError: - return None, None - else: - return result, current_size + run_rndc = Popen([self.rndc, 'stats'], shell=False) + run_rndc.wait() + + if run_rndc.returncode: + return None + with open(self.named_stats_path) as named_stats: + named_stats.seek(current_size) + result['stats'] = named_stats.readlines() + result['size'] = current_size + return result + except (OSError, IOError): + return None def _get_data(self): """ @@ -103,72 +144,98 @@ class Service(SimpleService): :return: dict """ - raw_data, size = self._get_raw_data() + raw_data = self._get_raw_data() if raw_data is None: return None - - rndc_stats = dict() - - # Result: dict. - # topic = Cache DB RRsets; body = A 178303 NS 86790 ... ; desc = A; value = 178303 - # {'Cache DB RRsets': [('A', 178303), ('NS', 286790), ...], - # {Incoming Queries': [('RESERVED0', 8), ('A', 4557317680), ...], - # ...... - for regex in self.regex_options: - rndc_stats.update(dict([(topic, [(desc, int(value)) for value, desc in self.regex_values.findall(body)]) - for topic, body in findall(regex, raw_data)])) - - nms = dict(rndc_stats.get('Name Server Statistics', [])) - - inc_queries = dict([('i' + k, 0) for k in QUERIES]) - inc_queries.update(dict([('i' + k, v) for k, v in rndc_stats.get('Incoming Queries', [])])) - out_queries = dict([('o' + k, 0) for k in QUERIES]) - out_queries.update(dict([('o' + k, v) for k, v in rndc_stats.get('Outgoing Queries', [])])) - - to_netdata = dict() - to_netdata['requests'] = sum([v for k, v in nms.items() if 'request' in k and 'received' in k]) - to_netdata['responses'] = sum([v for k, v in nms.items() if 'responses' in k and 'sent' in k]) - to_netdata['success'] = nms.get('queries resulted in successful answer', 0) - to_netdata['auth_answer'] = nms.get('queries resulted in authoritative answer', 0) - to_netdata['nonauth_answer'] = nms.get('queries resulted in non authoritative answer', 0) - to_netdata['nxrrset'] = nms.get('queries resulted in nxrrset', 0) - to_netdata['failure'] = sum([nms.get('queries resulted in SERVFAIL', 0), nms.get('other query failures', 0)]) - to_netdata['nxdomain'] = nms.get('queries resulted in NXDOMAIN', 0) - to_netdata['recursion'] = nms.get('queries caused recursion', 0) - to_netdata['duplicate'] = nms.get('duplicate queries received', 0) - to_netdata['rejections'] = nms.get('recursive queries rejected', 0) - to_netdata['stats_size'] = size - - to_netdata.update(inc_queries) - to_netdata.update(out_queries) - return to_netdata - - def create_charts(self): - self.order = ['stats_size', 'bind_stats', 'incoming_q', 'outgoing_q'] - self.definitions = { - 'bind_stats': { - 'options': [None, 'Name Server Statistics', 'stats', 'name server statistics', 'bind_rndc.stats', 'line'], - 'lines': [ - ]}, - 'incoming_q': { - 'options': [None, 'Incoming queries', 'queries','incoming queries', 'bind_rndc.incq', 'line'], - 'lines': [ - ]}, - 'outgoing_q': { - 'options': [None, 'Outgoing queries', 'queries','outgoing queries', 'bind_rndc.outq', 'line'], - 'lines': [ - ]}, - 'stats_size': { - 'options': [None, '%s file size' % split(self.named_stats_path)[1].capitalize(), 'megabytes', - '%s size' % split(self.named_stats_path)[1], 'bind_rndc.size', 'line'], - 'lines': [ - ["stats_size", None, "absolute", 1, 1048576] - ]} - } - for elem in QUERIES: - self.definitions['incoming_q']['lines'].append(['i' + elem, elem, 'incremental']) - self.definitions['outgoing_q']['lines'].append(['o' + elem, elem, 'incremental']) - - for elem in NMS: - self.definitions['bind_stats']['lines'].append([elem, None, 'incremental']) + parsed = dict() + for stat in STATS: + parsed[stat] = parse_stats(field=stat, + named_stats=raw_data['stats']) + + self.data.update(nms_mapper(data=parsed['Name Server Statistics'])) + + for elem in zip(['Incoming Queries', 'Outgoing Queries'], ['incoming_queries', 'outgoing_queries']): + parsed_key, chart_name = elem[0], elem[1] + for dimension_id, value in queries_mapper(data=parsed[parsed_key], + add=chart_name[:9]).items(): + if dimension_id not in self.data: + dimension = dimension_id.replace(chart_name[:9], '') + self._add_new_dimension(dimension_id=dimension_id, + dimension=dimension, + chart_name=chart_name, + priority=self.priority + self.order.index(chart_name)) + self.data[dimension_id] = value + + self.data['stats_size'] = raw_data['size'] + return self.data + + +def parse_stats(field, named_stats): + """ + :param field: str: + :param named_stats: list: + :return: dict + + Example: + filed: 'Incoming Queries' + names_stats (list of lines): + ++ Incoming Requests ++ + 1405660 QUERY + 3 NOTIFY + ++ Incoming Queries ++ + 1214961 A + 75 NS + 2 CNAME + 2897 SOA + 35544 PTR + 14 MX + 5822 TXT + 145974 AAAA + 371 SRV + ++ Outgoing Queries ++ + ... + + result: + {'A', 1214961, 'NS': 75, 'CNAME': 2, 'SOA': 2897, ...} + """ + data = dict() + ns = iter(named_stats) + for line in ns: + if field not in line: + continue + while True: + try: + line = next(ns) + except StopIteration: + break + if '++' not in line: + if '[' in line: + continue + v, k = line.strip().split(' ', 1) + data[k] = int(v) + continue + break + break + return data + + +def nms_mapper(data): + """ + :param data: dict + :return: dict(defaultdict) + """ + result = defaultdict(int) + for k, v in NMS.items(): + for elem in v: + result[k] += data.get(elem, 0) + return result + + +def queries_mapper(data, add): + """ + :param data: dict + :param add: str + :return: dict + """ + return dict([(add + k, v) for k, v in data.items()]) diff --git a/python.d/cpufreq.chart.py b/python.d/cpufreq.chart.py index e28bdea8..d5544b7b 100644 --- a/python.d/cpufreq.chart.py +++ b/python.d/cpufreq.chart.py @@ -100,7 +100,7 @@ class Service(SimpleService): self.error("couldn't find a method to read cpufreq statistics") return False - for name in self.assignment.keys(): + for name in sorted(self.assignment, key=lambda v: int(v[3:])): self.definitions[ORDER[0]]['lines'].append([name, name, 'absolute', 1, 1000]) return True diff --git a/python.d/dns_query_time.chart.py b/python.d/dns_query_time.chart.py new file mode 100644 index 00000000..9053d9a1 --- /dev/null +++ b/python.d/dns_query_time.chart.py @@ -0,0 +1,135 @@ +# -*- coding: utf-8 -*- +# Description: dns_query_time netdata python.d module +# Author: l2isbad + +try: + from time import monotonic as time +except ImportError: + from time import time +try: + import dns.message, dns.query, dns.name + DNS_PYTHON = True +except ImportError: + DNS_PYTHON = False +try: + from queue import Queue +except ImportError: + from Queue import Queue +from random import choice +from threading import Thread +from socket import gethostbyname, gaierror +from base import SimpleService + + +# default module values (can be overridden per job in `config`) +update_every = 5 +priority = 60000 +retries = 60 + + +class Service(SimpleService): + def __init__(self, configuration=None, name=None): + SimpleService.__init__(self, configuration=configuration, name=name) + self.order = list() + self.definitions = dict() + self.timeout = self.configuration.get('response_timeout', 4) + self.aggregate = self.configuration.get('aggregate', True) + self.domains = self.configuration.get('domains') + self.server_list = self.configuration.get('dns_servers') + + def check(self): + if not DNS_PYTHON: + self.error('\'python-dnspython\' package is needed to use dns_query_time.chart.py') + return False + + self.timeout = self.timeout if isinstance(self.timeout, int) else 4 + self.update_every = self.timeout + 1 if self.update_every <= self.timeout else self.update_every + + if not all([self.domains, self.server_list, + isinstance(self.server_list, str), isinstance(self.domains, str)]): + self.error('server_list and domain_list can\'t be empty') + return False + else: + self.domains, self.server_list = self.domains.split(), self.server_list.split() + + for ns in self.server_list: + if not check_ns(ns): + self.info('Bad NS: %s' % ns) + self.server_list.remove(ns) + if not self.server_list: + return False + + data = self._get_data(timeout=1) + + down_servers = [s for s in data if data[s] == -100] + for down in down_servers: + down = down[3:].replace('_', '.') + self.info('Removed due to non response %s' % down) + self.server_list.remove(down) + if not self.server_list: + return False + + self._data_from_check = data + self.order, self.definitions = create_charts(aggregate=self.aggregate, server_list=self.server_list) + self.info(str({'domains': len(self.domains), 'servers': self.server_list})) + return True + + def _get_data(self, timeout=None): + return dns_request(self.server_list, timeout or self.timeout, self.domains) + + +def dns_request(server_list, timeout, domains): + threads = list() + que = Queue() + result = dict() + + def dns_req(ns, t, q): + domain = dns.name.from_text(choice(domains)) + request = dns.message.make_query(domain, dns.rdatatype.A) + + try: + dns_start = time() + dns.query.udp(request, ns, timeout=t) + dns_end = time() + query_time = round((dns_end - dns_start) * 1000) + q.put({'_'.join(['ns', ns.replace('.', '_')]): query_time}) + except dns.exception.Timeout: + q.put({'_'.join(['ns', ns.replace('.', '_')]): -100}) + + for server in server_list: + th = Thread(target=dns_req, args=(server, timeout, que)) + th.start() + threads.append(th) + + for th in threads: + th.join() + result.update(que.get()) + + return result + + +def check_ns(ns): + try: + return gethostbyname(ns) + except gaierror: + return False + + +def create_charts(aggregate, server_list): + if aggregate: + order = ['dns_group'] + definitions = {'dns_group': {'options': [None, 'DNS Response Time', 'ms', 'name servers', + 'dns_query_time.response_time', 'line'], 'lines': []}} + for ns in server_list: + definitions['dns_group']['lines'].append(['_'.join(['ns', ns.replace('.', '_')]), ns, 'absolute']) + + return order, definitions + else: + order = [''.join(['dns_', ns.replace('.', '_')]) for ns in server_list] + definitions = dict() + for ns in server_list: + definitions[''.join(['dns_', ns.replace('.', '_')])] = {'options': [None, 'DNS Response Time', 'ms', ns, + 'dns_query_time.response_time', 'area'], + 'lines': [['_'.join(['ns', ns.replace('.', '_')]), + ns, 'absolute']]} + return order, definitions diff --git a/python.d/elasticsearch.chart.py b/python.d/elasticsearch.chart.py index 430227f6..9ec08719 100644 --- a/python.d/elasticsearch.chart.py +++ b/python.d/elasticsearch.chart.py @@ -85,6 +85,21 @@ HEALTH_STATS = [ ('active_shards_percent_as_number', 'health_active_shards_percent_as_number', None) ] +LATENCY = { + 'query_latency': + {'total': 'query_total', + 'spent_time': 'query_time_in_millis'}, + 'fetch_latency': + {'total': 'fetch_total', + 'spent_time': 'fetch_time_in_millis'}, + 'indexing_latency': + {'total': 'indexing_index_total', + 'spent_time': 'indexing_index_time_in_millis'}, + 'flushing_latency': + {'total': 'flush_total', + 'spent_time': 'flush_total_time_in_millis'} +} + # charts order (can be overridden if you want less charts, or different order) ORDER = ['search_perf_total', 'search_perf_current', 'search_perf_time', 'search_latency', 'index_perf_total', 'index_perf_current', 'index_perf_time', 'index_latency', 'jvm_mem_heap', 'jvm_gc_count', @@ -95,34 +110,34 @@ ORDER = ['search_perf_total', 'search_perf_current', 'search_perf_time', 'search CHARTS = { 'search_perf_total': { - 'options': [None, 'Total number of queries, fetches', 'number of', 'search performance', + 'options': [None, 'Queries And Fetches', 'number of', 'search performance', 'es.search_query_total', 'stacked'], 'lines': [ ['query_total', 'queries', 'incremental'], ['fetch_total', 'fetches', 'incremental'] ]}, 'search_perf_current': { - 'options': [None, 'Number of queries, fetches in progress', 'number of', 'search performance', + 'options': [None, 'Queries and Fetches In Progress', 'number of', 'search performance', 'es.search_query_current', 'stacked'], 'lines': [ ['query_current', 'queries', 'absolute'], ['fetch_current', 'fetches', 'absolute'] ]}, 'search_perf_time': { - 'options': [None, 'Time spent on queries, fetches', 'seconds', 'search performance', + 'options': [None, 'Time Spent On Queries And Fetches', 'seconds', 'search performance', 'es.search_time', 'stacked'], 'lines': [ ['query_time_in_millis', 'query', 'incremental', 1, 1000], ['fetch_time_in_millis', 'fetch', 'incremental', 1, 1000] ]}, 'search_latency': { - 'options': [None, 'Query and fetch latency', 'ms', 'search performance', 'es.search_latency', 'stacked'], + 'options': [None, 'Query And Fetch Latency', 'ms', 'search performance', 'es.search_latency', 'stacked'], 'lines': [ ['query_latency', 'query', 'absolute', 1, 1000], ['fetch_latency', 'fetch', 'absolute', 1, 1000] ]}, 'index_perf_total': { - 'options': [None, 'Total number of documents indexed, index refreshes, index flushes to disk', 'number of', + 'options': [None, 'Indexed Documents, Index Refreshes, Index Flushes To Disk', 'number of', 'indexing performance', 'es.index_performance_total', 'stacked'], 'lines': [ ['indexing_index_total', 'indexed', 'incremental'], @@ -130,13 +145,13 @@ CHARTS = { ['flush_total', 'flushes', 'incremental'] ]}, 'index_perf_current': { - 'options': [None, 'Number of documents currently being indexed', 'currently indexed', + 'options': [None, 'Number Of Documents Currently Being Indexed', 'currently indexed', 'indexing performance', 'es.index_performance_current', 'stacked'], 'lines': [ ['indexing_index_current', 'documents', 'absolute'] ]}, 'index_perf_time': { - 'options': [None, 'Time spent on indexing, refreshing, flushing', 'seconds', 'indexing performance', + 'options': [None, 'Time Spent On Indexing, Refreshing, Flushing', 'seconds', 'indexing performance', 'es.search_time', 'stacked'], 'lines': [ ['indexing_index_time_in_millis', 'indexing', 'incremental', 1, 1000], @@ -144,33 +159,33 @@ CHARTS = { ['flush_total_time_in_millis', 'flushing', 'incremental', 1, 1000] ]}, 'index_latency': { - 'options': [None, 'Indexing and flushing latency', 'ms', 'indexing performance', + 'options': [None, 'Indexing And Flushing Latency', 'ms', 'indexing performance', 'es.index_latency', 'stacked'], 'lines': [ ['indexing_latency', 'indexing', 'absolute', 1, 1000], ['flushing_latency', 'flushing', 'absolute', 1, 1000] ]}, 'jvm_mem_heap': { - 'options': [None, 'JVM heap currently in use/committed', 'percent/MB', 'memory usage and gc', + 'options': [None, 'JVM Heap Currently in Use/Committed', 'percent/MB', 'memory usage and gc', 'es.jvm_heap', 'area'], 'lines': [ ['jvm_heap_percent', 'inuse', 'absolute'], ['jvm_heap_commit', 'commit', 'absolute', -1, 1048576] ]}, 'jvm_gc_count': { - 'options': [None, 'Count of garbage collections', 'counts', 'memory usage and gc', 'es.gc_count', 'stacked'], + 'options': [None, 'Garbage Collections', 'counts', 'memory usage and gc', 'es.gc_count', 'stacked'], 'lines': [ ['young_collection_count', 'young', 'incremental'], ['old_collection_count', 'old', 'incremental'] ]}, 'jvm_gc_time': { - 'options': [None, 'Time spent on garbage collections', 'ms', 'memory usage and gc', 'es.gc_time', 'stacked'], + 'options': [None, 'Time Spent On Garbage Collections', 'ms', 'memory usage and gc', 'es.gc_time', 'stacked'], 'lines': [ ['young_collection_time_in_millis', 'young', 'incremental'], ['old_collection_time_in_millis', 'old', 'incremental'] ]}, 'thread_pool_qr_q': { - 'options': [None, 'Number of queued threads in thread pool', 'queued threads', 'queues and rejections', + 'options': [None, 'Number Of Queued Threads In Thread Pool', 'queued threads', 'queues and rejections', 'es.thread_pool_queued', 'stacked'], 'lines': [ ['bulk_queue', 'bulk', 'absolute'], @@ -179,7 +194,7 @@ CHARTS = { ['merge_queue', 'merge', 'absolute'] ]}, 'thread_pool_qr_r': { - 'options': [None, 'Number of rejected threads in thread pool', 'rejected threads', 'queues and rejections', + 'options': [None, 'Rejected Threads In Thread Pool', 'rejected threads', 'queues and rejections', 'es.thread_pool_rejected', 'stacked'], 'lines': [ ['bulk_rejected', 'bulk', 'absolute'], @@ -188,19 +203,19 @@ CHARTS = { ['merge_rejected', 'merge', 'absolute'] ]}, 'fdata_cache': { - 'options': [None, 'Fielddata cache size', 'MB', 'fielddata cache', 'es.fdata_cache', 'line'], + 'options': [None, 'Fielddata Cache', 'MB', 'fielddata cache', 'es.fdata_cache', 'line'], 'lines': [ ['index_fdata_memory', 'cache', 'absolute', 1, 1048576] ]}, 'fdata_ev_tr': { - 'options': [None, 'Fielddata evictions and circuit breaker tripped count', 'number of events', + 'options': [None, 'Fielddata Evictions And Circuit Breaker Tripped Count', 'number of events', 'fielddata cache', 'es.evictions_tripped', 'line'], 'lines': [ ['evictions', None, 'incremental'], ['tripped', None, 'incremental'] ]}, 'cluster_health_nodes': { - 'options': [None, 'Nodes and tasks statistics', 'units', 'cluster health API', + 'options': [None, 'Nodes And Tasks Statistics', 'units', 'cluster health API', 'es.cluster_health_nodes', 'stacked'], 'lines': [ ['health_number_of_nodes', 'nodes', 'absolute'], @@ -209,7 +224,7 @@ CHARTS = { ['health_number_of_in_flight_fetch', 'in_flight_fetch', 'absolute'] ]}, 'cluster_health_status': { - 'options': [None, 'Cluster status', 'status', 'cluster health API', + 'options': [None, 'Cluster Status', 'status', 'cluster health API', 'es.cluster_health_status', 'area'], 'lines': [ ['status_green', 'green', 'absolute'], @@ -220,7 +235,7 @@ CHARTS = { ['status_yellow', 'yellow', 'absolute'] ]}, 'cluster_health_shards': { - 'options': [None, 'Shards statistics', 'shards', 'cluster health API', + 'options': [None, 'Shards Statistics', 'shards', 'cluster health API', 'es.cluster_health_shards', 'stacked'], 'lines': [ ['health_active_shards', 'active_shards', 'absolute'], @@ -231,7 +246,7 @@ CHARTS = { ['health_active_shards_percent_as_number', 'active_percent', 'absolute'] ]}, 'cluster_stats_nodes': { - 'options': [None, 'Nodes statistics', 'nodes', 'cluster stats API', + 'options': [None, 'Nodes Statistics', 'nodes', 'cluster stats API', 'es.cluster_nodes', 'stacked'], 'lines': [ ['count_data_only', 'data_only', 'absolute'], @@ -241,46 +256,46 @@ CHARTS = { ['count_client', 'client', 'absolute'] ]}, 'cluster_stats_query_cache': { - 'options': [None, 'Query cache statistics', 'queries', 'cluster stats API', + 'options': [None, 'Query Cache Statistics', 'queries', 'cluster stats API', 'es.cluster_query_cache', 'stacked'], 'lines': [ ['query_cache_hit_count', 'hit', 'incremental'], ['query_cache_miss_count', 'miss', 'incremental'] ]}, 'cluster_stats_docs': { - 'options': [None, 'Docs statistics', 'count', 'cluster stats API', + 'options': [None, 'Docs Statistics', 'count', 'cluster stats API', 'es.cluster_docs', 'line'], 'lines': [ ['docs_count', 'docs', 'absolute'] ]}, 'cluster_stats_store': { - 'options': [None, 'Store statistics', 'MB', 'cluster stats API', + 'options': [None, 'Store Statistics', 'MB', 'cluster stats API', 'es.cluster_store', 'line'], 'lines': [ ['store_size_in_bytes', 'size', 'absolute', 1, 1048567] ]}, 'cluster_stats_indices_shards': { - 'options': [None, 'Indices and shards statistics', 'count', 'cluster stats API', + 'options': [None, 'Indices And Shards Statistics', 'count', 'cluster stats API', 'es.cluster_indices_shards', 'stacked'], 'lines': [ ['indices_count', 'indices', 'absolute'], ['shards_total', 'shards', 'absolute'] ]}, 'host_metrics_transport': { - 'options': [None, 'Cluster communication transport metrics', 'kbit/s', 'host metrics', + 'options': [None, 'Cluster Communication Transport Metrics', 'kilobit/s', 'host metrics', 'es.host_transport', 'area'], 'lines': [ ['transport_rx_size_in_bytes', 'in', 'incremental', 8, 1000], ['transport_tx_size_in_bytes', 'out', 'incremental', -8, 1000] ]}, 'host_metrics_file_descriptors': { - 'options': [None, 'Available file descriptors in percent', 'percent', 'host metrics', + 'options': [None, 'Available File Descriptors