diff options
author | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-05-06 01:22:31 +0000 |
---|---|---|
committer | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-05-06 01:22:31 +0000 |
commit | 8d4f58e49b9dc7d3545651023a36729de773ad86 (patch) | |
tree | 7bc7be4a8e9e298daa1349348400aa2a653866f2 /streaming | |
parent | Initial commit. (diff) | |
download | netdata-upstream/1.12.0.tar.xz netdata-upstream/1.12.0.zip |
Adding upstream version 1.12.0.upstream/1.12.0upstream
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to 'streaming')
-rw-r--r-- | streaming/Makefile.am | 12 | ||||
-rw-r--r-- | streaming/README.md | 417 | ||||
-rw-r--r-- | streaming/rrdpush.c | 1279 | ||||
-rw-r--r-- | streaming/rrdpush.h | 25 | ||||
-rw-r--r-- | streaming/stream.conf | 191 |
5 files changed, 1924 insertions, 0 deletions
diff --git a/streaming/Makefile.am b/streaming/Makefile.am new file mode 100644 index 0000000..8404894 --- /dev/null +++ b/streaming/Makefile.am @@ -0,0 +1,12 @@ +# SPDX-License-Identifier: GPL-3.0-or-later + +AUTOMAKE_OPTIONS = subdir-objects +MAINTAINERCLEANFILES = $(srcdir)/Makefile.in + +dist_libconfig_DATA = \ + stream.conf \ + $(NULL) + +dist_noinst_DATA = \ + README.md \ + $(NULL) diff --git a/streaming/README.md b/streaming/README.md new file mode 100644 index 0000000..e1ebad5 --- /dev/null +++ b/streaming/README.md @@ -0,0 +1,417 @@ +# Streaming and replication + +Each netdata is able to replicate/mirror its database to another netdata, by streaming collected +metrics, in real-time to it. This is quite different to [data archiving to third party time-series +databases](../backends). + +When a netdata streams metrics to another netdata, the receiving one is able to perform everything +a netdata performs: + +- visualize them with a dashboard +- run health checks that trigger alarms and send alarm notifications +- archive metrics to a backend time-series database + +## Supported configurations + +### netdata without a database or web API (headless collector) + +Local netdata (`slave`), **without any database or alarms**, collects metrics and sends them to +another netdata (`master`). + +The `my-netdata` menu shows a list of all "databases streamed to" the master. Clicking one of those links allows the user to view the full dashboard of the `slave` netdata. The URL has the form http://master-host:master-port/host/slave-host/. + +Alarms for the `slave` are served by the `master`. + +In this mode the `slave` is just a plain data collector. It spawns all external plugins, but instead +of maintaining a local database and accepting dashboard requests, it streams all metrics to the +`master`. The memory footprint is reduced significantly, to between 6 MiB and 40 MiB, depending on the enabled plugins. To reduce the memory usage as much as possible, refer to [running netdata in embedded devices](../docs/Performance.md#running-netdata-in-embedded-devices). + +The same `master` can collect data for any number of `slaves`. + +### database replication + +Local netdata (`slave`), **with a local database (and possibly alarms)**, collects metrics and +sends them to another netdata (`master`). + +The user can use all the functions **at both** http://slave-ip:slave-port/ and +http://master-host:master-port/host/slave-host/. + +The `slave` and the `master` may have different data retention policies for the same metrics. + +Alarms for the `slave` are triggered by **both** the `slave` and the `master` (and actually +each can have different alarms configurations or have alarms disabled). + +### netdata proxies + +Local netdata (`slave`), with or without a database, collects metrics and sends them to another +netdata (`proxy`), which may or may not maintain a database, which forwards them to another +netdata (`master`). + +Alarms for the slave can be triggered by any of the involved hosts that maintains a database. + +Any number of daisy chaining netdata servers are supported, each with or without a database and +with or without alarms for the `slave` metrics. + +### mix and match with backends + +All nodes that maintain a database can also send their data to a backend database. +This allows quite complex setups. + +Example: + +1. netdata `A`, `B` do not maintain a database and stream metrics to netdata `C`(live streaming functionality, i.e. this PR) +2. netdata `C` maintains a database for `A`, `B`, `C` and archives all metrics to `graphite` with 10 second detail (backends functionality) +3. netdata `C` also streams data for `A`, `B`, `C` to netdata `D`, which also collects data from `E`, `F` and `G` from another DMZ (live streaming functionality, i.e. this PR) +4. netdata `D` is just a proxy, without a database, that streams all data to a remote site at netdata `H` +5. netdata `H` maintains a database for `A`, `B`, `C`, `D`, `E`, `F`, `G`, `H` and sends all data to `opentsdb` with 5 seconds detail (backends functionality) +6. alarms are triggered by `H` for all hosts +7. users can use all the netdata that maintain a database to view metrics (i.e. at `H` all hosts can be viewed). + +## Configuration + +These are options that affect the operation of netdata in this area: + +``` +[global] + memory mode = none | ram | save | map +``` + +`[global].memory mode = none` disables the database at this host. This also disables health +monitoring (there cannot be health monitoring without a database). + +``` +[web] + mode = none | static-threaded + accept a streaming request every seconds = 0 +``` + +`[web].mode = none` disables the API (netdata will not listen to any ports). +This also disables the registry (there cannot be a registry without an API). + +`accept a streaming request every seconds` can be used to set a limit on how often a master Netdata server will accept streaming requests from the slaves. 0 sets no limit, 1 means maximum once every second. If this is set, you may see error log entries "... too busy to accept new streaming request. Will be allowed in X secs". + +``` +[backend] + enabled = yes | no + type = graphite | opentsdb + destination = IP:PORT ... + update every = 10 +``` + +`[backend]` configures data archiving to a backend (it archives all databases maintained on +this host). + +### streaming configuration + +A new file is introduced: [stream.conf](stream.conf) (to edit it on your system run +`/etc/netdata/edit-config stream.conf`). This file holds streaming configuration for both the +sending and the receiving netdata. + +API keys are used to authorize the communication of a pair of sending-receiving netdata. +Once the communication is authorized, the sending netdata can push metrics for any number of hosts. + +You can generate an API key with the command `uuidgen`. API keys are just random GUIDs. +You can use the same API key on all your netdata, or use a different API key for any pair of +sending-receiving netdata. + +##### options for the sending node + +This is the section for the sending netdata. On the receiving node, `[stream].enabled` can be `no`. +If it is `yes`, the receiving node will also stream the metrics to another node (i.e. it will be +a `proxy`). + +``` +[stream] + enabled = yes | no + destination = IP:PORT ... + api key = XXXXXXXXXXX +``` + +This is an overview of how these options can be combined: + +target | memory<br/>mode | web<br/>mode | stream<br/>enabled | backend | alarms | dashboard +-------|:-----------:|:---:|:------:|:-------:|:---------:|:----: +headless collector|`none`|`none`|`yes`|only for `data source = as collected`|not possible|no +headless proxy|`none`|not `none`|`yes`|only for `data source = as collected`|not possible|no +proxy with db|not `none`|not `none`|`yes`|possible|possible|yes +central netdata|not `none`|not `none`|`no`|possible|possible|yes + +##### options for the receiving node + +`stream.conf` looks like this: + +```sh +# replace API_KEY with your uuidgen generated GUID +[API_KEY] + enabled = yes + default history = 3600 + default memory mode = save + health enabled by default = auto + allow from = * +``` + +You can add many such sections, one for each API key. The above are used as default values for +all hosts pushed with this API key. + +You can also add sections like this: + +```sh +# replace MACHINE_GUID with the slave /var/lib/netdata/registry/netdata.public.unique.id +[MACHINE_GUID] + enabled = yes + history = 3600 + memory mode = save + health enabled = yes + allow from = * +``` + +The above is the receiver configuration of a single host, at the receiver end. `MACHINE_GUID` is +the unique id the netdata generating the metrics (i.e. the netdata that originally collects +them `/var/lib/netdata/registry/netdata.unique.id`). So, metrics for netdata `A` that pass through +any number of other netdata, will have the same `MACHINE_GUID`. + +##### allow from + +`allow from` settings are [netdata simple patterns](../libnetdata/simple_pattern): string matches +that use `*` as wildcard (any number of times) and a `!` prefix for a negative match. +So: `allow from = !10.1.2.3 10.*` will allow all IPs in `10.*` except `10.1.2.3`. The order is +important: left to right, the first positive or negative match is used. + +`allow from` is available in netdata v1.9+ + +##### tracing + +When a `slave` is trying to push metrics to a `master` or `proxy`, it logs entries like these: + +``` +2017-02-25 01:57:44: netdata: ERROR: Failed to connect to '10.11.12.1', port '19999' (errno 111, Connection refused) +2017-02-25 01:57:44: netdata: ERROR: STREAM costa-pc [send to 10.11.12.1:19999]: failed to connect +2017-02-25 01:58:04: netdata: INFO : STREAM costa-pc [send to 10.11.12.1:19999]: initializing communication... +2017-02-25 01:58:04: netdata: INFO : STREAM costa-pc [send to 10.11.12.1:19999]: waiting response from remote netdata... +2017-02-25 01:58:14: netdata: INFO : STREAM costa-pc [send to 10.11.12.1:19999]: established communication - sending metrics... +2017-02-25 01:58:14: netdata: ERROR: STREAM costa-pc [send]: discarding 1900 bytes of metrics already in the buffer. +2017-02-25 01:58:14: netdata: INFO : STREAM costa-pc [send]: ready - sending metrics... +``` + +The receiving end (`proxy` or `master`) logs entries like these: + +``` +2017-02-25 01:58:04: netdata: INFO : STREAM [receive from [10.11.12.11]:33554]: new client connection. +2017-02-25 01:58:04: netdata: INFO : STREAM costa-pc [10.11.12.11]:33554: receive thread created (task id 7698) +2017-02-25 01:58:14: netdata: INFO : Host 'costa-pc' with guid '12345678-b5a6-11e6-8a50-00508db7e9c9' initialized, os: linux, update every: 1, memory mode: ram, history entries: 3600, streaming: disabled, health: enabled, cache_dir: '/var/cache/netdata/12345678-b5a6-11e6-8a50-00508db7e9c9', varlib_dir: '/var/lib/netdata/12345678-b5a6-11e6-8a50-00508db7e9c9', health_log: '/var/lib/netdata/12345678-b5a6-11e6-8a50-00508db7e9c9/health/health-log.db', alarms default handler: '/usr/libexec/netdata/plugins.d/alarm-notify.sh', alarms default recipient: 'root' +2017-02-25 01:58:14: netdata: INFO : STREAM costa-pc [receive from [10.11.12.11]:33554]: initializing communication... +2017-02-25 01:58:14: netdata: INFO : STREAM costa-pc [receive from [10.11.12.11]:33554]: receiving metrics... +``` + +For netdata v1.9+, streaming can also be monitored via `access.log`. + + +## Viewing remote host dashboards, using mirrored databases + +On any receiving netdata, that