In Percent', 'percent', 'host metrics', 'es.host_descriptors', 'area'], 'lines': [ ['file_descriptors_used', 'used', 'absolute', 1, 10] ]}, 'host_metrics_http': { - 'options': [None, 'Opened HTTP connections', 'connections', 'host metrics', + 'options': [None, 'Opened HTTP Connections', 'connections', 'host metrics', 'es.host_http_connections', 'line'], 'lines': [ ['http_current_open', 'opened', 'absolute', 1, 1] @@ -300,12 +315,13 @@ class Service(UrlService): self.methods = list() def check(self): - # We can't start if <host> AND <port> not specified - if not all([self.host, self.port, isinstance(self.host, str), isinstance(self.port, (str, int))]): + if not all([self.host, + self.port, + isinstance(self.host, str), + isinstance(self.port, (str, int))]): self.error('Host is not defined in the module configuration file') return False - # It as a bad idea to use hostname. # Hostname -> ip address try: self.host = gethostbyname(self.host) @@ -313,45 +329,33 @@ class Service(UrlService): self.error(str(error)) return False - scheme = 'http' if self.scheme else 'https' + scheme = 'http' if self.scheme == 'http' else 'https' # Add handlers (auth, self signed cert accept) self.url = '%s://%s:%s' % (scheme, self.host, self.port) - self._UrlService__add_openers() + self.opener = self._build_opener() # Create URL for every Elasticsearch API url_node_stats = '%s://%s:%s/_nodes/_local/stats' % (scheme, self.host, self.port) url_cluster_health = '%s://%s:%s/_cluster/health' % (scheme, self.host, self.port) url_cluster_stats = '%s://%s:%s/_cluster/stats' % (scheme, self.host, self.port) - # Create list of enabled API calls user_choice = [bool(self.configuration.get('node_stats', True)), bool(self.configuration.get('cluster_health', True)), bool(self.configuration.get('cluster_stats', True))] - avail_methods = [METHODS(get_data_function=self._get_node_stats_, url=url_node_stats), - METHODS(get_data_function=self._get_cluster_health_, url=url_cluster_health), - METHODS(get_data_function=self._get_cluster_stats_, url=url_cluster_stats)] + avail_methods = [METHODS(get_data_function=self._get_node_stats_, + url=url_node_stats), + METHODS(get_data_function=self._get_cluster_health_, + url=url_cluster_health), + METHODS(get_data_function=self._get_cluster_stats_, + url=url_cluster_stats)] # Remove disabled API calls from 'avail methods' self.methods = [avail_methods[e[0]] for e in enumerate(avail_methods) if user_choice[e[0]]] - - # Run _get_data for ALL active API calls. - api_check_result = dict() - data_from_check = dict() - for method in self.methods: - try: - api_check_result[method.url] = method.get_data_function(None, method.url) - data_from_check.update(api_check_result[method.url] or dict()) - except KeyError as error: - self.error('Failed to parse %s. Error: %s' % (method.url, str(error))) - return False - - # We can start ONLY if all active API calls returned NOT None - if not all(api_check_result.values()): - self.error('Plugin could not get data from all APIs') + data = self._get_data() + if not data: return False - else: - self._data_from_check = data_from_check - return True + self._data_from_check = data + return True def _get_data(self): threads = list() @@ -359,7 +363,8 @@ class Service(UrlService): result = dict() for method in self.methods: - th = Thread(target=method.get_data_function, args=(queue, method.url)) + th = Thread(target=method.get_data_function, + args=(queue, method.url)) th.start() threads.append(th) @@ -378,18 +383,18 @@ class Service(UrlService): raw_data = self._get_raw_data(url) if not raw_data: - return queue.put(dict()) if queue else None - else: - data = loads(raw_data) + return queue.put(dict()) - to_netdata = fetch_data_(raw_data=data, metrics_list=HEALTH_STATS) + data = loads(raw_data) + to_netdata = fetch_data_(raw_data=data, + metrics_list=HEALTH_STATS) - to_netdata.update({'status_green': 0, 'status_red': 0, 'status_yellow': 0, - 'status_foo1': 0, 'status_foo2': 0, 'status_foo3': 0}) - current_status = 'status_' + data['status'] - to_netdata[current_status] = 1 + to_netdata.update({'status_green': 0, 'status_red': 0, 'status_yellow': 0, + 'status_foo1': 0, 'status_foo2': 0, 'status_foo3': 0}) + current_status = 'status_' + data['status'] + to_netdata[current_status] = 1 - return queue.put(to_netdata) if queue else to_netdata + return queue.put(to_netdata) def _get_cluster_stats_(self, queue, url): """ @@ -400,13 +405,13 @@ class Service(UrlService): raw_data = self._get_raw_data(url) if not raw_data: - return queue.put(dict()) if queue else None - else: - data = loads(raw_data) + return queue.put(dict()) - to_netdata = fetch_data_(raw_data=data, metrics_list=CLUSTER_STATS) + data = loads(raw_data) + to_netdata = fetch_data_(raw_data=data, + metrics_list=CLUSTER_STATS) - return queue.put(to_netdata) if queue else to_netdata + return queue.put(to_netdata) def _get_node_stats_(self, queue, url): """ @@ -417,47 +422,46 @@ class Service(UrlService): raw_data = self._get_raw_data(url) if not raw_data: - return queue.put(dict()) if queue else None - else: - data = loads(raw_data) - - node = list(data['nodes'].keys())[0] - to_netdata = fetch_data_(raw_data=data['nodes'][node], metrics_list=NODE_STATS) + return queue.put(dict()) - # Search performance latency - to_netdata['query_latency'] = self.find_avg_(to_netdata['query_total'], - to_netdata['query_time_in_millis'], 'query_latency') - to_netdata['fetch_latency'] = self.find_avg_(to_netdata['fetch_total'], - to_netdata['fetch_time_in_millis'], 'fetch_latency') + data = loads(raw_data) - # Indexing performance latency - to_netdata['indexing_latency'] = self.find_avg_(to_netdata['indexing_index_total'], - to_netdata['indexing_index_time_in_millis'], 'index_latency') - to_netdata['flushing_latency'] = self.find_avg_(to_netdata['flush_total'], - to_netdata['flush_total_time_in_millis'], 'flush_latency') + node = list(data['nodes'].keys())[0] + to_netdata = fetch_data_(raw_data=data['nodes'][node], + metrics_list=NODE_STATS) + # Search, index, flush, fetch performance latency + for key in LATENCY: + try: + to_netdata[key] = self.find_avg_(total=to_netdata[LATENCY[key]['total']], + spent_time=to_netdata[LATENCY[key]['spent_time']], + key=key) + except KeyError: + continue + if 'open_file_descriptors' in to_netdata and 'max_file_descriptors' in to_netdata: to_netdata['file_descriptors_used'] = round(float(to_netdata['open_file_descriptors']) / to_netdata['max_file_descriptors'] * 1000) - return queue.put(to_netdata) if queue else to_netdata + return queue.put(to_netdata) - def find_avg_(self, value1, value2, key): + def find_avg_(self, total, spent_time, key): if key not in self.latency: - self.latency.update({key: [value1, value2]}) + self.latency[key] = dict(total=total, + spent_time=spent_time) return 0 - else: - if not self.latency[key][0] == value1: - latency = round(float(value2 - self.latency[key][1]) / float(value1 - self.latency[key][0]) * 1000) - self.latency.update({key: [value1, value2]}) - return latency - else: - self.latency.update({key: [value1, value2]}) - return 0 + if self.latency[key]['total'] != total: + latency = float(spent_time - self.latency[key]['spent_time'])\ + / float(total - self.latency[key]['total']) * 1000 + self.latency[key]['total'] = total + self.latency[key]['spent_time'] = spent_time + return latency + self.latency[key]['spent_time'] = spent_time + return 0 def fetch_data_(raw_data, metrics_list): to_netdata = dict() - for metric, new_name, function in metrics_list: + for metric, new_name, func in metrics_list: value = raw_data for key in metric.split('.'): try: @@ -465,7 +469,7 @@ def fetch_data_(raw_data, metrics_list): except KeyError: break if not isinstance(value, dict) and key: - to_netdata[new_name or key] = value if not function else function(value) + to_netdata[new_name or key] = value if not func else func(value) return to_netdata diff --git a/python.d/fail2ban.chart.py b/python.d/fail2ban.chart.py index c7d24e8c..35761e89 100644 --- a/python.d/fail2ban.chart.py +++ b/python.d/fail2ban.chart.py @@ -2,131 +2,210 @@ # Description: fail2ban log netdata python.d module # Author: l2isbad -from base import LogService -from re import compile - -try: - from itertools import filterfalse -except ImportError: - from itertools import ifilterfalse as filterfalse +from re import compile as r_compile from os import access as is_accessible, R_OK -from os.path import isdir +from os.path import isdir, getsize from glob import glob +import bisect +from base import LogService priority = 60000 retries = 60 -REGEX = compile(r'\[([A-Za-z-_]+)][^\[\]]*?(?<!# )enabled = true') -ORDER = ['jails_group'] +REGEX_JAILS = r_compile(r'\[([A-Za-z-_0-9]+)][^\[\]]*?(?<!# )enabled = (?:(true|false))') +REGEX_DATA = r_compile(r'\[(?P<jail>[A-Za-z-_0-9]+)\] (?P<action>(?:(U|B)))[a-z]+ (?P<ipaddr>\d{1,3}(?:\.\d{1,3}){3})') +ORDER = ['jails_bans', 'jails_in_jail'] class Service(LogService): + """ + fail2ban log class + Reads logs line by line + Jail auto detection included + It produces following charts: + * Bans per second for every jail + * Banned IPs for every jail (since the last restart of netdata) + """ def __init__(self, configuration=None, name=None): LogService.__init__(self, configuration=configuration, name=name) self.order = ORDER + self.definitions = dict() self.log_path = self.configuration.get('log_path', '/var/log/fail2ban.log') self.conf_path = self.configuration.get('conf_path', '/etc/fail2ban/jail.local') - self.conf_dir = self.configuration.get('conf_dir', '') - try: - self.exclude = self.configuration['exclude'].split() - except (KeyError, AttributeError): - self.exclude = [] + self.conf_dir = self.configuration.get('conf_dir', '/etc/fail2ban/jail.d/') + self.exclude = self.configuration.get('exclude') def _get_data(self): """ Parse new log lines :return: dict """ - try: - raw = self._get_raw_data() - if raw is None: - return None - elif not raw: - return self.data - except (ValueError, AttributeError): + raw = self._get_raw_data() + if raw is None: return None + elif not raw: + return self.to_netdata # Fail2ban logs looks like # 2016-12-25 12:36:04,711 fail2ban.actions[2455]: WARNING [ssh] Ban 178.156.32.231 - data = dict( - zip( - self.jails_list, - [len(list(filterfalse(lambda line: (jail + '] Ban') not in line, raw))) for jail in self.jails_list] - )) - - for jail in data: - self.data[jail] += data[jail] - - return self.data + for row in raw: + match = REGEX_DATA.search(row) + if match: + match_dict = match.groupdict() + jail, action, ipaddr = match_dict['jail'], match_dict['action'], match_dict['ipaddr'] + if jail in self.jails_list: + if action == 'B': + self.to_netdata[jail] += 1 + if address_not_in_jail(self.banned_ips[jail], ipaddr, self.to_netdata[jail + '_in_jail']): + self.to_netdata[jail + '_in_jail'] += 1 + else: + if ipaddr in self.banned_ips[jail]: + self.banned_ips[jail].remove(ipaddr) + self.to_netdata[jail + '_in_jail'] -= 1 + + return self.to_netdata def check(self): + """ + :return: bool - # Check "log_path" is accessible. - # If NOT STOP plugin - if not is_accessible(self.log_path, R_OK): - self.error('Cannot access file %s' % self.log_path) - return False - jails_list = list() - - if self.conf_dir: - dir_jails, error = parse_conf_dir(self.conf_dir) - jails_list.extend(dir_jails) - if not dir_jails: - self.error(error) - - if self.conf_path: - path_jails, error = parse_conf_path(self.conf_path) - jails_list.extend(path_jails) - if not path_jails: - self.error(error) + Check if the "log_path" is not empty and readable + """ - # If for some reason parse failed we still can START with default jails_list. - self.jails_list = list(set(jails_list) - set(self.exclude)) or ['ssh'] - self.data = dict([(jail, 0) for jail in self.jails_list]) - self.create_dimensions() - self.info('Plugin successfully started. Jails: %s' % self.jails_list) + if not (is_accessible(self.log_path, R_OK) and getsize(self.log_path) != 0): + self.error('%s is not readable or empty' % self.log_path) + return False + self.jails_list, self.to_netdata, self.banned_ips = self.jails_auto_detection_() + self.definitions = create_definitions_(self.jails_list) + self.info('Jails: %s' % self.jails_list) return True - def create_dimensions(self): - self.definitions = { - 'jails_group': {'options': [None, "Jails ban statistics", "bans/s", 'jails', 'jail.ban', 'line'], - 'lines': []}} - for jail in self.jails_list: - self.definitions['jails_group']['lines'].append([jail, jail, 'incremental']) - + def jails_auto_detection_(self): + """ + return: <tuple> -def parse_conf_dir(conf_dir): - if not isdir(conf_dir): - return list(), '%s is not a directory' % conf_dir + * jails_list - list of enabled jails (['ssh', 'apache', ...]) + * to_netdata - dict ({'ssh': 0, 'ssh_in_jail': 0, ...}) + * banned_ips - here will be stored all the banned ips ({'ssh': ['1.2.3.4', '5.6.7.8', ...], ...}) + """ + raw_jails_list = list() + jails_list = list() - jail_local = list(filter(lambda local: is_accessible(local, R_OK), glob(conf_dir + '/*.local'))) - jail_conf = list(filter(lambda conf: is_accessible(conf, R_OK), glob(conf_dir + '/*.conf'))) + for raw_jail in parse_configuration_files_(self.conf_path, self.conf_dir, self.error): + raw_jails_list.extend(raw_jail) - if not (jail_local or jail_conf): - return list(), '%s is empty or not readable' % conf_dir + for jail, status in raw_jails_list: + if status == 'true' and jail not in jails_list: + jails_list.append(jail) + elif status == 'false' and jail in jails_list: + jails_list.remove(jail) - # According "man jail.conf" files could be *.local AND *.conf - # *.conf files parsed first. Changes in *.local overrides configuration in *.conf - if jail_conf: - jail_local.extend([conf for conf in jail_conf if conf[:-5] not in [local[:-6] for local in jail_local]]) + # If for some reason parse failed we still can START with default jails_list. + jails_list = list(set(jails_list) - set(self.exclude.split() + if isinstance(self.exclude, str) else list())) or ['ssh'] + + to_netdata = dict([(jail, 0) for jail in jails_list]) + to_netdata.update(dict([(jail + '_in_jail', 0) for jail in jails_list])) + banned_ips = dict([(jail, list()) for jail in jails_list]) + + return jails_list, to_netdata, banned_ips + + +def create_definitions_(jails_list): + """ + Chart definitions creating + """ + + definitions = { + 'jails_bans': {'options': [None, 'Jails Ban Statistics', 'bans/s', 'bans', 'jail.bans', 'line'], + 'lines': []}, + 'jails_in_jail': {'options': [None, 'Banned IPs (since the last restart of netdata)', 'IPs', + 'in jail', 'jail.in_jail', 'line'], + 'lines': []}} + for jail in jails_list: + definitions['jails_bans']['lines'].append([jail, jail, 'incremental']) + definitions['jails_in_jail']['lines'].append([jail + '_in_jail', jail, 'absolute']) + + return definitions + + +def parse_configuration_files_(jails_conf_path, jails_conf_dir, print_error): + """ + :param jails_conf_path: <str> + :param jails_conf_dir: <str> + :param print_error: <function> + :return: <tuple> + + Uses "find_jails_in_files" function to find all jails in the "jails_conf_dir" directory + and in the "jails_conf_path" + + All files must endswith ".local" or ".conf" + Return order is important. + According man jail.conf it should be + * jail.conf + * jail.d/*.conf (in alphabetical order) + * jail.local + * jail.d/*.local (in alphabetical order) + """ + path_conf, path_local, dir_conf, dir_local = list(), list(), list(), list() + + # Parse files in the directory + if not (isinstance(jails_conf_dir, str) and isdir(jails_conf_dir)): + print_error('%s is not a directory' % jails_conf_dir) + else: + dir_conf = list(filter(lambda conf: is_accessible(conf, R_OK), glob(jails_conf_dir + '/*.conf'))) + dir_local = list(filter(lambda local: is_accessible(local, R_OK), glob(jails_conf_dir + '/*.local'))) + if not (dir_conf or dir_local): + print_error('%s is empty or not readable' % jails_conf_dir) + else: + dir_conf, dir_local = (find_jails_in_files(dir_conf, print_error), + find_jails_in_files(dir_local, print_error)) + + # Parse .conf and .local files + if isinstance(jails_conf_path, str) and jails_conf_path.endswith(('.local', '.conf')): + path_conf, path_local = (find_jails_in_files([jails_conf_path.split('.')[0] + '.conf'], print_error), + find_jails_in_files([jails_conf_path.split('.')[0] + '.local'], print_error)) + + return path_conf, dir_conf, path_local, dir_local + + +def find_jails_in_files(list_of_files, print_error): + """ + :param list_of_files: <list> + :param print_error: <function> + :return: <list> + + Open a file and parse it to find all (enabled and disabled) jails + The output is a list of tuples: + [('ssh', 'true'), ('apache', 'false'), ...] + """ jails_list = list() - for conf in jail_local: - with open(conf, 'rt') as f: - raw_data = f.read() - - data = ' '.join(raw_data.split()) - jails_list.extend(REGEX.findall(data)) - jails_list = list(set(jails_list)) - - return jails_list, 'can\'t locate any jails in %s. Default jail is [\'ssh\']' % conf_dir - - -def parse_conf_path(conf_path): - if not is_accessible(conf_path, R_OK): - return list(), '%s is not readable' % conf_path - - with open(conf_path, 'rt') as jails_conf: - raw_data = jails_conf.read() - - data = raw_data.split() - jails_list = REGEX.findall(' '.join(data)) - return jails_list, 'can\'t locate any jails in %s. Default jail is [\'ssh\']' % conf_path + for conf in list_of_files: + if is_accessible(conf, R_OK): + with open(conf, 'rt') as conf: + raw_data = conf.read() + data = ' '.join(raw_data.split()) + jails_list.extend(REGEX_JAILS.findall(data)) + else: + print_error('%s is not readable or not exist' % conf) + return jails_list + + +def address_not_in_jail(pool, address, pool_size): + """ + :param pool: <list> + :param address: <str> + :param pool_size: <int> + :return: bool + + Checks if the address is in the pool. + If not address will be added + """ + index = bisect.bisect_left(pool, address) + if index < pool_size: + if pool[index] == address: + return False + bisect.insort_left(pool, address) + return True + else: + bisect.insort_left(pool, address) + return True diff --git a/python.d/go_expvar.chart.py b/python.d/go_expvar.chart.py new file mode 100644 index 00000000..e1a334cc --- /dev/null +++ b/python.d/go_expvar.chart.py @@ -0,0 +1,228 @@ +# -*- coding: utf-8 -*- +# Description: go_expvar netdata python.d module +# Author: Jan Kral (kralewitz) + +from __future__ import division +from base import UrlService +import json + +# default module values (can be overridden per job in `config`) +# update_every = 2 +priority = 60000 +retries = 60 + + +MEMSTATS_CHARTS = { + 'memstats_heap': { + 'options': ['heap', 'memory: size of heap memory structures', 'kB', 'memstats', 'expvar.memstats.heap', 'line'], + 'lines': [ + ['memstats_heap_alloc', 'alloc', 'absolute', 1, 1024], + ['memstats_heap_inuse', 'inuse', 'absolute', 1, 1024] + ]}, + 'memstats_stack': { + 'options': ['stack', 'memory: size of stack memory structures', 'kB', 'memstats', 'expvar.memstats.stack', 'line'], + 'lines': [ + ['memstats_stack_inuse', 'inuse', 'absolute', 1, 1024] + ]}, + 'memstats_mspan': { + 'options': ['mspan', 'memory: size of mspan memory structures', 'kB', 'memstats', 'expvar.memstats.mspan', 'line'], + 'lines': [ + ['memstats_mspan_inuse', 'inuse', 'absolute', 1, 1024] + ]}, + 'memstats_mcache': { + 'options': ['mcache', 'memory: size of mcache memory structures', 'kB', 'memstats', 'expvar.memstats.mcache', 'line'], + 'lines': [ + ['memstats_mcache_inuse', 'inuse', 'absolute', 1, 1024] + ]}, + 'memstats_live_objects': { + 'options': ['live_objects', 'memory: number of live objects', 'objects', 'memstats', 'expvar.memstats.live_objects', 'line'], + 'lines': [ + ['memstats_live_objects', 'live'] + ]}, + 'memstats_sys': { + 'options': ['sys', 'memory: size of reserved virtual address space', 'kB', 'memstats', 'expvar.memstats.sys', 'line'], + 'lines': [ + ['memstats_sys', 'sys', 'absolute', 1, 1024] + ]}, + 'memstats_gc_pauses': { + 'options': ['gc_pauses', 'memory: average duration of GC pauses', 'ns', 'memstats', 'expvar.memstats.gc_pauses', 'line'], + 'lines': [ + ['memstats_gc_pauses', 'avg'] + ]}, +} + +MEMSTATS_ORDER = ['memstats_heap', 'memstats_stack', 'memstats_mspan', 'memstats_mcache', 'memstats_sys', 'memstats_live_objects', 'memstats_gc_pauses'] + + +def flatten(d, top='', sep='.'): + items = [] + for key, val in d.items(): + nkey = top + sep + key if top else key + if isinstance(val, dict): + items.extend(flatten(val, nkey, sep=sep).items()) + else: + items.append((nkey, val)) + return dict(items) + + +class Service(UrlService): + def __init__(self, configuration=None, name=None): + UrlService.__init__(self, configuration=configuration, name=name) + + # if memstats collection is enabled, add the charts and their order + if self.configuration.get('collect_memstats'): + self.definitions = MEMSTATS_CHARTS + self.order = MEMSTATS_ORDER + else: + self.definitions = dict() + self.order = list() + + # if extra charts are defined, parse their config + extra_charts = self.configuration.get('extra_charts') + if extra_charts: + self._parse_extra_charts_config(extra_charts) + + def check(self): + """ + Check if the module can collect data: + 1) At least one JOB configuration has to be specified + 2) The JOB configuration needs to define the URL and either collect_memstats must be enabled or at least one + extra_chart must be defined. + + The configuration and URL check is provided by the UrlService class. + """ + + if not (self.configuration.get('extra_charts') or self.configuration.get('collect_memstats')): + self.error('Memstats collection is disabled and no extra_charts are defined, disabling module.') + return False + + return UrlService.check(self) + + def _parse_extra_charts_config(self, extra_charts_config): + + # a place to store the expvar keys and their types + self.expvars = dict() + + for chart in extra_charts_config: + + chart_dict = dict() + chart_id = chart.get('id') + chart_lines = chart.get('lines') + chart_opts = chart.get('options', dict()) + + if not all([chart_id, chart_lines]): + self.info('Chart {0} has no ID or no lines defined, skipping'.format(chart)) + continue + + chart_dict['options'] = [ + chart_opts.get('name', ''), + chart_opts.get('title', ''), + chart_opts.get('units', ''), + chart_opts.get('family', ''), + chart_opts.get('context', ''), + chart_opts.get('chart_type', 'line') + ] + chart_dict['lines'] = list() + + # add the lines to the chart + for line in chart_lines: + + ev_key = line.get('expvar_key') + ev_type = line.get('expvar_type') + line_id = line.get('id') + + if not all([ev_key, ev_type, line_id]): + self.info('Line missing expvar_key, expvar_type, or line_id, skipping: {0}'.format(line)) + continue + + if ev_type not in ['int', 'float']: + self.info('Unsupported expvar_type "{0}". Must be "int" or "float"'.format(ev_type)) + continue + + if ev_key in self.expvars: + self.info('Duplicate expvar key {0}: skipping line.'.format(ev_key)) + continue + + self.expvars[ev_key] = (ev_type, line_id) + + chart_dict['lines'].append( + [ + line.get('id', ''), + line.get('name', ''), + line.get('algorithm', ''), + line.get('multiplier', 1), + line.get('divisor', 100 if ev_type == 'float' else 1), + line.get('hidden', False) + ] + ) + + self.order.append(chart_id) + self.definitions[chart_id] = chart_dict + + def _get_data(self): + """ + Format data received from http request + :return: dict + """ + + raw_data = self._get_raw_data() + if not raw_data: + return None + + data = json.loads(raw_data) + + expvars = dict() + if self.configuration.get('collect_memstats'): + expvars.update(self._parse_memstats(data)) + + if self.configuration.get('extra_charts'): + # the memstats part of the data has been already parsed, so we remove it before flattening and checking + # the rest of the data, thus avoiding needless iterating over the multiply nested memstats dict. + del (data['memstats']) + flattened = flatten(data) + for k, v in flattened.items(): + ev = self.expvars.get(k) + if not ev: + # expvar is not defined in config, skip it + continue + try: + key_type, line_id = ev + if key_type == 'int': + expvars[line_id] = int(v) + elif key_type == 'float': + # if the value type is float, multiply it by 1000 and set line divisor to 1000 + expvars[line_id] = float(v) * 100 + except ValueError: + self.info('Failed to parse value for key {0} as {1}, ignoring key.'.format(k, key_type)) + del self.expvars[k] + + return expvars + + @staticmethod + def _parse_memstats(data): + + memstats = data['memstats'] + + # calculate the number of live objects in memory + live_objs = int(memstats['Mallocs']) - int(memstats['Frees']) + + # calculate GC pause times average + # the Go runtime keeps the last 256 GC pause durations in a circular buffer, + # so we need to filter out the 0 values before the buffer is filled + gc_pauses = memstats['PauseNs'] + try: + gc_pause_avg = sum(gc_pauses) / len([x for x in gc_pauses if x > 0]) + # no GC cycles have occured yet + except ZeroDivisionError: + gc_pause_avg = 0 + + return { + 'memstats_heap_alloc': memstats['HeapAlloc'], + 'memstats_heap_inuse': memstats['HeapInuse'], + 'memstats_stack_inuse': memstats['StackInuse'], + 'memstats_mspan_inuse': memstats['MSpanInuse'], + 'memstats_mcache_inuse': memstats['MCacheInuse'], + 'memstats_sys': memstats['Sys'], + 'memstats_live_objects': live_objs, + 'memstats_gc_pauses': gc_pause_avg, + } diff --git a/python.d/haproxy.chart.py b/python.d/haproxy.chart.py index 2fb97d75..67a6f782 100644 --- a/python.d/haproxy.chart.py +++ b/python.d/haproxy.chart.py @@ -3,6 +3,13 @@ # Author: l2isbad from base import UrlService, SocketService +from collections import defaultdict +from re import compile as re_compile + +try: + from urlparse import urlparse +except ImportError: + from urllib.parse import urlparse # default module values (can be overridden per job in `config`) # update_every = 2 @@ -10,135 +17,140 @@ priority = 60000 retries = 60 # charts order (can be overridden if you want less charts, or different order) -ORDER = ['fbin', 'fbout', 'fscur', 'fqcur', 'bbin', 'bbout', 'bscur', 'bqcur', 'health_sdown', 'health_bdown'] +ORDER = ['fbin', 'fbout', 'fscur', 'fqcur', 'bbin', 'bbout', 'bscur', 'bqcur', + 'health_sdown', 'health_bdown', 'health_idle'] CHARTS = { 'fbin': { - 'options': [None, "Kilobytes in", "kilobytes in/s", 'Frontend', 'haproxy_f.bin', 'line'], + 'options': [None, "Kilobytes In", "KB/s", 'frontend', 'haproxy_f.bin', 'line'], 'lines': [ ]}, 'fbout': { - 'options': [None, "Kilobytes out", "kilobytes out/s", 'Frontend', 'haproxy_f.bout', 'line'], + 'options': [None, "Kilobytes Out", "KB/s", 'frontend', 'haproxy_f.bout', 'line'], 'lines': [ ]}, 'fscur': { - 'options': [None, "Sessions active", "sessions", 'Frontend', 'haproxy_f.scur', 'line'], + 'options': [None, "Sessions Active", "sessions", 'frontend', 'haproxy_f.scur', 'line'], 'lines': [ ]}, 'fqcur': { - 'options': [None, "Session in queue", "sessions", 'Frontend', 'haproxy_f.qcur', 'line'], + 'options': [None, "Session In Queue", "sessions", 'frontend', 'haproxy_f.qcur', 'line'], 'lines': [ ]}, 'bbin': { - 'options': [None, "Kilobytes in", "kilobytes in/s", 'Backend', 'haproxy_b.bin', 'line'], + 'options': [None, "Kilobytes In", "KB/s", 'backend', 'haproxy_b.bin', 'line'], 'lines': [ ]}, 'bbout': { - 'options': [None, "Kilobytes out", "kilobytes out/s", 'Backend', 'haproxy_b.bout', 'line'], + 'options': [None, "Kilobytes Out", "KB/s", 'backend', 'haproxy_b.bout', 'line'], 'lines': [ ]}, 'bscur': { - 'options': [None, "Sessions active", "sessions", 'Backend', 'haproxy_b.scur', 'line'], + 'options': [None, "Sessions Active", "sessions", 'backend', 'haproxy_b.scur', 'line'], 'lines': [ ]}, 'bqcur': { - 'options': [None, "Sessions in queue", "sessions", 'Backend', 'haproxy_b.qcur', 'line'], + 'options': [None, "Sessions In Queue", "sessions", 'backend', 'haproxy_b.qcur', 'line'], 'lines': [ ]}, 'health_sdown': { - 'options': [None, "Number of servers in backend in DOWN state", "failed servers", 'Health', 'haproxy_hs.down', 'line'], + 'options': [None, "Backend Servers In DOWN State", "failed servers", 'health', + 'haproxy_hs.down', 'line'], 'lines': [ ]}, 'health_bdown': { - 'options': [None, "Is backend alive? 1 = DOWN", "failed backend", 'Health', 'haproxy_hb.down', 'line'], + 'options': [None, "Is Backend Alive? 1 = DOWN", "failed backend", 'health', 'haproxy_hb.down', 'line'], + 'lines': [ + ]}, + 'health_idle': { + 'options': [None, "The Ratio Of Polling Time Vs Total Time", "percent", 'health', 'haproxy.idle', 'line'], 'lines': [ + ['idle', None, 'absolute'] ]} } +METRICS = {'bin': {'algorithm': 'incremental', 'divisor': 1024}, + 'bout': {'algorithm': 'incremental', 'divisor': 1024}, + 'scur': {'algorithm': 'absolute', 'divisor': 1}, + 'qcur': {'algorithm': 'absolute', 'divisor': 1}} + +REGEX = dict(url=re_compile(r'idle = (?P<idle>[0-9]+)'), socket=re_compile(r'Idle_pct: (?P<idle>[0-9]+)')) + class Service(UrlService, SocketService): def __init__(self, configuration=None, name=None): - SocketService.__init__(self, configuration=configuration, name=name) - self.user = self.configuration.get('user') - self.password = self.configuration.get('pass') - self.request = 'show stat\n' - self.poll_method = (UrlService, SocketService) + if 'socket' in configuration: + SocketService.__init__(self, configuration=configuration, name=name) + self.poll = SocketService + self.options_ = dict(regex=REGEX['socket'], stat='show stat\n', info='show info\n') + else: + UrlService.__init__(self, configuration=configuration, name=name) + self.poll = UrlService + self.options_ = dict(regex=REGEX['url'], stat=self.url, info=url_remove_params(self.url)) self.order = ORDER - self.order_front = [_ for _ in ORDER if _.startswith('f')] - self.order_back = [_ for _ in ORDER if _.startswith('b')] self.definitions = CHARTS - self.charts = True def check(self): - if self.configuration.get('url'): - self.poll_method = self.poll_method[0] - url = self.configuration.get('url') - if not url.endswith(';csv;norefresh'): - self.error('Bad url(%s). Must be http://<ip.address>:<port>/<url>;csv;norefresh' % url) - return False - elif self.configuration.get('socket'): - self.poll_method = self.poll_method[1] - else: - self.error('No configuration is specified') - return False - - if self.poll_method.check(self): - self.info('Plugin was started succesfully. We are using %s.' % self.poll_method.__name__) + if self.poll.check(self): + self.create_charts() + self.info('We are using %s.' % self.poll.__name__) return True + return False - def create_charts(self, front_ends, back_ends): - for _ in range(len(front_ends)): - self.definitions['fbin']['lines'].append(['_'.join(['fbin', front_ends[_]['# pxname']]), front_ends[_]['# pxname'], 'incremental', 1, 1024]) - self.definitions['fbout']['lines'].append(['_'.join(['fbout', front_ends[_]['# pxname']]), front_ends[_]['# pxname'], 'incremental', 1, 1024]) - self.definitions['fscur']['lines'].append(['_'.join(['fscur', front_ends[_]['# pxname']]), front_ends[_]['# pxname'], 'absolute']) - self.definitions['fqcur']['lines'].append(['_'.join(['fqcur', front_ends[_]['# pxname']]), front_ends[_]['# pxname'], 'absolute']) - - for _ in range(len(back_ends)): - self.definitions['bbin']['lines'].append(['_'.join(['bbin', back_ends[_]['# pxname']]), back_ends[_]['# pxname'], 'incremental', 1, 1024]) - self.definitions['bbout']['lines'].append(['_'.join(['bbout', back_ends[_]['# pxname']]), back_ends[_]['# pxname'], 'incremental', 1, 1024]) - self.definitions['bscur']['lines'].append(['_'.join(['bscur', back_ends[_]['# pxname']]), back_ends[_]['# pxname'], 'absolute']) - self.definitions['bqcur']['lines'].append(['_'.join(['bqcur', back_ends[_]['# pxname']]), back_ends[_]['# pxname'], 'absolute']) - self.definitions['health_sdown']['lines'].append(['_'.join(['hsdown', back_ends[_]['# pxname']]), back_ends[_]['# pxname'], 'absolute']) - self.definitions['health_bdown']['lines'].append(['_'.join(['hbdown', back_ends[_]['# pxname']]), back_ends[_]['# pxname'], 'absolute']) - def _get_data(self): - """ - Format data received from http request - :return: dict - """ - try: - raw_data = self.poll_method._get_raw_data(self).splitlines() - except Exception as e: - self.error(str(e)) - return None - - all_instances = [dict(zip(raw_data[0].split(','), raw_data[_].split(','))) for _ in range(1, len(raw_data))] - - back_ends = list(filter(is_backend, all_instances)) - front_ends = list(filter(is_frontend, all_instances)) - servers = list(filter(is_server, all_instances)) - - if self.charts: - self.create_charts(front_ends, back_ends) - self.charts = False - to_netdata = dict() + self.request, self.url = self.options_['stat'], self.options_['stat'] + stat_data = self._get_stat_data() + self.request, self.url = self.options_['info'], self.options_['info'] + info_data = self._get_info_data(regex=self.options_['regex']) - for frontend in front_ends: - for _ in self.order_front: - to_netdata.update({'_'.join([_, frontend['# pxname']]): int(frontend[_[1:]]) if frontend.get(_[1:]) else 0}) + to_netdata.update(stat_data) + to_netdata.update(info_data) + return to_netdata or None - for backend in back_ends: - for _ in self.order_back: - to_netdata.update({'_'.join([_, backend['# pxname']]): int(backend[_[1:]]) if backend.get(_[1:]) else 0}) - - for _ in range(len(back_ends)): - to_netdata.update({'_'.join(['hsdown', back_ends[_]['# pxname']]): - len([server for server in servers if is_server_down(server, back_ends, _)])}) - to_netdata.update({'_'.join(['hbdown', back_ends[_]['# pxname']]): 1 if is_backend_down(back_ends, _) else 0}) + def _get_stat_data(self): + """ + :return: dict + """ + raw_data = self.poll._get_raw_data(self) + + if not raw_data: + return dict() + + raw_data = raw_data.splitlines() + self.data = parse_data_([dict(zip(raw_data[0].split(','), raw_data[_].split(','))) + for _ in range(1, len(raw_data))]) + if not self.data: + return dict() + + stat_data = dict() + + for frontend in self.data['frontend']: + for metric in METRICS: + idx = frontend['# pxname'].replace('.', '_') + stat_data['_'.join(['frontend', metric, idx])] = frontend.get(metric) or 0 + + for backend in self.data['backend']: + name, idx = backend['# pxname'], backend['# pxname'].replace('.', '_') + stat_data['hsdown_' + idx] = len([server for server in self.data['servers'] + if server_down(server, name)]) + stat_data['hbdown_' + idx] = 1 if backend.get('status') == 'DOWN' else 0 + for metric in METRICS: + stat_data['_'.join(['backend', metric, idx])] = backend.get(metric) or 0 + return stat_data + + def _get_info_data(self, regex): + """ + :return: dict + """ + raw_data = self.poll._get_raw_data(self) + if not raw_data: + return dict() - return to_netdata + match = regex.search(raw_data) + return match.groupdict() if match else dict() - def _check_raw_data(self, data): + @staticmethod + def _check_raw_data(data): """ Check if all data has been gathered from socket :param data: str @@ -146,32 +158,54 @@ class Service(UrlService, SocketService): """ return not bool(data) -def is_backend(backend): - try: - return backend['svname'] == 'BACKEND' and backend['# pxname'] != 'stats' - except Exception: - return False - -def is_frontend(frontend): - try: - return frontend['svname'] == 'FRONTEND' and frontend['# pxname'] != 'stats' - except Exception: - return False - -def is_server(server): - try: - return not server['svname'].startswith(('FRONTEND', 'BACKEND')) - except Exception: - return False - -def is_server_down(server, back_ends, _): - try: - return server['# pxname'] == back_ends[_]['# pxname'] and server['status'] == 'DOWN' - except Exception: - return False - -def is_backend_down(back_ends, _): - try: - return back_ends[_]['status'] == 'DOWN' - except Exception: - return False + def create_charts(self): + for front in self.data['frontend']: + name, idx = front['# pxname'], front['# pxname'].replace('.', '_') + for metric in METRICS: + self.definitions['f' + metric]['lines'].append(['_'.join(['frontend', metric, idx]), + name, METRICS[metric]['algorithm'], 1, + METRICS[metric]['divisor']]) + for back in self.data['backend']: + name, idx = back['# pxname'], back['# pxname'].replace('.', '_') + for metric in METRICS: + self.definitions['b' + metric]['lines'].append(['_'.join(['backend', metric, idx]), + name, METRICS[metric]['algorithm'], 1, + METRICS[metric]['divisor']]) + self.definitions['health_sdown']['lines'].append(['hsdown_' + idx, name, 'absolute']) + self.definitions['health_bdown']['lines'].append(['hbdown_' + idx, name, 'absolute']) + + +def parse_data_(data): + def is_backend(backend): + return backend.get('svname') == 'BACKEND' and backend.get('# pxname') != 'stats' + + def is_frontend(frontend): + return frontend.get('svname') == 'FRONTEND' and frontend.get('# pxname') != 'stats' + + def is_server(server): + return not server.get('svname', '').startswith(('FRONTEND', 'BACKEND')) + + if not data: + return None + + result = defaultdict(list) + for elem in data: + if is_backend(elem): + result['backend'].append(elem) + continue + elif is_frontend(elem): + result['frontend'].append(elem) + continue + elif is_server(elem): + result['servers'].append(elem) + + return result or None + + +def server_down(server, backend_name): + return server.get('# pxname') == backend_name and server.get('status') == 'DOWN' + + +def url_remove_params(url): + parsed = urlparse(url or str()) + return '%s://%s%s' % (parsed.scheme, parsed.netloc, parsed.path) diff --git a/python.d/isc_dhcpd.chart.py b/python.d/isc_dhcpd.chart.py index bb9ba5cb..a437f803 100644 --- a/python.d/isc_dhcpd.chart.py +++ b/python.d/isc_dhcpd.chart.py @@ -2,30 +2,56 @@ # Description: isc dhcpd lease netdata python.d module # Author: l2isbad -from base import SimpleService from time import mktime, strptime, gmtime, time -from os import stat +from os import stat, access, R_OK +from os.path import isfile try: - from ipaddress import IPv4Address as ipaddress - from ipaddress import ip_network + from ipaddress import ip_network, ip_address have_ipaddress = True except ImportError: have_ipaddress = False try: - from itertools import filterfalse as filterfalse + from itertools import filterfalse except ImportError: from itertools import ifilterfalse as filterfalse - +from base import SimpleService priority = 60000 retries = 60 -update_every = 60 +update_every = 5 + +ORDER = ['pools_utilization', 'pools_active_leases', 'leases_total', 'parse_time', 'leases_size'] + +CHARTS = { + 'pools_utilization': { + 'options': [None, 'Pools Utilization', 'used in percent', 'utilization', + 'isc_dhcpd.utilization', 'line'], + 'lines': []}, + 'pools_active_leases': { + 'options': [None, 'Active Leases', 'leases per pool', 'active leases', + 'isc_dhcpd.active_leases', 'line'], + 'lines': []}, + 'leases_total': { + 'options': [None, 'Total All Pools', 'number', 'active leases', + 'isc_dhcpd.leases_total', 'line'], + 'lines': [['leases_total', 'leases', 'absolute']]}, + 'parse_time': { + 'options': [None, 'Parse Time', 'ms', 'parse stats', + 'isc_dhcpd.parse_time', 'line'], + 'lines': [['parse_time', 'time', 'absolute']]}, + 'leases_size': { + 'options': [None, 'Dhcpd Leases File Size', 'kilobytes', + 'parse stats', 'isc_dhcpd.leases_size', 'line'], + 'lines': [['leases_size', 'size', 'absolute', 1, 1024]]}} + class Service(SimpleService): def __init__(self, configuration=None, name=None): SimpleService.