maintains remote databases and has its web server enabled, +`my-netdata` menu will include a list of the mirrored databases. + +![image](https://cloud.githubusercontent.com/assets/2662304/24080824/24cd2d3c-0caf-11e7-909d-a8dd1dbb95d7.png) + +Selecting any of these, the server will offer a dashboard using the mirrored metrics. + + +## Monitoring ephemeral nodes + +Auto-scaling is probably the most trendy service deployment strategy these days. + +Auto-scaling detects the need for additional resources and boots VMs on demand, based on a template. Soon after they start running the applications, a load balancer starts distributing traffic to them, allowing the service to grow horizontally to the scale needed to handle the load. When demands falls, auto-scaling starts shutting down VMs that are no longer needed. + +<p align="center"> +<img src="https://cloud.githubusercontent.com/assets/2662304/23627426/65a9074a-02b9-11e7-9664-cd8f258a00af.png"/> +</p> + +What a fantastic feature for controlling infrastructure costs! Pay only for what you need for the time you need it! + +In auto-scaling, all servers are ephemeral, they live for just a few hours. Every VM is a brand new instance of the application, that was automatically created based on a template. + +So, how can we monitor them? How can we be sure that everything is working as expected on all of them? + +### The netdata way + +We recently made a significant improvement at the core of netdata to support monitoring such setups. + +Following the netdata way of monitoring, we wanted: + +1. **real-time performance monitoring**, collecting **_thousands of metrics per server per second_**, visualized in interactive, automatically created dashboards. +2. **real-time alarms**, for all nodes. +3. **zero configuration**, all ephemeral servers should have exactly the same configuration, and nothing should be configured at any system for each of the ephemeral nodes. We shouldn't care if 10 or 100 servers are spawned to handle the load. +4. **self-cleanup**, so that nothing needs to be done for cleaning up the monitoring infrastructure from the hundreds of nodes that may have been monitored through time. + +### How it works + +All monitoring solutions, including netdata, work like this: + +1. `collect metrics`, from the system and the running applications +2. `store metrics`, in a time-series database +3. `examine metrics` periodically, for triggering alarms and sending alarm notifications +4. `visualize metrics`, so that users can see what exactly is happening + +netdata used to be self-contained, so that all these functions were handled entirely by each server. The changes we made, allow each netdata to be configured independently for each function. So, each netdata can now act as: + +- a `self contained system`, much like it used to be. +- a `data collector`, that collects metrics from a host and pushes them to another netdata (with or without a local database and alarms). +- a `proxy`, that receives metrics from other hosts and pushes them immediately to other netdata servers. netdata proxies can also be `store and forward proxies` meaning that they are able to maintain a local database for all metrics passing through them (with or without alarms). +- a `time-series database` node, where data are kept, alarms are run and queries are served to visualise the metrics. + +### Configuring an auto-scaling setup + +<p align="center"> +<img src="https://cloud.githubusercontent.com/assets/2662304/23627468/96daf7ba-02b9-11e7-95ac-1f767dd8dab8.png"/> +</p> + +You need a netdata `master`. This node should not be ephemeral. It will be the node where all ephemeral nodes (let's call them `slaves`) will be sending their metrics. + +The master will need to authorize the slaves for accepting