__init__(self, configuration=configuration, name=name) self.leases_path = self.configuration.get('leases_path', '/var/lib/dhcp/dhcpd.leases') - self.pools = self.configuration.get('pools') + self.order = ORDER + self.definitions = CHARTS + self.pools = dict() # Will work only with 'default' db-time-format (weekday year/month/day hour:minute:second) # TODO: update algorithm to parse correctly 'local' db-time-format @@ -33,50 +59,30 @@ class Service(SimpleService): # Also only ipv4 supported def check(self): - if not self._get_raw_data(): + if not have_ipaddress: + self.error('\'python-ipaddress\' module is needed') + return False + if not (isfile(self.leases_path) and access(self.leases_path, R_OK)): self.error('Make sure leases_path is correct and leases log file is readable by netdata') return False - elif not have_ipaddress: - self.error('No ipaddress module. Please install (py2-ipaddress in case of python2)') + if not self.configuration.get('pools'): + self.error('Pools are not defined') return False - else: + if not isinstance(self.configuration['pools'], dict): + self.error('Invalid \'pools\' format') + return False + + for pool in self.configuration['pools']: try: - self.pools = self.pools.split() - if not [ip_network(return_utf(pool)) for pool in self.pools]: - self.error('Pools list is empty') - return False - except (ValueError, IndexError, AttributeError, SyntaxError) as e: - self.error('Pools configurations is incorrect', str(e)) - return False - - # Creating static charts - self.order = ['parse_time', 'leases_size', 'utilization', 'total'] - self.definitions = {'utilization': - {'options': - [None, 'Pools utilization', 'used %', 'utilization', 'isc_dhcpd.util', 'line'], - 'lines': []}, - 'total': - {'options': - [None, 'Total all pools', 'leases', 'utilization', 'isc_dhcpd.total', 'line'], - 'lines': [['total', 'leases', 'absolute']]}, - 'parse_time': - {'options': - [None, 'Parse time', 'ms', 'parse stats', 'isc_dhcpd.parse', 'line'], - 'lines': [['ptime', 'time', 'absolute']]}, - 'leases_size': - {'options': - [None, 'dhcpd.leases file size', 'kilobytes', 'parse stats', 'isc_dhcpd.lsize', 'line'], - 'lines': [['lsize', 'size', 'absolute']]}} - # Creating dynamic charts - for pool in self.pools: - self.definitions['utilization']['lines'].append([''.join(['ut_', pool]), pool, 'absolute']) - self.order.append(''.join(['leases_', pool])) - self.definitions[''.join(['leases_', pool])] = \ - {'options': [None, 'Active leases', 'leases', 'pools', 'isc_dhcpd.lease', 'area'], - 'lines': [[''.join(['le_', pool]), pool, 'absolute']]} - - self.info('Plugin was started succesfully') - return True + net = ip_network(u'%s' % self.configuration['pools'][pool]) + self.pools[pool] = dict(net=net, num_hosts=net.num_addresses - 2) + except ValueError as error: + self.error('%s removed, error: %s' % (self.configuration['pools'][pool], error)) + + if not self.pools: + return False + self.create_charts() + return True def _get_raw_data(self): """ @@ -87,65 +93,60 @@ class Service(SimpleService): ) """ try: - with open(self.leases_path, 'rt') as dhcp_leases: + with open(self.leases_path) as leases: time_start = time() - part1 = filterfalse(find_lease, dhcp_leases) - part2 = filterfalse(find_ends, dhcp_leases) - raw_result = dict(zip(part1, part2)) + part1 = filterfalse(find_lease, leases) + part2 = filterfalse(find_ends, leases) + result = dict(zip(part1, part2)) time_end = time() file_parse_time = round((time_end - time_start) * 1000) - except Exception as e: - self.error("Failed to parse leases file:", str(e)) + return result, file_parse_time + except (OSError, IOError) as error: + self.error("Failed to parse leases file:", str(error)) return None - else: - result = (raw_result, file_parse_time) - return result def _get_data(self): """ :return: dict """ - raw_leases = self._get_raw_data() - if not raw_leases: + raw_data = self._get_raw_data() + if not raw_data: return None - # Result: {ipaddress: end lease time, ...} - all_leases = dict([(k[6:len(k)-3], v[7:len(v)-2]) for k, v in raw_leases[0].items()]) - - # Result: [active binding, active binding....]. (Expire time (ends date;) - current time > 0) - active_leases = [k for k, v in all_leases.items() if is_binding_active(all_leases[k])] - - # Result: {pool: number of active bindings in pool, ...} - pools_count = dict([(pool, len([lease for lease in active_leases if is_address_in(lease, pool)])) - for pool in self.pools]) - - # Result: {pool: number of host ip addresses in pool, ...} - pools_max = dict([(pool, (2 ** (32 - int(pool.split('/')[1])) - 2)) - for pool in self.pools]) - - # Result: {pool: % utilization, ....} (percent) - pools_util = dict([(pool, int(round(float(pools_count[pool]) / pools_max[pool] * 100, 0))) - for pool in self.pools]) - - # Bulding dicts to send to netdata - final_count = dict([(''.join(['le_', k]), v) for k, v in pools_count.items()]) - final_util = dict([(''.join(['ut_', k]), v) for k, v in pools_util.items()]) - - to_netdata = {'total': len(active_leases)} - to_netdata.update({'lsize': int(stat(self.leases_path)[6] / 1024)}) - to_netdata.update({'ptime': int(raw_leases[1])}) - to_netdata.update(final_util) - to_netdata.update(final_count) + raw_leases, parse_time = raw_data[0], raw_data[1] + # Result: {ipaddress: end lease time, ...} + active_leases, to_netdata = list(), dict() + current_time = mktime(gmtime()) + + for ip, lease_end_time in raw_leases.items(): + # Result: [active binding, active binding....]. (Expire time (ends date;) - current time > 0) + if binding_active(lease_end_time=lease_end_time[7:-2], + current_time=current_time): + active_leases.append(ip_address(u'%s' % ip[6:-3])) + + for pool in self.pools: + dim_id = pool.replace('.', '_') + pool_leases_count = len([ip for ip in active_leases if ip in self.pools[pool]['net']]) + to_netdata[dim_id + '_active_leases'] = pool_leases_count + to_netdata[dim_id + '_utilization'] = float(pool_leases_count) / self.pools[pool]['num_hosts'] * 10000 + + to_netdata['leases_total'] = len(active_leases) + to_netdata['leases_size'] = stat(self.leases_path)[6] + to_netdata['parse_time'] = parse_time return to_netdata + def create_charts(self): + for pool in self.pools: + dim, dim_id = pool, pool.replace('.', '_') + self.definitions['pools_utilization']['lines'].append([dim_id + '_utilization', + dim, 'absolute', 1, 100]) + self.definitions['pools_active_leases']['lines'].append([dim_id + '_active_leases', + dim, 'absolute']) -def is_binding_active(binding): - return mktime(strptime(binding, '%w %Y/%m/%d %H:%M:%S')) - mktime(gmtime()) > 0 - -def is_address_in(address, pool): - return ipaddress(return_utf(address)) in ip_network(return_utf(pool)) +def binding_active(lease_end_time, current_time): + return mktime(strptime(lease_end_time, '%w %Y/%m/%d %H:%M:%S')) - current_time > 0 def find_lease(value): @@ -155,10 +156,3 @@ def find_lease(value): def find_ends(value): return value[2:6] != 'ends' - -def return_utf(s): - # python2 returns "<type 'str'>" for simple strings - # python3 returns "<class 'str'>" for unicode strings - if str(type(s)) == "<type 'str'>": - return unicode(s, 'utf-8') - return s diff --git a/python.d/mdstat.chart.py b/python.d/mdstat.chart.py index 0f7d2b44..ca9aba56 100644 --- a/python.d/mdstat.chart.py +++ b/python.d/mdstat.chart.py @@ -2,98 +2,125 @@ # Description: mdstat netdata python.d module # Author: l2isbad +from re import compile as re_compile +from collections import defaultdict from base import SimpleService -from re import compile + priority = 60000 retries = 60 update_every = 1 +OPERATIONS = ('check', 'resync', 'reshape', 'recovery', 'finish', 'speed') + class Service(SimpleService): def __init__(self, configuration=None, name=None): SimpleService.__init__(self, configuration=configuration, name=name) - self.order = ['agr_health'] - self.definitions = {'agr_health': - {'options': - [None, 'Faulty devices in MD', 'failed disks', 'health', 'md.health', 'line'], - 'lines': []}} - self.proc_mdstat = '/proc/mdstat' - self.regex_disks = compile(r'((?<=\ )[a-zA-Z_0-9]+(?= : active)).*?((?<= \[)[0-9]+)/([0-9]+(?=\] ))') - self.regex_status = compile(r'([a-zA-Z_0-9]+)( : active)[^:]*?([a-z]+) = ([0-9.]+(?=%)).*?((?<=finish=)[0-9.]+)min speed=([0-9]+)') + self.regex = dict(disks=re_compile(r' (?P<array>[a-zA-Z_0-9]+) : active .+\[' + r'(?P<total_disks>[0-9]+)/' + r'(?P<inuse_disks>[0-9])\]'), + status=re_compile(r' (?P<array>[a-zA-Z_0-9]+) : active .+ ' + r'(?P<operation>[a-z]+) = ' + r'(?P<operation_status>[0-9.]+).+finish=' + r'(?P<finish>([0-9.]+))min speed=' + r'(?P<speed>[0-9]+)')) def check(self): - raw_data = self._get_raw_data() - if not raw_data: - self.error('Cant read mdstat data from %s' % (self.proc_mdstat)) - return False - - md_list = [md[0] for md in self.regex_disks.findall(raw_data)] - - if not md_list: - self.error('No active arrays in %s' % (self.proc_mdstat)) - return False - else: - for md in md_list: - self.order.append(md) - self.order.append(''.join([md, '_status'])) - self.order.append(''.join([md, '_rate'])) - self.definitions['agr_health']['lines'].append([''.join([md, '_health']), md, 'absolute']) - self.definitions[md] = {'options': - [None, '%s disks stats' % md, 'disks', md, 'md.disks', 'stacked'], - 'lines': [[''.join([md, '_total']), 'total', 'absolute'], - [''.join([md, '_inuse']), 'inuse', 'absolute']]} - self.definitions[''.join([md, '_status'])] = {'options': - [None, '%s current status' % md, 'percent', md, 'md.status', 'line'], - 'lines': [[''.join([md, '_resync']), 'resync', 'absolute', 1, 100], - [''.join([md, '_recovery']), 'recovery', 'absolute', 1, 100], - [''.join([md, '_reshape']), 'reshape', 'absolute', 1, 100], - [''.join([md, '_check']), 'check', 'absolute', 1, 100]]} - self.definitions[''.join([md, '_rate'])] = {'options': - [None, '%s operation status' % md, 'rate', md, 'md.rate', 'line'], - 'lines': [[''.join([md, '_finishin']), 'finish min', 'absolute', 1, 100], - [''.join([md, '_rate']), 'megabyte/s', 'absolute', -1, 100]]} - self.info('Plugin was started successfully. MDs to monitor %s' % (md_list)) - - return True - - def _get_raw_data(self): + arrays = find_arrays(self._get_raw_data(), self.regex) + if not arrays: + self.error('Failed to read data from /proc/mdstat or there is no active arrays') + return None + + self.order, self.definitions = create_charts(arrays.keys()) + return True + + @staticmethod + def _get_raw_data(): """ Read data from /proc/mdstat :return: str """ try: - with open(self.proc_mdstat, 'rt') as proc_mdstat: - raw_result = proc_mdstat.read() - except Exception: + with open('/proc/mdstat', 'rt') as proc_mdstat: + return proc_mdstat.readlines() or None + except (OSError, IOError): return None - else: - raw_result = ' '.join(raw_result.split()) - return raw_result def _get_data(self): """ Parse data from _get_raw_data() :return: dict """ - raw_mdstat = self._get_raw_data() - mdstat_disks = self.regex_disks.findall(raw_mdstat) - mdstat_status = self.regex_status.findall(raw_mdstat) - to_netdata = {} - - for md in mdstat_disks: - to_netdata[''.join([md[0], '_total'])] = int(md[1]) - to_netdata[''.join([md[0], '_inuse'])] = int(md[2]) - to_netdata[''.join([md[0], '_health'])] = int(md[1]) - int(md[2]) - to_netdata[''.join([md[0], '_check'])] = 0 - to_netdata[''.join([md[0], '_resync'])] = 0 - to_netdata[''.join([md[0], '_reshape'])] = 0 - to_netdata[''.join([md[0], '_recovery'])] = 0 - to_netdata[''.join([md[0], '_finishin'])] = 0 - to_netdata[''.join([md[0], '_rate'])] = 0 - - for md in mdstat_status: - to_netdata[''.join([md[0], '_' + md[2]])] = round(float(md[3]) * 100) - to_netdata[''.join([md[0], '_finishin'])] = round(float(md[4]) * 100) - to_netdata[''.join([md[0], '_rate'])] = round(float(md[5]) / 1000 * 100) + raw_data = self._get_raw_data() + arrays = find_arrays(raw_data, self.regex) + if not arrays: + return None + + to_netdata = dict() + for array, values in arrays.items(): + for key, value in values.items(): + to_netdata['_'.join([array, key])] = value return to_netdata + + +def find_arrays(raw_data, regex): + if raw_data is None: + return None + data = defaultdict(str) + counter = 1 + + for row in (elem.strip() for elem in raw_data): + if not row: + counter += 1 + continue + data[counter] = ' '.join([data[counter], row]) + + arrays = dict() + for value in data.values(): + match = regex['disks'].search(value) + if not match: + continue + + match = match.groupdict() + array = match.pop('array') + arrays[array] = match + arrays[array]['health'] = int(match['total_disks']) - int(match['inuse_disks']) + for operation in OPERATIONS: + arrays[array][operation] = 0 + + match = regex['status'].search(value) + if match: + match = match.groupdict() + if match['operation'] in OPERATIONS: + arrays[array][match['operation']] = float(match['operation_status']) * 100 + arrays[array]['finish'] = float(match['finish']) * 100 + arrays[array]['speed'] = float(match['speed']) / 1000 * 100 + + return arrays or None + + +def create_charts(arrays): + order = ['mdstat_health'] + definitions = dict(mdstat_health={'options': [None, 'Faulty devices in MD', 'failed disks', + 'health', 'md.health', 'line'], + 'lines': []}) + for md in arrays: + order.append(md) + order.append(md + '_status') + order.append(md + '_rate') + definitions['mdstat_health']['lines'].append([md + '_health', md, 'absolute']) + definitions[md] = {'options': [None, '%s disks stats' % md, 'disks', md, 'md.disks', 'stacked'], + 'lines': [[md + '_total_disks', 'total', 'absolute'], + [md + '_inuse_disks', 'inuse', 'absolute']]} + definitions[md + '_status'] = {'options': [None, '%s current status' % md, + 'percent', md, 'md.status', 'line'], + 'lines': [[md + '_resync', 'resync', 'absolute', 1, 100], + [md + '_recovery', 'recovery', 'absolute', 1, 100], + [md + '_reshape', 'reshape', 'absolute', 1, 100], + [md + '_check', 'check', 'absolute', 1, 100]]} + definitions[md + '_rate'] = {'options': [None, '%s operation status' % md, + 'rate', md, 'md.rate', 'line'], + 'lines': [[md + '_finish', 'finish min', 'absolute', 1, 100], + [md + '_speed', 'MB/s', 'absolute', -1, 100]]} + return order, definitions diff --git a/python.d/mongodb.chart.py b/python.d/mongodb.chart.py index c01bd293..bb4c44b0 100644 --- a/python.d/mongodb.chart.py +++ b/python.d/mongodb.chart.py @@ -19,7 +19,7 @@ except ImportError: priority = 60000 retries = 60 -REPLSET_STATES = [ +REPL_SET_STATES = [ ('1', 'primary'), ('8', 'down'), ('2', 'secondary'), @@ -358,8 +358,8 @@ CHARTS = { 'options': [None, 'Lock on the oplog. Number of times the lock was acquired in the specified mode', 'locks', 'locks metrics', 'mongodb.locks_oplog', 'stacked'], 'lines': [ - ['Metadata_r', 'intent_shared', 'incremental'], - ['Metadata_w', 'intent_exclusive', 'incremental'] + ['oplog_r', 'intent_shared', 'incremental'], + ['oplog_w', 'intent_exclusive', 'incremental'] ]} } @@ -391,13 +391,16 @@ class Service(SimpleService): self.build_metrics_to_collect_(server_status) try: - self._get_data() + data = self._get_data() except (LookupError, SyntaxError, AttributeError): self.error('Type: %s, error: %s' % (str(exc_info()[0]), str(exc_info()[1]))) return False - else: + if isinstance(data, dict) and data: + self._data_from_check = data self.create_charts_(server_status) return True + self.error('_get_data() returned no data or type is not <dict>') + return False def build_metrics_to_collect_(self, server_status): @@ -473,7 +476,7 @@ class Service(SimpleService): lines.append([dim_id, description, 'absolute', 1, 1]) return lines - all_hosts = server_status['repl']['hosts'] + all_hosts = server_status['repl']['hosts'] + server_status['repl'].get('arbiters', list()) this_host = server_status['repl']['me'] other_hosts = [host for host in all_hosts if host != this_host] @@ -503,19 +506,19 @@ class Service(SimpleService): self.definitions[chart_name] = { 'options': [None, 'Replica set member (%s) current state' % host, 'state', 'replication and oplog', 'mongodb.replication_state', 'line'], - 'lines': create_state_lines(REPLSET_STATES)} + 'lines': create_state_lines(REPL_SET_STATES)} def _get_raw_data(self): raw_data = dict() - raw_data.update(self.get_serverstatus_() or dict()) - raw_data.update(self.get_dbstats_() or dict()) - raw_data.update(self.get_replsetgetstatus_() or dict()) - raw_data.update(self.get_getreplicationinfo_() or dict()) + raw_data.update(self.get_server_status() or dict()) + raw_data.update(self.get_db_stats() or dict()) + raw_data.update(self.get_repl_set_get_status() or dict()) + raw_data.update(self.get_get_replication_info() or dict()) return raw_data or None - def get_serverstatus_(self): + def get_server_status(self): raw_data = dict() try: raw_data['serverStatus'] = self.connection.admin.command('serverStatus') @@ -524,7 +527,7 @@ class Service(SimpleService): else: return raw_data - def get_dbstats_(self): + def get_db_stats(self): if not self.databases: return None @@ -533,24 +536,22 @@ class Service(SimpleService): try: for dbase in self.databases: raw_data['dbStats'][dbase] = self.connection[dbase].command('dbStats') + return raw_data except PyMongoError: return None - else: - return raw_data - def get_replsetgetstatus_(self): + def get_repl_set_get_status(self): if not self.do_replica: return None raw_data = dict() try: raw_data['replSetGetStatus'] = self.connection.admin.command('replSetGetStatus') + return raw_data except PyMongoError: return None - else: - return raw_data - def get_getreplicationinfo_(self): + def get_get_replication_info(self): if not (self.do_replica and 'local' in self.databases): return None @@ -561,10 +562,9 @@ class Service(SimpleService): "$natural", ASCENDING).limit(1)[0] raw_data['getReplicationInfo']['DESCENDING'] = self.connection.local.oplog.rs.find().sort( "$natural", DESCENDING).limit(1)[0] + return raw_data except PyMongoError: return None - else: - return raw_data def _get_data(self): """ @@ -583,7 +583,7 @@ class Service(SimpleService): utc_now = datetime.utcnow() # serverStatus - for metric, new_name, function in self.metrics_to_collect: + for metric, new_name, func in self.metrics_to_collect: value = serverStatus for key in metric.split('.'): try: @@ -592,7 +592,7 @@ class Service(SimpleService): break if not isinstance(value, dict) and key: - to_netdata[new_name or key] = value if not function else function(value) + to_netdata[new_name or key] = value if not func else func(value) to_netdata['nonmapped'] = to_netdata['virtual'] - serverStatus['mem'].get('mappedWithJournal', to_netdata['mapped']) @@ -620,13 +620,13 @@ class Service(SimpleService): if not member.get('self'): other_hosts.append(member) # Replica set time diff between current time and time when last entry from the oplog was applied - if member['optimeDate'] != unix_epoch: + if member.get('optimeDate', unix_epoch) != unix_epoch: member_optimedate = member['name'] + '_optimedate' to_netdata.update({member_optimedate: int(delta_calculation(delta=utc_now - member['optimeDate'], multiplier=1000))}) # Replica set members state member_state = member['name'] + '_state' - for elem in REPLSET_STATES: + for elem in REPL_SET_STATES: state = elem[0] to_netdata.update({'_'.join([member_state, state]): 0}) to_netdata.update({'_'.join([member_state, str(member['state'])]): member['state']}) @@ -668,5 +668,4 @@ class Service(SimpleService): def delta_calculation(delta, multiplier=1): if hasattr(delta, 'total_seconds'): return delta.total_seconds() * multiplier - else: - return (delta.microseconds + (delta.seconds + delta.days * 24 * 3600) * 10 ** 6) / 10.0 ** 6 * multiplier + return (delta.microseconds + (delta.seconds + delta.days * 24 * 3600) * 10 ** 6) / 10.0 ** 6 * multiplier diff --git a/python.d/mysql.chart.py b/python.d/mysql.chart.py index abf6bf71..cdabe971 100644 --- a/python.d/mysql.chart.py +++ b/python.d/mysql.chart.py @@ -117,7 +117,7 @@ GLOBAL_STATS = [ 'Connection_errors_tcpwrap'] def slave_seconds(value): - return value if value else -1 + return value if value is not '' else -1 def slave_running(value): return 1 if value == 'Yes' else -1 @@ -278,10 +278,10 @@ CHARTS = { 'innodb_rows': { 'options': [None, 'mysql InnoDB Row Operations', 'operations/s', 'innodb', 'mysql.innodb_rows', 'area'], 'lines': [ - ['Innodb_rows_inserted', 'read', 'incremental'], - ['Innodb_rows_read', 'deleted', 'incremental', -1, 1], - ['Innodb_rows_updated', 'inserted', 'incremental', 1, 1], - ['Innodb_rows_deleted', 'updated', 'incremental', -1, 1], + ['Innodb_rows_inserted', 'inserted', 'incremental'], + ['Innodb_rows_read', 'read', 'incremental', 1, 1], + ['Innodb_rows_updated', 'updated', 'incremental', 1, 1], + ['Innodb_rows_deleted', 'deleted', 'incremental', -1, 1], ]}, 'innodb_buffer_pool_pages': { 'options': [None, 'mysql InnoDB Buffer Pool Pages', 'pages', 'innodb', 'mysql.innodb_buffer_pool_pages', 'line'], diff --git a/python.d/ovpn_status_log.chart.py b/python.d/ovpn_status_log.chart.py index c5fca002..3a7e8200 100644 --- a/python.d/ovpn_status_log.chart.py +++ b/python.d/ovpn_status_log.chart.py @@ -2,8 +2,10 @@ # Description: openvpn status log netdata python.d module # Author: l2isbad +from re import compile as r_compile +from collections import defaultdict from base import SimpleService -from re import compile, findall, search, subn + priority = 60000 retries = 60 update_every = 10 @@ -11,67 +13,101 @@ update_every = 10 ORDER = ['users', 'traffic'] CHARTS = { 'users': { - 'options': [None, 'OpenVPN active users', 'active users', 'Users', 'openvpn_status.users', 'line'], + 'options': [None, 'OpenVPN Active Users', 'active users', 'users', 'openvpn_status.users', 'line'], 'lines': [ - ["users", None, "absolute"], + ['users', None, 'absolute'], ]}, 'traffic': { - 'options': [None, 'OpenVPN traffic', 'kilobit/s', 'Traffic', 'openvpn_status.traffic', 'area'], + 'options': [None, 'OpenVPN Traffic', 'KB/s', 'traffic', 'openvpn_status.traffic', 'area'], 'lines': [ - ["in", None, "incremental", 8, 1000], ["out", None, "incremental", 8, -1000] + ['bytes_in', 'in', 'incremental', 1, 1 << 10], ['bytes_out', 'out', 'incremental', 1, -1 << 10] ]}, } + class Service(SimpleService): def __init__(self, configuration=None, name=None): SimpleService.