their metrics. This is done with an API key. + +#### API keys + +API keys are just random GUIDs. Use the Linux command `uuidgen` to generate one. You can use the same API key for all your `slaves`, or you can configure one API for each of them. This is entirely your decision. + +We suggest to use the same API key for each ephemeral node template you have, so that all replicas of the same ephemeral node will have exactly the same configuration. + +I will use this API_KEY: `11111111-2222-3333-4444-555555555555`. Replace it with your own. + +#### Configuring the `master` + +On the master, edit `/etc/netdata/stream.conf` (to edit it on your system run `/etc/netdata/edit-config stream.conf`) and set these: + +```bash +[11111111-2222-3333-4444-555555555555] + # enable/disable this API key + enabled = yes + + # one hour of data for each of the slaves + default history = 3600 + + # do not save slave metrics on disk + default memory = ram + + # alarms checks, only while the slave is connected + health enabled by default = auto +``` +*`stream.conf` on master, to enable receiving metrics from slaves using the API key.* + +If you used many API keys, you can add one such section for each API key. + +When done, restart netdata on the `master` node. It is now ready to receive metrics. + +#### Configuring the `slaves` + +On each of the slaves, edit `/etc/netdata/stream.conf` (to edit it on your system run `/etc/netdata/edit-config stream.conf`) and set these: + +```bash +[stream] + # stream metrics to another netdata + enabled = yes + + # the IP and PORT of the master + destination = 10.11.12.13:19999 + + # the API key to use + api key = 11111111-2222-3333-4444-555555555555 +``` +*`stream.conf` on slaves, to enable pushing metrics to master at `10.11.12.13:19999`.* + +Using just the above configuration, the `slaves` will be pushing their metrics to the `master` netdata, but they will still maintain a local database of the metrics and run health checks. To disable them, edit `/etc/netdata/netdata.conf` and set: + +```bash +[global] + # disable the local database + memory mode = none + +[health] + # disable health checks + enabled = no +``` +*`netdata.conf` configuration on slaves, to disable the local database and health checks.* + +Keep in mind that setting `memory mode = none` will also force `[health].enabled = no` (health checks require access to a local database). But you can keep the database and disable health checks if you need to. You are however sending all the metrics to the master server, which can handle the health checking (`[health].enabled = yes`) + +#### netdata unique id + +The file `/var/lib/netdata/registry/netdata.public.unique.id` contains a random GUID that **uniquely identifies each netdata**. This file is automatically generated, by netdata, the first time it is started and remains unaltaired forever. + +> If you are building an image to be used for automated provisioning of autoscaled VMs, it important to delete that file from the image, so that each instance of your image will generate its own. + +#### Troubleshooting metrics streaming + +Both the sender and the receiver of metrics log information at `/var/log/netdata/error.log`. + + +On both master and slave do this: + +``` +tail -f /var/log/netdata/error.log | grep STREAM +``` + +If the slave manages to connect to the master you will see something like (on the master): + +``` +2017-03-09 09:38:52: netdata: INFO : STREAM [receive from [10.11.12.86]:38564]: new client connection. +2017-03-09 09:38:52: netdata: INFO : STREAM xxx [10.11.12.86]:38564: receive thread created (task id 27721) +2017-03-09 09:38:52: netdata: INFO : STREAM xxx [receive from [10.11.12.86]:38564]: client willing to stream metrics for host 'xxx' with machine_guid '1234567-1976-11e6-ae19-7cdd9077342a': update every = 1, history = 3600, memory mode = ram, health auto +2017-03-09 09:38:52: netdata: INFO : STREAM xxx [receive from [10.11.12.86]:38564]: initializing communication... +2017-03-09 09:38:52: netdata: INFO : STREAM xxx [receive from [10.11.12.86]:38564]: receiving metrics... +``` + +and something like this on the slave: + +``` +2017-03-09 09:38:28: netdata: INFO : STREAM xxx [send to box:19999]: connecting... +2017-03-09 09:38:28: netdata: INFO : STREAM xxx [send to box:19999]: initializing communication... +2017-03-09 09:38:28: netdata: INFO : STREAM xxx [send to box:19999]: waiting response from remote netdata... +2017-03-09 09:38:28: netdata: INFO : STREAM xxx [send to box:19999]: established communication - sending metrics... +``` + +### Archiving to a time-series database + +The `master` netdata node can also archive metrics, for all `slaves`, to a time-series database. At the time of this writing, netdata supports: + +- graphite +- opentsdb +- prometheus +- json document DBs +- all the compatibles to the above (e.g. kairosdb, influxdb, etc) + +Check the netdata [backends documentation](../backends) for configuring this. + +This is how such a solution will work: + +<p align="center"> +<img src="https://cloud.githubusercontent.com/assets/2662304/23627295/e3569adc-02b8-11e7-9d55-4014bf98c1b3.png"/> +</p> + +### An advanced setup + +netdata also supports `proxies` with and without a local database, and data retention can be different between all nodes. + +This means a setup like the following is also possible: + +<p align="center"> +<img src="https://cloud.githubusercontent.com/assets/2662304/23629551/bb1fd9c2-02c0-11e7-90f5-cab5a3ed4c53.png"/> +</p> + + +## proxies + +A proxy is a netdata that is receiving metrics from a netdata, and streams them to another netdata. + +netdata proxies may or may not maintain a database for the metrics passing through them. +When they maintain a database, they can also run health checks (alarms and notifications) +for the remote host that is streaming the metrics. + +To configure a proxy, configure it as a receiving and a sending netdata at the same time, +using [stream.conf](stream.conf). + +The sending side of a netdata proxy, connects and disconnects to the final destination of the +metrics, following the same pattern of the receiving side. + +For a practical example see [Monitoring ephemeral nodes](#monitoring-ephemeral-nodes). + +[![analytics](https://www.google-analytics.com/collect?v=1&aip=1&t=pageview&_s=1&ds=github&dr=https%3A%2F%2Fgithub.com%2Fnetdata%2Fnetdata&dl=https%3A%2F%2Fmy-netdata.io%2Fgithub%2Fstreaming%2FREADME&_u=MAC~&cid=5792dfd7-8dc4-476b-af31-da2fdb9f93d2&tid=UA-64295674-3)]() diff --git a/streaming/rrdpush.c b/streaming/rrdpush.c new file mode 100644 index 0000000..617484c --- /dev/null +++ b/streaming/rrdpush.c @@ -0,0 +1,1279 @@ +// SPDX-License-Identifier: GPL-3.0-or-later + +#include "rrdpush.h" + +/* + * rrdpush + * + * 3 threads are involved for all stream operations + * + * 1. a random data collection thread, calling rrdset_done_push() + * this is called for each chart. + * + * the output of this work is kept in a BUFFER in RRDHOST + * the sender thread is signalled via a pipe (also in RRDHOST) + * + * 2. a sender thread running at the sending netdata + * this is spawned automatically on the first chart to be pushed + * + * It tries to push the metrics to the remote netdata, as fast + * as possible (i.e. immediately after they are collected). + * + * 3. a receiver thread, running at the receiving netdata + * this is spawned automatically when the sender connects to + * the receiver. + * + */ + +#define START_STREAMING_PROMPT "Hit me baby, push them over..." + +typedef enum { + RRDPUSH_MULTIPLE_CONNECTIONS_ALLOW, + RRDPUSH_MULTIPLE_CONNECTIONS_DENY_NEW +} RRDPUSH_MULTIPLE_CONNECTIONS_STRATEGY; + +static struct config stream_config = { + .sections = NULL, + .mutex = NETDATA_MUTEX_INITIALIZER, + .index = { + .avl_tree = { + .root = NULL, + .compar = appconfig_section_compare + }, + .rwlock = AVL_LOCK_INITIALIZER + } +}; + +unsigned int default_rrdpush_enabled = 0; +char *default_rrdpush_destination = NULL; +char *default_rrdpush_api_key = NULL; +char *default_rrdpush_send_charts_matching = NULL; + +static void load_stream_conf() { + errno = 0; + char *filename = strdupz_path_subpath(netdata_configured_user_config_dir, "stream.conf"); + if(!appconfig_load(&stream_config, filename, 0)) { + info("CONFIG: cannot load user config '%s'. Will try stock config.", filename); + freez(filename); + + filename = strdupz_path_subpath(netdata_configured_stock_config_dir, "stream.conf"); + if(!appconfig_load(&stream_config, filename, 0)) + info("CONFIG: cannot load stock config '%s'. Running with internal defaults.", filename); + } + freez(filename); +} + +int rrdpush_init() { + // -------------------------------------------------------------------- + // load stream.conf + load_stream_conf(); + + default_rrdpush_enabled = (unsigned int)appconfig_get_boolean(&stream_config, CONFIG_SECTION_STREAM, "enabled", default_rrdpush_enabled); + default_rrdpush_destination = appconfig_get(&stream_config, CONFIG_SECTION_STREAM, "destination", ""); + default_rrdpush_api_key = appconfig_get(&stream_config, CONFIG_SECTION_STREAM, "api key", ""); + default_rrdpush_send_charts_matching = appconfig_get(&stream_config, CONFIG_SECTION_STREAM, "send charts matching", "*"); + rrdhost_free_orphan_time = config_get_number(CONFIG_SECTION_GLOBAL, "cleanup orphan hosts after seconds", rrdhost_free_orphan_time); + + if(default_rrdpush_enabled && (!default_rrdpush_destination || !*default_rrdpush_destination || !default_rrdpush_api_key || !*default_rrdpush_api_key)) { + error("STREAM [send]: cannot enable sending thread - information is missing."); + default_rrdpush_enabled = 0; + } + + return default_rrdpush_enabled; +} + +#define CONNECTED_TO_SIZE 100 + +// data collection happens from multiple threads +// each of these threads calls rrdset_done() +// which in turn calls rrdset_done_push() +// which uses this pipe to notify the streaming thread +// that there are more data ready to be sent +#define PIPE_READ 0 +#define PIPE_WRITE 1 + +// to have the remote netdata re-sync the charts +// to its current clock, we send for this many +// iterations a BEGIN line without microseconds +// this is for the first iterations of each chart +unsigned int remote_clock_resync_iterations = 60; + +#define rrdpush_buffer_lock(host) netdata_mutex_lock(&((host)->rrdpush_sender_buffer_mutex)) +#define rrdpush_buffer_unlock(host) netdata_mutex_unlock(&((host)->rrdpush_sender_buffer_mutex)) + +static inline int should_send_chart_matching(RRDSET *st) { + if(unlikely(!rrdset_flag_check(st, RRDSET_FLAG_ENABLED))) { + rrdset_flag_clear(st, RRDSET_FLAG_UPSTREAM_SEND); + rrdset_flag_set(st, RRDSET_FLAG_UPSTREAM_IGNORE); + } + else if(!rrdset_flag_check(st, RRDSET_FLAG_UPSTREAM_SEND|RRDSET_FLAG_UPSTREAM_IGNORE)) { + RRDHOST *host = st->rrdhost; + + if(simple_pattern_matches(host->rrdpush_send_charts_matching, st->id) || + simple_pattern_matches(host->rrdpush_send_charts_matching, st->name)) { + rrdset_flag_clear(st, RRDSET_FLAG_UPSTREAM_IGNORE); + rrdset_flag_set(st, RRDSET_FLAG_UPSTREAM_SEND); + } + else { + rrdset_flag_clear(st, RRDSET_FLAG_UPSTREAM_SEND); + rrdset_flag_set(st, RRDSET_FLAG_UPSTREAM_IGNORE); + } + } + + return(rrdset_flag_check(st, RRDSET_FLAG_UPSTREAM_SEND)); +} + +// checks if the current chart definition has been sent +static inline int need_to_send_chart_definition(RRDSET *st) { + rrdset_check_rdlock(st); + + if(unlikely(!