__init__(self, configuration=configuration, name=name) self.order = ORDER self.definitions = CHARTS self.log_path = self.configuration.get('log_path') - self.regex_data_inter = compile(r'(?<=Since ).*?(?=.ROUTING)') - self.regex_data_final = compile(r'\d{1,3}(?:\.\d{1,3}){3}[:0-9,. ]*') - self.regex_users = compile(r'\d{1,3}(?:\.\d{1,3}){3}:\d+') - self.regex_traffic = compile(r'(?<=(?:,| ))\d+(?=(?:,| ))') + self.regex = dict(tls=r_compile(r'\d{1,3}(?:\.\d{1,3}){3}(?::\d+)? (?P<bytes_in>\d+) (?P<bytes_out>\d+)'), + static_key=r_compile(r'TCP/[A-Z]+ (?P<direction>(?:read|write)) bytes,(?P<bytes>\d+)')) + self.to_netdata = dict(bytes_in=0, bytes_out=0) def check(self): - if not self._get_raw_data(): - self.error('Make sure that the openvpn status log file exists and netdata has permission to read it') + if not (self.log_path and isinstance(self.log_path, str)): + self.error('\'log_path\' is not defined') return False - else: - self.info('Plugin was started succesfully') + + data = False + for method in (self._get_data_tls, self._get_data_static_key): + data = method() + if data: + self._get_data = method + self._data_from_check = data + break + + if data: return True + self.error('Make sure that the openvpn status log file exists and netdata has permission to read it') + return False def _get_raw_data(self): """ Open log file :return: str """ + try: - with open(self.log_path, 'rt') as log: - result = log.read() - except Exception: + with open(self.log_path) as log: + raw_data = log.readlines() or None + except OSError: return None else: - return result + return raw_data - def _get_data(self): + def _get_data_static_key(self): """ Parse openvpn-status log file. - Current regex version is ok for status-version 1, 2 and 3. Both users and bytes in/out are collecting. """ raw_data = self._get_raw_data() - try: - data_inter = self.regex_data_inter.search(' '.join(raw_data.splitlines())).group() - except AttributeError: - data_inter = '' + if not raw_data: + return None + + data = defaultdict(lambda: 0) + + for row in raw_data: + match = self.regex['static_key'].search(row) + if match: + match = match.groupdict() + if match['direction'] == 'read': + data['bytes_in'] += int(match['bytes']) + else: + data['bytes_out'] += int(match['bytes']) + + return data or None + + def _get_data_tls(self): + """ + Parse openvpn-status log file. + """ + + raw_data = self._get_raw_data() + if not raw_data: + return None - data_final = ' '.join(self.regex_data_final.findall(data_inter)) - users = self.regex_users.subn('', data_final)[1] - traffic = self.regex_traffic.findall(data_final) + data = defaultdict(lambda: 0) + for row in raw_data: + row = ' '.join(row.split(',')) if ',' in row else ' '.join(row.split()) + match = self.regex['tls'].search(row) + if match: + match = match.groupdict() + data['users'] += 1 + data['bytes_in'] += int(match['bytes_in']) + data['bytes_out'] += int(match['bytes_out']) - bytes_in = sum([int(traffic[i]) for i in range(len(traffic)) if (i + 1) % 2 is 1]) - bytes_out = sum([int(traffic[i]) for i in range(len(traffic)) if (i + 1) % 2 is 0]) + return data or None - return {'users': users, 'in': bytes_in, 'out': bytes_out} diff --git a/python.d/postgres.chart.py b/python.d/postgres.chart.py index d359bb4f..ef710cb8 100644 --- a/python.d/postgres.chart.py +++ b/python.d/postgres.chart.py @@ -99,8 +99,8 @@ SELECT sum(conflicts) AS conflicts, pg_database_size(datname) AS size FROM pg_stat_database -WHERE NOT datname ~* '^template\d+' -GROUP BY database_name; +WHERE datname IN %(databases)s +GROUP BY datname; """, BGWRITER=""" SELECT @@ -146,7 +146,6 @@ SELECT current_setting('is_superuser') = 'on' AS is_superuser; QUERY_STATS = { QUERIES['DATABASE']: METRICS['DATABASE'], QUERIES['BACKENDS']: METRICS['BACKENDS'], - QUERIES['ARCHIVE']: METRICS['ARCHIVE'], QUERIES['LOCKS']: METRICS['LOCKS'] } @@ -242,6 +241,7 @@ class Service(SimpleService): self.definitions = deepcopy(CHARTS) self.table_stats = configuration.pop('table_stats', False) self.index_stats = configuration.pop('index_stats', False) + self.database_poll = configuration.pop('database_poll', None) self.configuration = configuration self.connection = False self.is_superuser = False @@ -281,6 +281,9 @@ class Service(SimpleService): is_superuser = check_if_superuser_(cursor, QUERIES['IF_SUPERUSER']) cursor.close() + if (self.database_poll and isinstance(self.database_poll, str)): + self.databases = [dbase for dbase in self.databases if dbase in self.database_poll.split()] or self.databases + self.locks_zeroed = populate_lock_types(self.databases) self.add_additional_queries_(is_superuser) self.create_dynamic_charts_() @@ -296,6 +299,7 @@ class Service(SimpleService): QUERY_STATS[QUERIES['TABLE_STATS']] = METRICS['TABLE_STATS'] if is_superuser: QUERY_STATS[QUERIES['BGWRITER']] = METRICS['BGWRITER'] + QUERY_STATS[QUERIES['ARCHIVE']] = METRICS['ARCHIVE'] def create_dynamic_charts_(self): @@ -328,7 +332,7 @@ class Service(SimpleService): return None def query_stats_(self, cursor, query, metrics): - cursor.execute(query) + cursor.execute(query, dict(databases=tuple(self.databases))) for row in cursor: for metric in metrics: dimension_id = '_'.join([row['database_name'], metric]) if 'database_name' in row else metric diff --git a/python.d/python_modules/base.py b/python.d/python_modules/base.py index 859300ec..a643cc6a 100644 --- a/python.d/python_modules/base.py +++ b/python.d/python_modules/base.py @@ -20,23 +20,20 @@ import time import os import socket -import select import threading -import msg import ssl from subprocess import Popen, PIPE from sys import exc_info - +from glob import glob +import re try: from urlparse import urlparse except ImportError: from urllib.parse import urlparse - try: import urllib.request as urllib2 except ImportError: import urllib2 - try: import MySQLdb PYMYSQL = True @@ -46,6 +43,7 @@ except ImportError: PYMYSQL = True except ImportError: PYMYSQL = False +import msg try: PATH = os.getenv('PATH').split(':') @@ -175,14 +173,15 @@ class SimpleService(threading.Thread): # it is important to do this in a loop # sleep() is interruptable while now < next: - self.debug("sleeping for", str(next - now), "secs to reach frequency of", str(step), "secs, now:", str(now), " next:", str(next), " penalty:", str(penalty)) + self.debug("sleeping for", str(next - now), "secs to reach frequency of", + str(step), "secs, now:", str(now), " next:", str(next), " penalty:", str(penalty)) time.sleep(next - now) now = float(time.time()) # do the job try: status = self._run_once() - except Exception as e: + except Exception: status = False if status: @@ -202,10 +201,12 @@ class SimpleService(threading.Thread): penalty = 600 self.retries_left = self.retries - self.alert("failed to collect data for " + str(self.retries) + " times - increasing penalty to " + str(penalty) + " sec and trying again") + self.alert("failed to collect data for " + str(self.retries) + + " times - increasing penalty to " + str(penalty) + " sec and trying again") else: - self.error("failed to collect data - " + str(self.retries_left) + " retries left - penalty: " + str(penalty) + " sec") + self.error("failed to collect data - " + str(self.retries_left) + + " retries left - penalty: " + str(penalty) + " sec") # --- CHART --- @@ -460,11 +461,42 @@ class SimpleService(threading.Thread): return next(('/'.join([p, binary]) for p in PATH if os.path.isfile('/'.join([p, binary])) and os.access('/'.join([p, binary]), os.X_OK))) - else: - return None + return None except StopIteration: return None + def _add_new_dimension(self, dimension_id, chart_name, dimension=None, algorithm='incremental', + multiplier=1, divisor=1, priority=65000): + """ + :param dimension_id: + :param chart_name: + :param dimension: + :param algorithm: + :param multiplier: + :param divisor: + :param priority: + :return: + """ + if not all([dimension_id not in self._dimensions, + chart_name in self.order, + chart_name in self.definitions]): + return + self._dimensions.append(dimension_id) + dimension_list = list(map(str, [dimension_id, + dimension if dimension else dimension_id, + algorithm, + multiplier, + divisor])) + self.definitions[chart_name]['lines'].append(dimension_list) + add_to_name = self.override_name or self.name + job_name = ('_'.join([self.__module__, re.sub('\s+', '_', add_to_name)]) + if add_to_name != 'None' else self.__module__) + chart = 'CHART {0}.{1} '.format(job_name, chart_name) + options = '"" "{0}" {1} "{2}" {3} {4} '.format(*self.definitions[chart_name]['options'][1:6]) + other = '{0} {1}\n'.format(priority, self.update_every) + new_dimension = "DIMENSION {0}\n".format(' '.join(dimension_list)) + print(chart + options + other + new_dimension) + class UrlService(SimpleService): def __init__(self, configuration=None, name=None): @@ -473,47 +505,73 @@ class UrlService(SimpleService): self.user = self.configuration.get('user') self.password = self.configuration.get('pass') self.ss_cert = self.configuration.get('ss_cert') + self.proxy = self.configuration.get('proxy') - def __add_openers(self): - def self_signed_cert(ss_cert): - if ss_cert: - try: - ctx = ssl.create_default_context() - ctx.check_hostname = False - ctx.verify_mode = ssl.CERT_NONE - return urllib2.build_opener(urllib2.HTTPSHandler(context=ctx)) - except AttributeError: - return None - else: - return None + def __add_openers(self, user=None, password=None, ss_cert=None, proxy=None, url=None): + user = user or self.user + password = password or self.password + ss_cert = ss_cert or self.ss_cert + proxy = proxy or self.proxy - self.opener = self_signed_cert(self.ss_cert) or urllib2.build_opener() + handlers = list() - # HTTP Basic Auth - if self.user and self.password: - url_parse = urlparse(self.url) + # HTTP Basic Auth handler + if all([user, password, isinstance(user, str), isinstance(password, str)]): + url = url or self.url + url_parse = urlparse(url) top_level_url = '://'.join([url_parse.scheme, url_parse.netloc]) passman = urllib2.HTTPPasswordMgrWithDefaultRealm() - passman.add_password(None, top_level_url, self.user, self.password) - self.opener.add_handler(urllib2.HTTPBasicAuthHandler(passman)) + passman.add_password(None, top_level_url, user, password) + handlers.append(urllib2.HTTPBasicAuthHandler(passman)) self.debug("Enabling HTTP basic auth") - def _get_raw_data(self, custom_url=None): + # HTTPS handler + # Self-signed certificate ignore + if ss_cert: + try: + ctx = ssl.create_default_context() + ctx.check_hostname = False + ctx.verify_mode = ssl.CERT_NONE + except AttributeError: + self.error('HTTPS self-signed certificate ignore not enabled') + else: + handlers.append(urllib2.HTTPSHandler(context=ctx)) + self.debug("Enabling HTTP self-signed certificate ignore") + + # PROXY handler + if proxy and isinstance(proxy, str) and not ss_cert: + handlers.append(urllib2.ProxyHandler(dict(http=proxy))) + self.debug("Enabling HTTP proxy handler (%s)" % proxy) + + opener = urllib2.build_opener(*handlers) + return opener + + def _build_opener(self, **kwargs): + try: + return self.__add_openers(**kwargs) + except TypeError as error: + self.error('build_opener() error:', str(error)) + return None + + def _get_raw_data(self, url=None, opener=None): """ Get raw data from http request :return: str """ - raw_data = None - f = None + data = None try: - f = self.opener.open(custom_url or self.url, timeout=self.update_every * 2) - raw_data = f.read().decode('utf-8', 'ignore') + opener = opener or self.opener + data = opener.open(url or self.url, timeout=self.update_every * 2) + raw_data = data.read().decode('utf-8', 'ignore') + except urllib2.URLError as error: + self.error('Url: %s. Error: %s' % (url or self.url, str(error))) + return None except Exception as error: - self.error('Url: %s. Error: %s' %(custom_url or self.url, str(error))) + self.error(str(error)) return None finally: - if f is not None: f.close() - + if data is not None: + data.close() return raw_data or None def check(self): @@ -525,7 +583,7 @@ class UrlService(SimpleService): self.error('URL is not defined or type is not <str>') return False - self.__add_openers() + self.opener = self.__add_openers() try: data = self._get_data() @@ -781,57 +839,69 @@ class SocketService(SimpleService): class LogService(SimpleService): def __init__(self, configuration=None, name=None): - self.log_path = "" - self._last_position = 0 - # self._log_reader = None SimpleService.__init__(self, configuration=configuration, name=name) + self.log_path = self.configuration.get('path') + self.__glob_path = self.log_path + self._last_position = 0 self.retries = 100000 # basically always retry + self.__re_find = dict(current=0, run=0, maximum=60) def _get_raw_data(self): """ Get log lines since last poll :return: list """ - lines = [] + lines = list() try: - if os.path.getsize(self.log_path) < self._last_position: + if self.__re_find['current'] == self.__re_find['run']: + self._find_recent_log_file() + size = os.path.getsize(self.log_path) + if size == self._last_position: + self.__re_find['current'] += 1 + return list() # return empty list if nothing has changed + elif size < self._last_position: self._last_position = 0 # read from beginning if file has shrunk - elif os.path.getsize(self.log_path) == self._last_position: - self.debug("Log file hasn't changed. No new data.") - return [] # return empty list if nothing has changed - with open(self.log_path, "r") as fp: + + with open(self.log_path) as fp: fp.seek(self._last_position) - for i, line in enumerate(fp): + for line in fp: lines.append(line) self._last_position = fp.tell() - except Exception as e: - self.error(str(e)) + self.__re_find['current'] = 0 + except (OSError, IOError) as error: + self.__re_find['current'] += 1 + self.error(str(error)) - if len(lines) != 0: - return lines - else: - self.error("No data collected.") - return None + return lines or None + + def _find_recent_log_file(self): + """ + :return: + """ + self.__re_find['run'] = self.__re_find['maximum'] + self.__re_find['current'] = 0 + self.__glob_path = self.__glob_path or self.log_path # workaround for modules w/o config files + path_list = glob(self.__glob_path) + if path_list: + self.log_path = max(path_list) + return True + return False def check(self): """ Parse basic configuration and check if log file exists :return: boolean """ - if self.name is not None or self.name != str(None): - self.name = "" - else: - self.name = str(self.name) - try: - self.log_path = str(self.configuration['path']) - except (KeyError, TypeError): - self.info("No path to log specified. Using: '" + self.log_path + "'") + if not self.log_path: + self.error("No path to log specified") + return None - if os.access(self.log_path, os.R_OK): + if all([self._find_recent_log_file(), + os.access(self.log_path, os.R_OK), + os.path.isfile(self.log_path)]): return True - else: - self.error("Cannot access file: '" + self.log_path + "'") - return False + self.error("Cannot access %s" % self.log_path) + return False def create(self): # set cursor at last byte of log file @@ -847,7 +917,7 @@ class ExecutableService(SimpleService): SimpleService.__init__(self, configuration=configuration, name=name) self.command = None - def _get_raw_data(self): + def _get_raw_data(self, stderr=False): """ Get raw data from executed command :return: <list> @@ -855,10 +925,11 @@ class ExecutableService(SimpleService): try: p = Popen(self.command, stdout=PIPE, stderr=PIPE) except Exception as error: - self.error("Executing command", self.command, "resulted in error:", str(error)) + self.error("Executing command", " ".join(self.command), "resulted in error:", str(error)) return None data = list() - for line in p.stdout.readlines(): + std = p.stderr if stderr else p.stdout + for line in std.readlines(): data.append(line.decode()) return data or None diff --git a/python.d/rabbitmq.chart.py b/python.d/rabbitmq.chart.py new file mode 100644 index 00000000..15a6a80f --- /dev/null +++ b/python.d/rabbitmq.chart.py @@ -0,0 +1,187 @@ +# -*- coding: utf-8 -*- +# Description: rabbitmq netdata python.d module +# Author: l2isbad + +from base import UrlService +from socket import gethostbyname, gaierror +try: + from queue import Queue +except ImportError: + from Queue import Queue +from threading import Thread +from collections import namedtuple +from json import loads + +# default module values (can be overridden per job in `config`) +update_every = 1 +priority = 60000 +retries = 60 + +METHODS = namedtuple('METHODS', ['get_data_function', 'url', 'stats']) + +NODE_STATS = [('fd_used', None), + ('mem_used', None), + ('sockets_used', None), + ('proc_used', None), + ('disk_free', None) + ] +OVERVIEW_STATS = [('object_totals.channels', None), + ('object_totals.consumers', None), + ('object_totals.connections', None), + ('object_totals.queues', None), + ('object_totals.exchanges', None), + ('queue_totals.messages_ready', None), + ('queue_totals.messages_unacknowledged', None), + ('message_stats.ack', None), + ('message_stats.redeliver', None), + ('message_stats.deliver', None), + ('message_stats.publish', None) + ] +ORDER = ['queued_messages', 'message_rates', 'global_counts', + 'file_descriptors', 'socket_descriptors', 'erlang_processes', 'memory', 'disk_space'] + +CHARTS = { + 'file_descriptors': { + 'options': [None, 'File Descriptors', 'descriptors', 'overview', + 'rabbitmq.file_descriptors', 'line'], + 'lines': [ + ['fd_used', 'used', 'absolute'] + ]}, + 'memory': { + 'options': [None, 'Memory', 'MB', 'overview', + 'rabbitmq.memory', 'line'], + 'lines': [ + ['mem_used', 'used', 'absolute', 1, 1024 << 10] + ]}, + 'disk_space': { + 'options': [None, 'Disk Space', 'GB', 'overview', + 'rabbitmq.disk_space', 'line'], + 'lines': [ + ['disk_free', 'free', 'absolute', 1, 1024 ** 3] + ]}, + 'socket_descriptors': { + 'options': [None, 'Socket Descriptors', 'descriptors', 'overview', + 'rabbitmq.sockets', 'line'], + 'lines': [ + ['sockets_used', 'used', 'absolute'] + ]}, + 'erlang_processes': { + 'options': [None, 'Erlang Processes', 'processes', 'overview', + 'rabbitmq.processes', 'line'], + 'lines': [ + ['proc_used', 'used', 'absolute'] + ]}, + 'global_counts': { + 'options': [None, 'Global Counts', 'counts', 'overview', + 'rabbitmq.global_counts', 'line'], + 'lines': [ + ['channels', None, 'absolute'], + ['consumers', None, 'absolute'], + ['connections', None, 'absolute'], + ['queues', None, 'absolute'], + ['exchanges', None, 'absolute'] + ]}, + 'queued_messages': { + 'options': [None, 'Queued Messages', 'messages', 'overview', + 'rabbitmq.queued_messages', 'stacked'], + 'lines': [ + ['messages_ready', 'ready', 'absolute'], + ['messages_unacknowledged', 'unacknowledged', 'absolute'] + ]}, + 'message_rates': { + 'options': [None, 'Message Rates', 'messages/s', 'overview', + 'rabbitmq.message_rates', 'stacked'], + 'lines': [ + ['ack', None, 'incremental'], + ['redeliver', None, 'incremental'], + ['deliver', None, 'incremental'], + ['publish', None, 'incremental'] + ]} +} + + +class Service(UrlService): + def __init__(self, configuration=None, name=None): + UrlService.__init__(self, configuration=configuration, name=name) + self.order = ORDER + self.definitions = CHARTS + self.host = self.configuration.get('host', '127.0.0.1') + self.port = self.configuration.get('port', 15672) + self.scheme = self.configuration.get('scheme', 'http') + + def check(self): + # We can't start if <host> AND <port> not specified + if not (self.host and self.port): + self.error('Host is not defined in the module configuration file') + return False + + # Hostname -> ip address + try: + self.host = gethostbyname(self.host) + except gaierror as error: + self.error(str(error)) + return False + + # Add handlers (auth, self signed cert accept) + url = '%s://%s:%s/api' % (self.scheme, self.host, self.port) + self.opener = self._build_opener(url=url) + if not self.opener: + return False + # Add methods + api_node = url + '/nodes' + api_overview = url + '/overview' + self.methods = [METHODS(get_data_function=self._get_overview_stats, url=api_node, stats=NODE_STATS), + METHODS(get_data_function=self._get_overview_stats, url=api_overview, stats=OVERVIEW_STATS)] + + result = self._get_data() + if not result: + self.error('_get_data() returned no data') + return False + self._data_from_check = result + return True + + def _get_data(self): + threads = list() + queue = Queue() + result = dict() + + for method in self.methods: + th = Thread(target=method.get_data_function, args=(queue, method.url, method.stats)) + th.start() + threads.append(th) + + for thread in threads: + thread.join() + result.update(queue.get()) + + return result or None + + def _get_overview_stats(self, queue, url, stats): + """ + Format data received from http request + :return: dict + """ + + raw_data = self._get_raw_data(url) + + if not raw_data: + return queue.put(dict()) + data = loads(raw_data) + data = data[0] if isinstance(data, list) else data + + to_netdata = fetch_data_(raw_data=data, metrics_list=stats) + return queue.put(to_netdata) + + +def fetch_data_(raw_data, metrics_list): + to_netdata = dict() + for metric, new_name in metrics_list: + value = raw_data + for key in metric.split('.'): + try: + value = value[key] + except (KeyError, TypeError): + break + if not isinstance(value, dict): + to_netdata[new_name or key] = value + return to_netdata diff --git a/python.d/redis.chart.py b/python.d/redis.chart.py index 61f4f6d6..4bc1d41f 100644 --- a/python.d/redis.chart.py +++ b/python.d/redis.chart.py @@ -100,13 +100,12 @@ class Service(SocketService): :return: dict """ if self.passwd: - info_request = self.request self.request = "AUTH " + self.passwd + "\r\n" raw = self._get_raw_data().strip() if raw != "+OK": self.error("invalid password") return None - self.request = info_request + self.request = "INFO\r\n" response = self._get_raw_data() if response is None: # error has already been logged @@ -155,7 +154,7 @@ class Service(SocketService): :return: boolean """ length = len(data) - supposed = data.split('\n')[0][1:] + supposed = data.split('\n')[0][1:-1] offset = len(supposed) + 4 # 1 dollar sing, 1 new line character + 1 ending sequence '\r\n' if not supposed.isdigit(): return True diff --git a/python.d/samba.chart.py b/python.d/samba.chart.py new file mode 100644 index 00000000..767c9746 --- /dev/null +++ b/python.d/samba.chart.py @@ -0,0 +1,124 @@ +# -*- coding: utf-8 -*- +# Description: samba netdata python.d module +# Author: Christopher Cox <chris_cox@endlessnow.com> +# +# The netdata user needs to be able to be able to sudo the smbstatus program +# without password: +# netdata ALL=(ALL) NOPASSWD: /usr/bin/smbstatus -P +# +# This makes calls to smbstatus -P +# +# This just looks at a couple of values out of syscall, and some from smb2. +# +# The Lesser Ops chart is merely a display of current counter values. They +# didn't seem to change much to me. However, if you notice something changing +# a lot there, bring one or more out into its own chart and make it incremental +# (like find and notify... good examples). + +from base import ExecutableService +import re + +# default module values (can be overridden per job in `config`) +update_every = 5 +priority = 60000 +retries = 60 + +ORDER = ['syscall_rw','smb2_rw','smb2_create_close','smb2_info','smb2_find','smb2_notify','smb2_sm_count'] + +CHARTS = { + 'syscall_rw': { + 'lines': [ + ['syscall_sendfile_bytes', 'sendfile', 'incremental', 1, 1024], + ['syscall_recvfile_bytes', 'recvfile', 'incremental', -1, 1024] + ], + 'options': [None, 'R/Ws', 'kilobytes/s', 'syscall', 'syscall.rw', 'area'] + }, + 'smb2_rw': { + 'lines': [ + ['smb2_read_outbytes', 'readout', 'incremental', 1, 1024], + ['smb2_write_inbytes', 'writein', 'incremental', -1, 1024], + ['smb2_read_inbytes', 'readin', 'incremental', 1, 1024], + ['smb2_write_outbytes', 'writeout', 'incremental', -1, 1024] + ], + 'options': [None, 'R/Ws', 'kilobytes/s', 'smb2', 'smb2.rw', 'area'] + }, + 'smb2_create_close': { + 'lines': [ + ['smb2_create_count', 'create', 'incremental', 1, 1], + ['smb2_close_count', 'close', 'incremental', -1, 1] + ], + 'options': [None, 'Create/Close', 'operations/s', 'smb2', 'smb2.create_close', 'line'] + }, + 'smb2_info': { + 'lines': [ + ['smb2_getinfo_count', 'getinfo', 'incremental', 1, 1], + ['smb2_setinfo_count', 'setinfo', 'incremental', -1, 1] + ], + 'options': [None, 'Info', 'operations/s', 'smb2', 'smb2.get_set_info', 'line'] + }, + 'smb2_find': { + 'lines': [ + ['smb2_find_count', 'find', 'incremental', 1, 1] + ], + 'options': [None, 'Find', 'operations/s', 'smb2', 'smb2.find', 'line'] + }, + 'smb2_notify': { + 'lines': [ + ['smb2_notify_count', 'notify', 'incremental', 1, 1] + ], + 'options': [None, 'Notify', 'operations/s', 'smb2', 'smb2.notify', 'line'] + }, + 'smb2_sm_count': { + 'lines': [ + ['smb2_tcon_count', 'tcon', 'absolute', 1, 1], + ['smb2_negprot_count', 'negprot', 'absolute', 1, 1], + ['smb2_tdis_count', 'tdis', 'absolute', 1, 1], + ['smb2_cancel_count', 'cancel', 'absolute', 1, 1], + ['smb2_logoff_count', 'logoff', 'absolute', 1, 1], + ['smb2_flush_count', 'flush', 'absolute', 1, 1], + ['smb2_lock_count', 'lock', 'absolute', 1, 1], + ['smb2_keepalive_count', 'keepalive', 'absolute', 1, 1], + ['smb2_break_count', 'break', 'absolute', 1, 1], + ['smb2_sessetup_count', 'sessetup', 'absolute', 1, 1] + ], + 'options': [None, 'Lesser Ops', 'count', 'smb2', 'smb2.sm_counters', 'stacked'] + } + } + + +class Service(ExecutableService): + def __init__(self, configuration=None, name=None): + ExecutableService.__init__(self, configuration=configuration, name=name) + self.order = ORDER + self.definitions = CHARTS + self.rgx_smb2 = re.compile(r'(smb2_[^:]+|syscall_.*file_bytes):\s+(\d+)') + + def check(self): + sudo_binary, smbstatus_binary = self.find_binary('sudo'), self.find_binary('smbstatus') + + if not (sudo_binary and smbstatus_binary): + self.error('Can\'t locate \'sudo\' or \'smbstatus\' binary') + return False + + self.command = [sudo_binary, '-v'] + err = self._get_raw_data(stderr=True) + if err: + self.error(''.join(err)) + return False + + self.command = ' '.join([sudo_binary, '-n', smbstatus_binary, '-P']) + + return ExecutableService.check(self) + + def _get_data(self): + """ + Format data received from shell command + :return: dict + """ + raw_data = self._get_raw_data() + if not raw_data: + return None + + parsed = self.rgx_smb2.findall(' '.join(raw_data)) + + return dict(parsed) or None diff --git a/python.d/smartd_log.chart.py b/python.d/smartd_log.chart.py index e8037237..4039c153 100644 --- a/python.d/smartd_log.chart.py +++ b/python.d/smartd_log.chart.py @@ -2,7 +2,7 @@ # Description: smart netdata python.d module # Author: l2isbad, vorph1 -from re import compile +from re import compile as r_compile from os import listdir, access, R_OK from os.path import isfile, join, getsize, basename, isdir try: @@ -101,7 +101,7 @@ NAMED_DISKS = namedtuple('disks', ['name', 'size', 'number']) class Service(SimpleService): def __init__(self, configuration=None, name=None): SimpleService.__init__(self, configuration=configuration, name=name) - self.regex = compile(r'(\d+);(\d+);(\d+)') + self.regex = r_compile(r'(\d+);(\d+);(\d+)') self.log_path = self.configuration.get('log_path', '/var/log/smartd') self.raw_values = self.configuration.get('raw_values') self.attr = self.configuration.get('smart_attributes', []) @@ -197,18 +197,19 @@ class Service(SimpleService): result.append(['_'.join([name, attrid]), name[:name.index('.')], 'absolute']) return result - # Add additional smart attributes to the ORDER. If something goes wrong we don't care. + # Use configured attributes, if present. If something goes wrong we don't care. + order = ORDER try: - ORDER.extend(list(set(self.attr.split()) & SMART_ATTR.keys() - set(ORDER))) + order = [attr for attr in self.attr.split() if attr in SMART_ATTR.keys()] or ORDER except Exception: pass - self.order = [''.join(['attrid', i]) for i in ORDER] + self.order = [''.join(['attrid', i]) for i in order] self.definitions = dict() units = 'raw' if self.raw_values else 'normalized' for k, v in dict([(k, v) for k, v in SMART_ATTR.items() if k in ORDER]).items(): self.definitions.update({''.join(['attrid', k]): { - 'options': [None, v, units, v, 'smartd.attrid' + k, 'line'], + 'options': [None, v, units, v.lower(), 'smartd.attrid' + k, 'line'], 'lines': create_lines(k)}}) def find_disks_in_log_path(log_path): diff --git a/python.d/web_log.chart.py b/python.d/web_log.chart.py index cbc8cd23..564c9f1d 100644 --- a/python.d/web_log.chart.py +++ b/python.d/web_log.chart.py @@ -1,37 +1,51 @@ # -*- coding: utf-8 -*- # Description: web log netdata python.d module # Author: l2isbad - -from base import LogService import re import bisect from os import access, R_OK from os.path import getsize -from collections import namedtuple +from collections import namedtuple, defaultdict from copy import deepcopy +try: + from itertools import filterfalse +except ImportError: + from itertools import ifilterfalse as filterfalse +from base import LogService +import msg + priority = 60000 retries = 60 -ORDER = ['response_statuses', 'response_codes', 'bandwidth', 'response_time', 'requests_per_url', 'http_method', - 'http_version', 'requests_per_ipproto', 'clients', 'clients_all'] -CHARTS = { +ORDER_APACHE_CACHE = ['apache_cache'] + +ORDER_WEB = ['response_statuses', 'response_codes', 'bandwidth', 'response_time', 'response_time_upstream', + 'requests_per_url', 'requests_per_user_defined', 'http_method', 'http_version', + 'requests_per_ipproto', 'clients', 'clients_all'] + +ORDER_SQUID = ['squid_response_statuses', 'squid_response_codes', 'squid_detailed_response_codes', + 'squid_method', 'squid_mime_type', 'squid_hier_code', 'squid_transport_methods', + 'squid_transport_errors', 'squid_code', 'squid_handling_opts', 'squid_object_types', + 'squid_cache_events', 'squid_bytes', 'squid_duration', 'squid_clients', 'squid_clients_all'] + +CHARTS_WEB = { 'response_codes': { 'options': [None, 'Response Codes', 'requests/s', 'responses', 'web_log.response_codes', 'stacked'], 'lines': [ - ['2xx', '2xx', 'incremental'], - ['5xx', '5xx', 'incremental'], - ['3xx', '3xx', 'incremental'], - ['4xx', '4xx', 'incremental'], - ['1xx', '1xx', 'incremental'], + ['2xx', None, 'incremental'], + ['5xx', None, 'incremental'], + ['3xx', None, 'incremental'], + ['4xx', None, 'incremental'], + ['1xx', None, 'incremental'], ['0xx', 'other', 'incremental'], - ['unmatched', 'unmatched', 'incremental'] + ['unmatched', None, 'incremental'] ]}, 'bandwidth': { - 'options': [None, 'Bandwidth', 'KB/s', 'bandwidth', 'web_log.bandwidth', 'area'], + 'options': [None, 'Bandwidth', 'kilobits/s', 'bandwidth', 'web_log.bandwidth', 'area'], 'lines': [ - ['resp_length', 'received', 'incremental', 1, 1024], - ['bytes_sent', 'sent', 'incremental', -1, 1024] + ['resp_length', 'received', 'incremental', 8, 1000], + ['bytes_sent', 'sent', 'incremental', -8, 1000] ]}, 'response_time': { 'options': [None, 'Processing Time', 'milliseconds', 'timings', 'web_log.response_time', 'area'], @@ -40,6 +54,14 @@ CHARTS = { ['resp_time_max', 'max', 'incremental', 1, 1000], ['resp_time_avg', 'avg', 'incremental', 1, 1000] ]}, + 'response_time_upstream': { + 'options': [None, 'Processing Time Upstream', 'milliseconds', 'timings', + 'web_log.response_time_upstream', 'area'], + 'lines': [ + ['resp_time_upstream_min', 'min', 'incremental', 1, 1000], + ['resp_time_upstream_max', 'max', 'incremental', 1, 1000], + ['resp_time_upstream_avg', 'avg', 'incremental', 1, 1000] + ]}, 'clients': { 'options': [None, 'Current Poll Unique Client IPs', 'unique ips', 'clients', 'web_log.clients', 'stacked'], 'lines': [ @@ -77,36 +99,160 @@ CHARTS = { ['redirects', 'redirect', 'incremental', 1, 1], ['bad_requests', 'bad', 'incremental', 1, 1], ['other_requests', 'other', 'incremental', 1, 1] + ]}, + 'requests_per_url': { + 'options': [None, 'Requests Per Url', 'requests/s', 'urls', 'web_log.requests_per_url', + 'stacked'], + 'lines': [ + ['url_pattern_other', 'other', 'incremental', 1, 1] + ]}, + 'requests_per_user_defined': { + 'options': [None, 'Requests Per User Defined Pattern', 'requests/s', 'user defined', + 'web_log.requests_per_user_defined', 'stacked'], + 'lines': [ + ['user_pattern_other', 'other', 'incremental', 1, 1] ]} } -NAMED_URL_PATTERN = namedtuple('URL_PATTERN', ['description', 'pattern']) +CHARTS_APACHE_CACHE = { + 'apache_cache': { + 'options': [None, 'Apache Cached Responses', 'percent cached', 'cached', 'web_log.apache_cache_cache', + 'stacked'], + 'lines': [ + ["hit", 'cache', "percentage-of-absolute-row"], + ["miss", None, "percentage-of-absolute-row"], + ["other", None, "percentage-of-absolute-row"] + ]} +} + +CHARTS_SQUID = { + 'squid_duration': { + 'options': [None, 'Elapsed Time The Transaction Busied The Cache', + 'milliseconds', 'squid_timings', 'web_log.squid_duration', 'area'], + 'lines': [ + ['duration_min', 'min', 'incremental', 1, 1000], + ['duration_max', 'max', 'incremental', 1, 1000], + ['duration_avg', 'avg', 'incremental', 1, 1000] + ]}, + 'squid_bytes': { + 'options': [None, 'Amount Of Data Delivered To The Clients', + 'kilobits/s', 'squid_bandwidth', 'web_log.squid_bytes', 'area'], + 'lines': [ + ['bytes', 'sent', 'incremental', 8, 1000] + ]}, + 'squid_response_statuses': { + 'options': [None, 'Response Statuses', 'responses/s', 'squid_responses', 'web_log.squid_response_statuses', + 'stacked'], + 'lines': [ + ['successful_requests', 'success', 'incremental', 1, 1], + ['server_errors', 'error', 'incremental', 1, 1], + ['redirects', 'redirect', 'incremental', 1, 1], + ['bad_requests', 'bad', 'incremental', 1, 1], + ['other_requests', 'other', 'incremental', 1, 1] + ]}, + 'squid_response_codes': { + 'options': [None, 'Response Codes', 'responses/s', 'squid_responses', + 'web_log.squid_response_codes', 'stacked'], + 'lines': [ + ['2xx', None, 'incremental'], + ['5xx', None, 'incremental'], + ['3xx', None, 'incremental'], + ['4xx', None, 'incremental'], + ['1xx', None, 'incremental'], + ['0xx', None, 'incremental'], + ['other', None, 'incremental'], + ['unmatched', None, 'incremental'] + ]}, + 'squid_code': { + 'options': [None, 'Responses Per Cache Result Of The Request', + 'requests/s', 'squid_squid_cache', 'web_log.squid_code', 'stacked'], + 'lines': [ + ]}, + 'squid_detailed_response_codes': { + 'options': [None, 'Detailed Response Codes', + 'responses/s', 'squid_responses', 'web_log.squid_detailed_response_codes', 'stacked'], + 'lines': [ + ]}, + 'squid_hier_code': { + 'options': [None, 'Responses Per Hierarchy Code', + 'requests/s', 'squid_hierarchy', 'web_log.squid_hier_code', 'stacked'], + 'lines': [ + ]}, + 'squid_method': { + 'options': [None, 'Requests Per Method', + 'requests/s', 'squid_requests', 'web_log.squid_method', 'stacked'], + 'lines': [ + ]}, + 'squid_mime_type': { + 'options': [None, 'Requests Per MIME Type', + 'requests/s', 'squid_requests', 'web_log.squid_mime_type', 'stacked'], + 'lines': [ + ]}, + 'squid_clients': { + 'options': [None, 'Current Poll Unique Client IPs', 'unique ips', 'squid_clients', + 'web_log.squid_clients', 'stacked'], + 'lines': [ + ['unique_ipv4', 'ipv4', 'incremental'], + ['unique_ipv6', 'ipv6', 'incremental'] + ]}, + 'squid_clients_all': { + 'options': [None, 'All Time Unique Client IPs', 'unique ips', 'squid_clients', + 'web_log.squid_clients_all', 'stacked'], + 'lines': [ + ['unique_tot_ipv4', 'ipv4', 'absolute'], + ['unique_tot_ipv6', 'ipv6', 'absolute'] + ]}, + 'squid_transport_methods': { + 'options': [None, 'Transport Methods', 'requests/s', 'squid_squid_transport', + 'web_log.squid_transport_methods', 'stacked'], + 'lines': [ + ]}, + 'squid_transport_errors': { + 'options': [None, 'Transport Errors', 'requests/s', 'squid_squid_transport', + 'web_log.squid_transport_errors', 'stacked'], + 'lines': [ + ]}, + 'squid_handling_opts': { + 'options': [None, 'Handling Opts', 'requests/s', 'squid_squid_cache', + 'web_log.squid_handling_opts', 'stacked'], + 'lines': [ + ]}, + 'squid_object_types': { + 'options': [None, 'Object Types', 'objects/s', 'squid_squid_cache', + 'web_log.squid_object_types', 'stacked'], + 'lines': [ + ]}, + 'squid_cache_events': { + 'options': [None, 'Cache Events', 'events/s', 'squid_squid_cache', + 'web_log.squid_cache_events', 'stacked'], + 'lines': [ + ]} +} + +NAMED_PATTERN = namedtuple('PATTERN', ['description', 'func']) DET_RESP_AGGR = ['', '_1xx', '_2xx', '_3xx', '_4xx', '_5xx', '_Other'] +SQUID_CODES = dict(TCP='squid_transport_methods', UDP='squid_transport_methods', NONE='squid_transport_methods', + CLIENT='squid_handling_opts', IMS='squid_handling_opts', ASYNC='squid_handling_opts', + SWAPFAIL='squid_handling_opts', REFRESH='squid_handling_opts', SHARED='squid_handling_opts', + REPLY='squid_handling_opts', NEGATIVE='squid_object_types', STALE='squid_object_types', + OFFLINE='squid_object_types', INVALID='squid_object_types', FAIL='squid_object_types', + MODIFIED='squid_object_types', UNMODIFIED='squid_object_types', REDIRECT='squid_object_types', + HIT='squid_cache_events', MEM='squid_cache_events', MISS='squid_cache_events', + DENIED='squid_cache_events', NOFETCH='squid_cache_events', TUNNEL='squid_cache_events', + ABORTED='squid_transport_errors', TIMEOUT='squid_transport_errors') + class Service(LogService): def __init__(self, configuration=None, name=None): """ :param configuration: :param name: - # self._get_data = None # will be assigned in 'check' method. - # self.order = None # will be assigned in 'create_*_method' method. - # self.definitions = None # will be assigned in 'create_*_method' method. """ LogService.__init__(self, configuration=configuration, name=name) - # Variables from module configuration file - self.log_type = self.configuration.get('type', 'web_access') + self.log_type = self.configuration.get('type', 'web') self.log_path = self.configuration.get('path') - self.url_pattern = self.configuration.get('categories') # dict - self.custom_log_format = self.configuration.get('custom_log_format') # dict - # Instance variables - self.regex = None # will be assigned in 'find_regex' or 'find_regex_custom' method - self.data = {'bytes_sent': 0, 'resp_length': 0, 'resp_time_min': 0, 'resp_time_max': 0, - 'resp_time_avg': 0, 'unique_cur_ipv4': 0, 'unique_cur_ipv6': 0, '2xx': 0, - '5xx': 0, '3xx': 0, '4xx': 0, '1xx': 0, '0xx': 0, 'unmatched': 0, 'req_ipv4': 0, - 'req_ipv6': 0, 'unique_tot_ipv4': 0, 'unique_tot_ipv6': 0, 'successful_requests': 0, - 'redirects': 0, 'bad_requests': 0, 'server_errors': 0, 'other_requests': 0, 'GET': 0} def check(self): """ @@ -117,11 +263,18 @@ class Service(LogService): 3. "log_path' must not be empty. We need at least 1 line to find appropriate pattern to parse 4. other checks depends on log "type" """ + + log_types = dict(web=Web, apache_cache=ApacheCache, squid=Squid) + + if self.log_type not in log_types: + self.error('bad log type (%s). Supported types: %s' % (self.log_type, log_types.keys())) + return False + if not self.log_path: self.error('log path is not specified') return False - if not access(self.log_path, R_OK): + if not (self._find_recent_log_file() and access(self.log_path, R_OK)): self.error('%s not readable or not exist' % self.log_path) return False @@ -129,357 +282,237 @@ class Service(LogService): self.error('%s is empty' % self.log_path) return False + self.configuration['update_every'] = self.update_every + self.configuration['name'] = self.name + self.configuration['override_name'] = self.override_name + self.configuration['_dimensions'] = self._dimensions + self.configuration['path'] = self.log_path + + cls = log_types[self.log_type] + self.Job = cls(configuration=self.configuration) + if self.Job.check(): + self.order = self.Job.order + self.definitions = self.Job.definitions + self.info('Current log file: %s' % self.log_path) + return True + return False + + def _get_data(self): + return self.Job.get_data(self._get_raw_data()) + + +class Mixin: + def filter_data(self, raw_data): + """ + :param raw_data: list + :return: + """ + if not self.pre_filter: + return raw_data + filtered = raw_data + for elem in self.pre_filter: + if elem.description == 'filter_include': + filtered = filter(elem.func, filtered) + elif elem.description == 'filter_exclude': + filtered = filterfalse(elem.func, filtered) + return filtered + + def add_new_dimension(self, dimension_id, chart_key, dimension=None, + algorithm='incremental', multiplier=1, divisor=1): + """ + :param dimension: + :param chart_key: + :param dimension_id: + :param algorithm: + :param multiplier: + :param divisor: + :return: + """ + + self.data[dimension_id] = 0 + # SET method check if dim in _dimensions + self.conf['_dimensions'].append(dimension_id) + # UPDATE method do SET only if dim in definitions + dimension_list = list(map(str, [dimension_id, + dimension if dimension else dimension_id, + algorithm, + multiplier, + divisor])) + self.definitions[chart_key]['lines'].append(dimension_list) + job_name = find_job_name(self.conf['override_name'], self.conf['name']) + opts = self.definitions[chart_key]['options'] + chart = 'CHART %s.%s "" "%s" %s "%s" %s %s 60000 %s\n' % (job_name, chart_key, + opts[1], opts[2], opts[3], + opts[4], opts[5], self.conf['update_every']) + print(chart + "DIMENSION %s\n" % ' '.join(dimension_list)) + + def get_last_line(self): + """ + Reads last line from the log file + :return: str: + """ # Read last line (or first if there is only one line) - with open(self.log_path, 'rb') as logs: + with open(self.conf['path'], 'rb') as logs: logs.seek(-2, 2) while logs.read(1) != b'\n': logs.seek(-2, 1) if logs.tell() == 0: break last_line = logs.readline() - try: - last_line = last_line.decode() + return last_line.decode() except UnicodeDecodeError: try: - last_line = last_line.decode(encoding='utf-8') + return last_line.decode(encoding='utf-8') except (TypeError, UnicodeDecodeError) as error: - self.error(str(error)) + msg.error('web_log', str(error)) return False - if self.log_type == 'web_access': - self.unique_all_time = list() # sorted list of unique IPs - self.detailed_response_codes = self.configuration.get('detailed_response_codes', True) - self.detailed_response_aggregate = self.configuration.get('detailed_response_aggregate', True) - self.all_time = self.configuration.get('all_time', True) + @staticmethod + def error(*params): + msg.error('web_log', ' '.join(map(str, params))) - # Custom_log_format or predefined log format. - if self.custom_log_format: - match_dict, error = self.find_regex_custom(last_line) - else: - match_dict, error = self.find_regex(last_line) + @staticmethod + def info(*params): + msg.info('web_log', ' '.join(map(str, params))) - # "match_dict" is None if there are any problems - if match_dict is None: - self.error(str(error)) - return False - # self.url_pattern check - if self.url_pattern: - self.url_pattern = check_req_per_url_pattern('rpu', self.url_pattern) +class Web(Mixin): + def __init__(self, configuration): + self.conf = configuration + self.pre_filter = check_patterns('filter', self.conf.get('filter')) + self.storage = dict() + self.data = {'bytes_sent': 0, 'resp_length': 0, 'resp_time_min': 0, 'resp_time_max': 0, + 'resp_time_avg': 0, 'resp_time_upstream_min': 0, 'resp_time_upstream_max': 0, + 'resp_time_upstream_avg': 0, 'unique_cur_ipv4': 0, 'unique_cur_ipv6': 0, '2xx': 0, + '5xx': 0, '3xx': 0, '4xx': 0, '1xx': 0, '0xx': 0, 'unmatched': 0, 'req_ipv4': 0, + 'req_ipv6': 0, 'unique_tot_ipv4': 0, 'unique_tot_ipv6': 0, 'successful_requests': 0, + 'redirects': 0, 'bad_requests': 0, 'server_errors': 0, 'other_requests': 0, 'GET': 0} - self.create_access_charts(match_dict) # Create charts - self._get_data = self._get_access_data # _get_data assignment - else: - self.error('Not implemented') + def check(self): + last_line = self.get_last_line() + if not last_line: return False - - # Double check - if not self.regex: - self.error('That can not happen, but it happened. "regex" is None') - - self.info('Collected data: %s' % list(match_dict.keys())) - return True - - def find_regex_custom(self, last_line): - """ - :param last_line: str: literally last line from log file - :return: tuple where: - [0]: dict or None: match_dict or None - [1]: str: error description - - We are here only if "custom_log_format" is in logs. We need to make sure: - 1. "custom_log_format" is a dict - 2. "pattern" in "custom_log_format" and pattern is <str> instance - 3. if "time_multiplier" is in "custom_log_format" it must be <int> instance - - If all parameters is ok we need to make sure: - 1. Pattern search is success - 2. Pattern search contains named subgroups (?P<subgroup_name>) (= "match_dict") - - If pattern search is success we need to make sure: - 1. All mandatory keys ['address', 'code', 'bytes_sent', 'method', 'url'] are in "match_dict" - - If this is True we need to make sure: - 1. All mandatory key values from "match_dict" have the correct format - ("code" is integer, "method" is uppercase word, etc) - - If non mandatory keys in "match_dict" we need to make sure: - 1. All non mandatory key values from match_dict ['resp_length', 'resp_time'] have the correct format - ("resp_length" is integer or "-", "resp_time" is integer or float) - - """ - if not hasattr(self.custom_log_format, 'keys'): - return find_regex_return(msg='Custom log: "custom_log_format" is not a <dict>') - - pattern = self.custom_log_format.get('pattern') - if not (pattern and isinstance(pattern, str)): - return find_regex_return(msg='Custom log: "pattern" option is not specified or type is not <str>') - - resp_time_func = self.custom_log_format.get('time_multiplier') or 0 - - if not isinstance(resp_time_func, int): - return find_regex_return(msg='Custom log: "time_multiplier" is not an integer') - - try: - regex = re.compile(pattern) - except re.error as error: - return find_regex_return(msg='Pattern compile error: %s' % str(error)) - - match = regex.search(last_line) - if match: - match_dict = match.groupdict() or None + # Custom_log_format or predefined log format. + if self.conf.get('custom_log_format'): + match_dict, error = self.find_regex_custom(last_line) else: - return find_regex_return(msg='Custom log: pattern search FAILED') + match_dict, error = self.find_regex(last_line) + # "match_dict" is None if there are any problems if match_dict is None: - find_regex_return(msg='Custom log: search OK but contains no named subgroups' - ' (you need to use ?P<subgroup_name>)') - else: - mandatory_dict = {'address': r'[\da-f.:]+', - 'code': r'[1-9]\d{2}', - 'method': r'[A-Z]+', - 'bytes_sent': r'\d+|-'} - optional_dict = {'resp_length': r'\d+', - 'resp_time': r'[\d.]+', - 'http_version': r'\d\.\d'} - - mandatory_values = set(mandatory_dict) - set(match_dict) - if mandatory_values: - return find_regex_return(msg='Custom log: search OK but some mandatory keys (%s) are missing' - % list(mandatory_values)) - else: - for key in mandatory_dict: - if not re.search(mandatory_dict[key], match_dict[key]): - return find_regex_return(msg='Custom log: can\'t parse "%s": %s' - % (key, match_dict[key])) - - optional_values = set(optional_dict) & set(match_dict) - for key in optional_values: - if not re.search(optional_dict[key], match_dict[key]): - return find_regex_return(msg='Custom log: can\'t parse "%s": %s' - % (key, match_dict[key])) - - dot_in_time = '.' in match_dict.get('resp_time', '') - if dot_in_time: - self.resp_time_func = lambda time: time * (resp_time_func or 1000000) - else: - self.resp_time_func = lambda time: time * (resp_time_func or 1) - - self.regex = regex - return find_regex_return(match_dict=match_dict) - - def find_regex(self, last_line): - """ - :param last_line: str: literally last line from log file - :return: tuple where: - [0]: dict or None: match_dict or None - [1]: str: error description - We need to find appropriate pattern for current log file - All logic is do a regex search through the string for all predefined patterns - until we find something or fail. - """ - # REGEX: 1.IPv4 address 2.HTTP method 3. URL 4. Response code - # 5. Bytes sent 6. Response length 7. Response process time - acs_default = re.compile(r'(?P<address>[\da-f.:]+)' - r' -.*?"(?P<method>[A-Z]+)' - r' (?P<url>[^ ]+)' - r' [A-Z]+/(?P<http_version>\d\.\d)"' - r' (?P<code>[1-9]\d{2})' - r' (?P<bytes_sent>\d+|-)') - - acs_apache_ext_insert = re.compile(r'(?P<address>[\da-f.:]+)' - r' -.*?"(?P<method>[A-Z]+)' - r' (?P<url>[^ ]+)' - r' [A-Z]+/(?P<http_version>\d\.\d)"' - r' (?P<code>[1-9]\d{2})' - r' (?P<bytes_sent>\d+|-)' - r' (?P<resp_length>\d+)' - r' (?P<resp_time>\d+) ') - - acs_apache_ext_append = re.compile(r'(?P<address>[\da-f.:]+)' - r' -.*?"(?P<method>[A-Z]+)' - r' (?P<url>[^ ]+)' - r' [A-Z]+/(?P<http_version>\d\.\d)"' - r' (?P<code>[1-9]\d{2})' - r' (?P<bytes_sent>\d+|-)' - r' .*?' - r' (?P<resp_length>\d+)' - r' (?P<resp_time>\d+)' - r'(?: |$)') - - acs_nginx_ext_insert = re.compile(r'(?P<address>[\da-f.:]+)' - r' -.*?"(?P<method>[A-Z]+)' - r' (?P<url>[^ ]+)' - r' [A-Z]+/(?P<http_version>\d\.\d)"' - r' (?P<code>[1-9]\d{2})' - r' (?P<bytes_sent>\d+)' - r' (?P<resp_length>\d+)' - r' (?P<resp_time>\d+\.\d+) ') - - acs_nginx_ext_append = re.compile(r'(?P<address>[\da-f.:]+)' - r' -.*?"(?P<method>[A-Z]+)' - r' (?P<url>[^ ]+)' - r' [A-Z]+/(?P<http_version>\d\.\d)"' - r' (?P<code>[1-9]\d{2})' - r' (?P<bytes_sent>\d+)' - r' .*?' - r' (?P<resp_length>\d+)' - r' (?P<resp_time>\d+\.\d+)') - - def func_usec(time): - return time - - def func_sec(time): - return time * 1000000 - - r_regex = [acs_apache_ext_insert, acs_apache_ext_append, acs_nginx_ext_insert, - acs_nginx_ext_append, acs_default] - r_function = [func_usec, func_usec, func_sec, func_sec, func_usec] - regex_function = zip(r_regex, r_function) - - match_dict = dict() - for regex, function in regex_function: - match = regex.search(last_line) - if match: - self.regex = regex - self.resp_time_func = function - match_dict = match.groupdict() - break + self.error(str(error)) + return False + self.storage['unique_all_time'] = list() + self.storage['url_pattern'] = check_patterns('url_pattern', self.conf.get('categories')) + self.storage['user_pattern'] = check_patterns('user_pattern', self.conf.get('user_defined')) - return find_regex_return(match_dict=match_dict or None, - msg='Unknown log format. You need to use "custom_log_format" feature.') + self.create_web_charts(match_dict) # Create charts + self.info('Collected data: %s' % list(match_dict.keys())) + return True - def create_access_charts(self, match_dict): + def create_web_charts(self, match_dict): """ :param match_dict: dict: regex.search.groupdict(). Ex. {'address': '127.0.0.1', 'code': '200', 'method': 'GET'} :return: - Create additional charts depending on the 'match_dict' keys and configuration file options - 1. 'time_response' chart is removed if there is no 'resp_time' in match_dict. - 2. Other stuff is just remove/add chart depending on yes/no in conf + Create/remove additional charts depending on the 'match_dict' keys and configuration file options """ + self.order = ORDER_WEB[:] + self.definitions = deepcopy(CHARTS_WEB) - def find_job_name(override_name, name): - """ - :param override_name: str: 'name' var from configuration file - :param name: str: 'job_name' from configuration file - :return: str: new job name - We need this for dynamic charts. Actually same logic as in python.d.plugin. - """ - add_to_name = override_name or name - if add_to_name: - return '_'.join(['web_log', re.sub('\s+', '_', add_to_name)]) - else: - return 'web_log' - - self.order = ORDER[:] - self.definitions = deepcopy(CHARTS) - - job_name = find_job_name(self.override_name, self.name) - - self.http_method_chart = 'CHART %s.http_method' \ - ' "" "Requests Per HTTP Method" requests/s "http methods"' \ - ' web_log.http_method stacked 11 %s\n' \ - 'DIMENSION GET GET incremental\n' % (job_name, self.update_every) - self.http_version_chart = 'CHART %s.http_version' \ - ' "" "Requests Per HTTP Version" requests/s "http versions"' \ - ' web_log.http_version stacked 12 %s\n' % (job_name, self.update_every) - - # Remove 'request_time' chart from ORDER if resp_time not in match_dict if 'resp_time' not in match_dict: self.order.remove('response_time') - # Remove 'clients_all' chart from ORDER if specified in the configuration - if not self.all_time: + if 'resp_time_upstream' not in match_dict: + self.order.remove('response_time_upstream') + + if not self.conf.get('all_time', True): self.order.remove('clients_all') + # Add 'detailed_response_codes' chart if specified in the configuration - if self.detailed_response_codes: - self.detailed_chart = list() - for prio, add_to_dim in enumerate(DET_RESP_AGGR): - self.detailed_chart.append('CHART %s.detailed_response_codes%s ""' - ' "Detailed Response Codes %s" requests/s responses' - ' web_log.detailed_response_codes%s stacked %s %s\n' - % (job_name, add_to_dim, add_to_dim[1:], add_to_dim, - str(prio), self.update_every)) - - codes = DET_RESP_AGGR[:1] if self.detailed_response_aggregate else DET_RESP_AGGR[1:] + if self.conf.get('detailed_response_codes', True): + codes = DET_RESP_AGGR[:1] if self.conf.get('detailed_response_aggregate', True) else DET_RESP_AGGR[1:] for code in codes: self.order.append('detailed_response_codes%s' % code) - self.definitions['detailed_response_codes%s' % code] = {'options': - [None, - 'Detailed Response Codes %s' % code[1:], - 'requests/s', - 'responses', - 'web_log.detailed_response_codes%s' % code, - 'stacked'], - 'lines': []} + self.definitions['detailed_response_codes%s' % code] \ + = {'options': [None, 'Detailed Response Codes %s' % code[1:], 'requests/s', 'responses', + 'web_log.detailed_response_codes%s' % code, 'stacked'], + 'lines': []} # Add 'requests_per_url' chart if specified in the configuration - if self.url_pattern: - self.definitions['requests_per_url'] = {'options': [None, 'Requests Per Url', 'requests/s', - 'urls', 'web_log.requests_per_url', 'stacked'], - 'lines': [['rpu_other', 'other', 'incremental']]} - for elem in self.url_pattern: - self.definitions['requests_per_url']['lines'].append([elem.description, elem.description[4:], + if self.storage['url_pattern']: + for elem in self.storage['url_pattern']: + self.definitions['requests_per_url']['lines'].append([elem.description, + elem.description[12:], 'incremental']) - self.data.update({elem.description: 0}) - self.data.update({'rpu_other': 0}) + self.data[elem.description] = 0 + self.data['url_pattern_other'] = 0 else: self.order.remove('requests_per_url') - def add_new_dimension(self, dimension, line_list, chart_string, key): - """ - :param dimension: str: response status code. Ex.: '202', '499' - :param line_list: list: Ex.: ['202', '202', 'incremental'] - :param chart_string: Current string we need to pass to netdata to rebuild the chart - :param key: str: CHARTS dict key (chart name). Ex.: 'response_time' - :return: str: new chart string = previous + new dimensions - """ - self.data.update({dimension: 0}) - # SET method check if dim in _dimensions - self._dimensions.append(dimension) - # UPDATE method do SET only if dim in definitions - self.definitions[key]['lines'].append(line_list) - chart = chart_string - chart += "%s %s\n" % ('DIMENSION', ' '.join(line_list)) - print(chart) - return chart + # Add 'requests_per_user_defined' chart if specified in the configuration + if self.storage['user_pattern'] and 'user_defined' in match_dict: + for elem in self.storage['user_pattern']: + self.definitions['requests_per_user_defined']['lines'].append([elem.description, + elem.description[13:], + 'incremental']) + self.data[elem.description] = 0 + self.data['user_pattern_other'] = 0 + else: + self.order.remove('requests_per_user_defined') - def _get_access_data(self): + def get_data(self, raw_data=None): """ - Parse new log lines + Parses new log lines :return: dict OR None None if _get_raw_data method fails. In all other cases - dict. """ - raw = self._get_raw_data() - if raw is None: - return None + if not raw_data: + return None if raw_data is None else self.data + + filtered_data = self.filter_data(raw_data=raw_data) + + unique_current = set() + timings = defaultdict(lambda: dict(minimum=None, maximum=0, summary=0, count=0)) - request_time, unique_current = list(), list() - request_counter = {'count': 0, 'sum': 0} ip_address_counter = {'unique_cur_ip': 0} - for line in raw: - match = self.regex.search(line) + for line in filtered_data: + match = self.storage['regex'].search(line) if match: match_dict = match.groupdict() try: - code = ''.join([match_dict['code'][0], 'xx']) + code = match_dict['code'][0] + 'xx' self.data[code] += 1 except KeyError: self.data['0xx'] += 1 # detailed response code - if self.detailed_response_codes: - self._get_data_detailed_response_codes(match_dict['code']) + if self.conf.get('detailed_response_codes', True): + self.get_data_per_response_codes_detailed(code=match_dict['code']) # response statuses - self._get_data_statuses(match_dict['code']) + self.get_data_per_statuses(code=match_dict['code']) # requests per url - if self.url_pattern: - self._get_data_per_url(match_dict['url']) + if self.storage['url_pattern']: + self.get_data_per_pattern(row=match_dict['url'], + other='url_pattern_other', + pattern=self.storage['url_pattern']) + # requests per user defined pattern + if self.storage['user_pattern'] and 'user_defined' in match_dict: + self.get_data_per_pattern(row=match_dict['user_defined'], + other='user_pattern_other', + pattern=self.storage['user_pattern']) # requests per http method - self._get_data_http_method(match_dict['method']) + if match_dict['method'] not in self.data: + self.add_new_dimension(dimension_id=match_dict['method'], + chart_key='http_method') + self.data[match_dict['method']] += 1 # requests per http version if 'http_version' in match_dict: - self._get_data_http_version(match_dict['http_version']) + dim_id = match_dict['http_version'].replace('.', '_') + if dim_id not in self.data: + self.add_new_dimension(dimension_id=dim_id, + chart_key='http_version', + dimension=match_dict['http_version']) + self.data[dim_id] += 1 # bandwidth sent bytes_sent = match_dict['bytes_sent'] if '-' not in match_dict['bytes_sent'] else 0 self.data['bytes_sent'] += int(bytes_sent) @@ -487,92 +520,245 @@ class Service(LogService): if 'resp_length' in match_dict: self.data['resp_length'] += int(match_dict['resp_length']) if 'resp_time' in match_dict: - resp_time = self.resp_time_func(float(match_dict['resp_time'])) - bisect.insort_left(request_time, resp_time) - request_counter['count'] += 1 - request_counter['sum'] += resp_time + get_timings(timings=timings['resp_time'], + time=self.storage['func_resp_time'](float(match_dict['resp_time']))) + if 'resp_time_upstream' in match_dict and match_dict['resp_time_upstream'] != '-': + get_timings(timings=timings['resp_time_upstream'], + time=self.storage['func_resp_time'](float(match_dict['resp_time_upstream']))) # requests per ip proto proto = 'ipv4' if '.' in match_dict['address'] else 'ipv6' self.data['req_' + proto] += 1 # unique clients ips - if address_not_in_pool(self.unique_all_time, match_dict['address'], - self.data['unique_tot_ipv4'] + self.data['unique_tot_ipv6']): + if address_not_in_pool(pool=self.storage['unique_all_time'], + address=match_dict['address'], + pool_size=self.data['unique_tot_ipv4'] + self.data['unique_tot_ipv6']): self.data['unique_tot_' + proto] += 1 - if address_not_in_pool(unique_current, match_dict['address'], ip_address_counter['unique_cur_ip']): + if match_dict['address'] not in unique_current: self.data['unique_cur_' + proto] += 1 - ip_address_counter['unique_cur_ip'] += 1 + unique_current.add(match_dict['address']) else: self.data['unmatched'] += 1 # timings - if request_time: - self.data['resp_time_min'] += int(request_time[0]) - self.data['resp_time_avg'] += int(round(float(request_counter['sum']) / request_counter['count'])) - self.data['resp_time_max'] += int(request_time[-1]) + for elem in timings: + self.data[elem + '_min'] += timings[elem]['minimum'] + self.data[elem + '_avg'] += timings[elem]['summary'] / timings[elem]['count'] + self.data[elem + '_max'] += timings[elem]['maximum'] return self.data - def _get_data_detailed_response_codes(self, code): + def find_regex(self, last_line): """ - :param code: str: CODE from parsed line. Ex.: '202, '499' - :return: - Calls add_new_dimension method If the value is found for the first time + :param last_line: str: literally last line from log file + :return: tuple where: + [0]: dict or None: match_dict or None + [1]: str: error description + We need to find appropriate pattern for current log file + All logic is do a regex search through the string for all predefined patterns + until we find something or fail. """ - if code not in self.data: - if self.detailed_response_aggregate: - chart_string_copy = self.detailed_chart[0] - self.detailed_chart[0] = self.add_new_dimension(code, [code, code, 'incremental'], - chart_string_copy, 'detailed_response_codes') - else: - code_index = int(code[0]) if int(code[0]) < 6 else 6 - chart_string_copy = self.detailed_chart[code_index] - chart_name = 'detailed_response_codes' + DET_RESP_AGGR[code_index] - self.detailed_chart[code_index] = self.add_new_dimension(code, [code, code, 'incremental'], - chart_string_copy, chart_name) - self.data[code] += 1 + # REGEX: 1.IPv4 address 2.HTTP method 3. URL 4. Response code + # 5. Bytes sent 6. Response length 7. Response process time + default = re.compile(r'(?P<address>[\da-f.:]+)' + r' -.*?"(?P<method>[A-Z]+)' + r' (?P<url>[^ ]+)' + r' [A-Z]+/(?P<http_version>\d\.\d)"' + r' (?P<code>[1-9]\d{2})' + r' (?P<bytes_sent>\d+|-)') + + apache_ext_insert = re.compile(r'(?P<address>[\da-f.:]+)' + r' -.*?"(?P<method>[A-Z]+)' + r' (?P<url>[^ ]+)' + r' [A-Z]+/(?P<http_version>\d\.\d)"' + r' (?P<code>[1-9]\d{2})' + r' (?P<bytes_sent>\d+|-)' + r' (?P<resp_length>\d+)' + r' (?P<resp_time>\d+) ') + + apache_ext_append = re.compile(r'(?P<address>[\da-f.:]+)' + r' -.*?"(?P<method>[A-Z]+)' + r' (?P<url>[^ ]+)' + r' [A-Z]+/(?P<http_version>\d\.\d)"' + r' (?P<code>[1-9]\d{2})' + r' (?P<bytes_sent>\d+|-)' + r' .*?' + r' (?P<resp_length>\d+)' + r' (?P<resp_time>\d+)' + r'(?: |$)') + + nginx_ext_insert = re.compile(r'(?P<address>[\da-f.:]+)' + r' -.*?"(?P<method>[A-Z]+)' + r' (?P<url>[^ ]+)' + r' [A-Z]+/(?P<http_version>\d\.\d)"' + r' (?P<code>[1-9]\d{2})' + r' (?P<bytes_sent>\d+)' + r' (?P<resp_length>\d+)' + r' (?P<resp_time>\d+\.\d+) ') + + nginx_ext2_insert = re.compile(r'(?P<address>[\da-f.:]+)' + r' -.*?"(?P<method>[A-Z]+)' + r' (?P<url>[^ ]+)' + r' [A-Z]+/(?P<http_version>\d\.\d)"' + r' (?P<code>[1-9]\d{2})' + r' (?P<bytes_sent>\d+)' + r' (?P<resp_length>\d+)' + r' (?P<resp_time>\d+\.\d+)' + r' (?P<resp_time_upstream>[\d.-]+) ') + + nginx_ext_append = re.compile(r'(?P<address>[\da-f.:]+)' + r' -.*?"(?P<method>[A-Z]+)' + r' (?P<url>[^ ]+)' + r' [A-Z]+/(?P<http_version>\d\.\d)"' + r' (?P<code>[1-9]\d{2})' + r' (?P<bytes_sent>\d+)' + r' .*?' + r' (?P<resp_length>\d+)' + r' (?P<resp_time>\d+\.\d+)') + + def func_usec(time): + return time + + def func_sec(time): + return time * 1000000 + + r_regex = [apache_ext_insert, apache_ext_append, + nginx_ext2_insert, nginx_ext_insert, nginx_ext_append, + default] + r_function = [func_usec, func_usec, func_sec, func_sec, func_sec, func_usec] + regex_function = zip(r_regex, r_function) + + match_dict = dict() + for regex, func in regex_function: + match = regex.search(last_line) + if match: + self.storage['regex'] = regex + self.storage['func_resp_time'] = func + match_dict = match.groupdict() + break + + return find_regex_return(match_dict=match_dict or None, + msg='Unknown log format. You need to use "custom_log_format" feature.') - def _get_data_http_method(self, method): + def find_regex_custom(self, last_line): """ - :param method: str: METHOD from parsed line. Ex.: 'GET', 'POST' - :return: - Calls add_new_dimension method If the value is found for the first time + :param last_line: str: literally last line from log file + :return: tuple where: + [0]: dict or None: match_dict or None + [1]: str: error description + + We are here only if "custom_log_format" is in logs. We need to make sure: + 1. "custom_log_format" is a dict + 2. "pattern" in "custom_log_format" and pattern is <str> instance + 3. if "time_multiplier" is in "custom_log_format" it must be <int> instance + + If all parameters is ok we need to make sure: + 1. Pattern search is success + 2. Pattern search contains named subgroups (?P<subgroup_name>) (= "match_dict") + + If pattern search is success we need to make sure: + 1. All mandatory keys ['address', 'code', 'bytes_sent', 'method', 'url'] are in "match_dict" + + If this is True we need to make sure: + 1. All mandatory key values from "match_dict" have the correct format + ("code" is integer, "method" is uppercase word, etc) + + If non mandatory keys in "match_dict" we need to make sure: + 1. All non mandatory key values from match_dict ['resp_length', 'resp_time'] have the correct format + ("resp_length" is integer or "-", "resp_time" is integer or float) + """ - if method not in self.data: - chart_string_copy = self.http_method_chart - self.http_method_chart = self.add_new_dimension(method, [method, method, 'incremental'], - chart_string_copy, 'http_method') - self.data[method] += 1 + if not hasattr(self.conf.get('custom_log_format'), 'keys'): + return find_regex_return(msg='Custom log: "custom_log_format" is not a <dict>') + + pattern = self.conf.get('custom_log_format', dict()).get('pattern') + if not (pattern and isinstance(pattern, str)): + return find_regex_return(msg='Custom log: "pattern" option is not specified or type is not <str>') + + resp_time_func = self.conf.get('custom_log_format', dict()).get('time_multiplier') or 0 + + if not isinstance(resp_time_func, int): + return find_regex_return(msg='Custom log: "time_multiplier" is not an integer') + + try: + regex = re.compile(pattern) + except re.error as error: + return find_regex_return(msg='Pattern compile error: %s' % str(error)) + match = regex.search(last_line) + if not match: + return find_regex_return(msg='Custom log: pattern search FAILED') - def _get_data_http_version(self, http_version): + match_dict = match.groupdict() or None + if match_dict is None: + return find_regex_return(msg='Custom log: search OK but contains no named subgroups' + ' (you need to use ?P<subgroup_name>)') + mandatory_dict = {'address': r'[\da-f.:]+', + 'code': r'[1-9]\d{2}', + 'method': r'[A-Z]+', + 'bytes_sent': r'\d+|-'} + optional_dict = {'resp_length': r'\d+', + 'resp_time': r'[\d.]+', + 'resp_time_upstream': r'[\d.-]+', + 'http_version': r'\d\.\d'} + + mandatory_values = set(mandatory_dict) - set(match_dict) + if mandatory_values: + return find_regex_return(msg='Custom log: search OK but some mandatory keys (%s) are missing' + % list(mandatory_values)) + for key in mandatory_dict: + if not re.search(mandatory_dict[key], match_dict[key]): + return find_regex_return(msg='Custom log: can\'t parse "%s": %s' + % (key, match_dict[key])) + + optional_values = set(optional_dict) & set(match_dict) + for key in optional_values: + if not re.search(optional_dict[key], match_dict[key]): + return find_regex_return(msg='Custom log: can\'t parse "%s": %s' + % (key, match_dict[key])) + + dot_in_time = '.' in match_dict.get('resp_time', '') + if dot_in_time: + self.storage['func_resp_time'] = lambda time: time * (resp_time_func or 1000000) + else: + self.storage['func_resp_time'] = lambda time: time * (resp_time_func or 1) + + self.storage['regex'] = regex + return find_regex_return(match_dict=match_dict) + + def get_data_per_response_codes_detailed(self, code): """ - :param http_version: str: METHOD from parsed line. Ex.: '1.1', '1.0' + :param code: str: CODE from parsed line. Ex.: '202, '499' :return: Calls add_new_dimension method If the value is found for the first time """ - http_version_dim_id = http_version.replace('.', '_') - if http_version_dim_id not in self.data: - chart_string_copy = self.http_version_chart - self.http_version_chart = self.add_new_dimension(http_version_dim_id, - [http_version_dim_id, http_version, 'incremental'], - chart_string_copy, 'http_version') - self.data[http_version_dim_id] += 1 - - def _get_data_per_url(self, url): + if code not in self.data: + if self.conf.get('detailed_response_aggregate', True): + self.add_new_dimension(dimension_id=code, + chart_key='detailed_response_codes') + else: + code_index = int(code[0]) if int(code[0]) < 6 else 6 + chart_key = 'detailed_response_codes' + DET_RESP_AGGR[code_index] + self.add_new_dimension(dimension_id=code, + chart_key=chart_key) + self.data[code] += 1 + + def get_data_per_pattern(self, row, other, pattern): """ - :param url: str: URL from parsed line + :param row: str: + :param other: str: + :param pattern: named tuple: (['pattern_description', 'regular expression']) :return: Scan through string looking for the first location where patterns produce a match for all user defined patterns """ match = None - for elem in self.url_pattern: - if elem.pattern.search(url): + for elem in pattern: + if elem.func(row): self.data[elem.description] += 1 match = True break if not match: - self.data['rpu_other'] += 1 + self.data[other] += 1 - def _get_data_statuses(self, code): + def get_data_per_statuses(self, code): """ :param code: str: response status code. Ex.: '202', '499' :return: @@ -590,23 +776,209 @@ class Service(LogService): self.data['other_requests'] += 1 +class ApacheCache: + def __init__(self, configuration): + self.conf = configuration + self.order = ORDER_APACHE_CACHE + self.definitions = CHARTS_APACHE_CACHE + + @staticmethod + def check(): + return True + + @staticmethod + def get_data(raw_data=None): + data = dict(hit=0, miss=0, other=0) + if not raw_data: + return None if raw_data is None else data + + for line in raw_data: + if 'cache hit' in line: + data['hit'] += 1 + elif 'cache miss' in line: + data['miss'] += 1 + else: + data['other'] += 1 + return data + + +class Squid(Mixin): + def __init__(self, configuration): + self.conf = configuration + self.order = ORDER_SQUID + self.definitions = CHARTS_SQUID + self.pre_filter = check_patterns('filter', self.conf.get('filter')) + self.storage = dict() + self.data = {'duration_max': 0, 'duration_avg': 0, 'duration_min': 0, 'bytes': 0, + '0xx': 0, '1xx': 0, '2xx': 0, '3xx': 0, '4xx': 0, '5xx': 0, + 'other': 0, 'unmatched': 0, 'unique_ipv4': 0, 'unique_ipv6': 0, + 'unique_tot_ipv4': 0, 'unique_tot_ipv6': 0, 'successful_requests': 0, + 'redirects': 0, 'bad_requests': 0, 'server_errors': 0, 'other_requests': 0 + } + + def check(self): + last_line = self.get_last_line() + if not last_line: + return False + self.storage['unique_all_time'] = list() + self.storage['regex'] = re.compile(r'[0-9.]+\s+(?P<duration>[0-9]+)' + r' (?P<client_address>[\da-f.:]+)' + r' (?P<squid_code>[A-Z_]+)/' + r'(?P<http_code>[0-9]+)' + r' (?P<bytes>[0-9]+)' + r' (?P<method>[A-Z_]+)' + r' (?P<url>[^ ]+)' + r' (?P<user>[^ ]+)' + r' (?P<hier_code>[A-Z_]+)/[\da-f.:-]+' + r' (?P<mime_type>[^\n]+)') + + match = self.storage['regex'].search(last_line) + if not match: + self.error('Regex not matches (%s)' % self.storage['regex'].pattern) + return False + self.storage['dynamic'] = { + 'http_code': + {'chart': 'squid_detailed_response_codes', + 'func_dim_id': None, + 'func_dim': None}, + 'hier_code': { + 'chart': 'squid_hier_code', + 'func_dim_id': None, + 'func_dim': lambda v: v.replace('HIER_', '')}, + 'method': { + 'chart': 'squid_method', + 'func_dim_id': None, + 'func_dim': None}, + 'mime_type': { + 'chart': 'squid_mime_type', + 'func_dim_id': lambda v: v.split('/')[0], + 'func_dim': None}} + return True + + def get_data(self, raw_data=None): + if not raw_data: + return None if raw_data is None else self.data + + filtered_data = self.filter_data(raw_data=raw_data) + + unique_ip = set() + timings = defaultdict(lambda: dict(minimum=None, maximum=0, summary=0, count=0)) + + for row in filtered_data: + match = self.storage['regex'].search(row) + if match: + match = match.groupdict() + if match['duration'] != '0': + get_timings(timings=timings['duration'], time=float(match['duration']) * 1000) + try: + self.data[match['http_code'][0] + 'xx'] += 1 + except KeyError: + self.data['other'] += 1 + + self.get_data_per_statuses(match['http_code']) + + self.get_data_per_squid_code(match['squid_code']) + + self.data['bytes'] += int(match['bytes']) + + proto = 'ipv4' if '.' in match['client_address'] else 'ipv6' + # unique clients ips + if address_not_in_pool(pool=self.storage['unique_all_time'], + address=match['client_address'], + pool_size=self.data['unique_tot_ipv4'] + self.data['unique_tot_ipv6']): + self.data['unique_tot_' + proto] += 1 + + if match['client_address'] not in unique_ip: + self.data['unique_' + proto] += 1 + unique_ip.add(match['client_address']) + + for key, values in self.storage['dynamic'].items(): + if match[key] == '-': + continue + dimension_id = values['func_dim_id'](match[key]) if values['func_dim_id'] else match[key] + if dimension_id not in self.data: + dimension = values['func_dim'](match[key]) if values['func_dim'] else dimension_id + self.add_new_dimension(dimension_id=dimension_id, + chart_key=values['chart'], + dimension=dimension) + self.data[dimension_id] += 1 + else: + self.data['unmatched'] += 1 + + for elem in timings: + self.data[elem + '_min'] += timings[elem]['minimum'] + self.data[elem + '_avg'] += timings[elem]['summary'] / timings[elem]['count'] + self.data[elem + '_max'] += timings[elem]['maximum'] + return self.data + + def get_data_per_statuses(self, code): + """ + :param code: str: response status code. Ex.: '202', '499' + :return: + """ + code_class = code[0] + if code_class == '2' or code == '304' or code_class == '1' or code == '000': + self.data['successful_requests'] += 1 + elif code_class == '3': + self.data['redirects'] += 1 + elif code_class == '4': + self.data['bad_requests'] += 1 + elif code_class == '5' or code_class == '6': + self.data['server_errors'] += 1 + else: + self.data['other_requests'] += 1 + + def get_data_per_squid_code(self, code): + """ + :param code: str: squid response code. Ex.: 'TCP_MISS', 'TCP_MISS_ABORTED' + :return: + """ + if code not in self.data: + self.add_new_dimension(dimension_id=code, chart_key='squid_code') + self.data[code] += 1 + if '_' not in code: + return + for tag in code.split('_'): + try: + chart_key = SQUID_CODES[tag] + except KeyError: + continue + if tag not in self.data: + self.add_new_dimension(dimension_id=tag, chart_key=chart_key) + self.data[tag] += 1 + + +def get_timings(timings, time): + """ + :param timings: + :param time: + :return: + """ + if timings['minimum'] is None: + timings['minimum'] = time + if time > timings['maximum']: + timings['maximum'] = time + elif time < timings['minimum']: + timings['minimum'] = time + timings['summary'] += time + timings['count'] += 1 + + def address_not_in_pool(pool, address, pool_size): """ :param pool: list of ip addresses :param address: ip address :param pool_size: current pool size - :return: True if address not in pool. False if address in pool. + :return: True if address not in pool. False otherwise. """ index = bisect.bisect_left(pool, address) if index < pool_size: if pool[index] == address: return False - else: - bisect.insort_left(pool, address) - return True - else: bisect.insort_left(pool, address) return True + bisect.insort_left(pool, address) + return True def find_regex_return(match_dict=None, msg='Generic error message'): @@ -618,36 +990,53 @@ def find_regex_return(match_dict=None, msg='Generic error message'): return match_dict, msg -def check_req_per_url_pattern(string, url_pattern): +def check_patterns(string, dimension_regex_dict): """ :param string: str: - :param url_pattern: dict: ex. {'dim1': 'pattern1>', 'dim2': '<pattern2>'} + :param dimension_regex_dict: dict: ex. {'dim1': '<pattern1>', 'dim2': '<pattern2>'} :return: list of named tuples or None: We need to make sure all patterns are valid regular expressions """ - if not hasattr(url_pattern, 'keys'): + if not hasattr(dimension_regex_dict, 'keys'): return None result = list() - def is_valid_pattern(pattern): + def valid_pattern(pattern): """ :param pattern: str :return: re.compile(pattern) or None """ if not isinstance(pattern, str): return False - else: - try: - compile_pattern = re.compile(pattern) - except re.error: - return False - else: - return compile_pattern + try: + return re.compile(pattern) + except re.error: + return False - for dimension, regex in url_pattern.items(): - valid_pattern = is_valid_pattern(regex) - if isinstance(dimension, str) and valid_pattern: - result.append(NAMED_URL_PATTERN(description='_'.join([string, dimension]), pattern=valid_pattern)) + def func_search(pattern): + def closure(v): + return pattern.search(v) + + return closure + for dimension, regex in dimension_regex_dict.items(): + valid = valid_pattern(regex) + if isinstance(dimension, str) and valid_pattern: + func = func_search(valid) + result.append(NAMED_PATTERN(description='_'.join([string, dimension]), + func=func)) return result or None + + +def find_job_name(override_name, name): + """ + :param override_name: str: 'name' var from configuration file + :param name: str: 'job_name' from configuration file + :return: str: new job name + We need this for dynamic charts. Actually the same logic as in python.d.plugin. + """ + add_to_name = override_name or name + if add_to_name: + return '_'.join(['web_log', re.sub('\s+', '_', add_to_name)]) + return 'web_log' |