(rrdset_flag_check(st, RRDSET_FLAG_UPSTREAM_EXPOSED)))) + return 1; + + RRDDIM *rd; + rrddim_foreach_read(rd, st) { + if(unlikely(!rd->exposed)) { + #ifdef NETDATA_INTERNAL_CHECKS + info("host '%s', chart '%s', dimension '%s' flag 'exposed' triggered chart refresh to upstream", st->rrdhost->hostname, st->id, rd->id); + #endif + return 1; + } + } + + return 0; +} + +// sends the current chart definition +static inline void rrdpush_send_chart_definition_nolock(RRDSET *st) { + RRDHOST *host = st->rrdhost; + + rrdset_flag_set(st, RRDSET_FLAG_UPSTREAM_EXPOSED); + + // properly set the name for the remote end to parse it + char *name = ""; + if(unlikely(strcmp(st->id, st->name))) { + // they differ + name = strchr(st->name, '.'); + if(name) + name++; + else + name = ""; + } + + // info("CHART '%s' '%s'", st->id, name); + + // send the chart + buffer_sprintf( + host->rrdpush_sender_buffer + , "CHART \"%s\" \"%s\" \"%s\" \"%s\" \"%s\" \"%s\" \"%s\" %ld %d \"%s %s %s %s\" \"%s\" \"%s\"\n" + , st->id + , name + , st->title + , st->units + , st->family + , st->context + , rrdset_type_name(st->chart_type) + , st->priority + , st->update_every + , rrdset_flag_check(st, RRDSET_FLAG_OBSOLETE)?"obsolete":"" + , rrdset_flag_check(st, RRDSET_FLAG_DETAIL)?"detail":"" + , rrdset_flag_check(st, RRDSET_FLAG_STORE_FIRST)?"store_first":"" + , rrdset_flag_check(st, RRDSET_FLAG_HIDDEN)?"hidden":"" + , (st->plugin_name)?st->plugin_name:"" + , (st->module_name)?st->module_name:"" + ); + + // send the dimensions + RRDDIM *rd; + rrddim_foreach_read(rd, st) { + buffer_sprintf( + host->rrdpush_sender_buffer + , "DIMENSION \"%s\" \"%s\" \"%s\" " COLLECTED_NUMBER_FORMAT " " COLLECTED_NUMBER_FORMAT " \"%s %s\"\n" + , rd->id + , rd->name + , rrd_algorithm_name(rd->algorithm) + , rd->multiplier + , rd->divisor + , rrddim_flag_check(rd, RRDDIM_FLAG_HIDDEN)?"hidden":"" + , rrddim_flag_check(rd, RRDDIM_FLAG_DONT_DETECT_RESETS_OR_OVERFLOWS)?"noreset":"" + ); + rd->exposed = 1; + } + + // send the chart local custom variables + RRDSETVAR *rs; + for(rs = st->variables; rs ;rs = rs->next) { + if(unlikely(rs->type == RRDVAR_TYPE_CALCULATED && rs->options & RRDVAR_OPTION_CUSTOM_CHART_VAR)) { + calculated_number *value = (calculated_number *) rs->value; + + buffer_sprintf( + host->rrdpush_sender_buffer + , "VARIABLE CHART %s = " CALCULATED_NUMBER_FORMAT "\n" + , rs->variable + , *value + ); + } + } + + st->upstream_resync_time = st->last_collected_time.tv_sec + (remote_clock_resync_iterations * st->update_every); +} + +// sends the current chart dimensions +static inline void rrdpush_send_chart_metrics_nolock(RRDSET *st) { + RRDHOST *host = st->rrdhost; + buffer_sprintf(host->rrdpush_sender_buffer, "BEGIN \"%s\" %llu\n", st->id, (st->last_collected_time.tv_sec > st->upstream_resync_time)?st->usec_since_last_update:0); + + RRDDIM *rd; + rrddim_foreach_read(rd, st) { + if(rd->updated && rd->exposed) + buffer_sprintf(host->rrdpush_sender_buffer + , "SET \"%s\" = " COLLECTED_NUMBER_FORMAT "\n" + , rd->id + , rd->collected_value + ); + } + + buffer_strcat(host->rrdpush_sender_buffer, "END\n"); +} + +static void rrdpush_sender_thread_spawn(RRDHOST *host); + +void rrdset_push_chart_definition_now(RRDSET *st) { + RRDHOST *host = st->rrdhost; + + if(unlikely(!host->rrdpush_send_enabled || !should_send_chart_matching(st))) + return; + + rrdset_rdlock(st); + rrdpush_buffer_lock(host); + rrdpush_send_chart_definition_nolock(st); + rrdpush_buffer_unlock(host); + rrdset_unlock(st); +} + +void rrdset_done_push(RRDSET *st) { + if(unlikely(!should_send_chart_matching(st))) + return; + + RRDHOST *host = st->rrdhost; + + rrdpush_buffer_lock(host); + + if(unlikely(host->rrdpush_send_enabled && !host->rrdpush_sender_spawn)) + rrdpush_sender_thread_spawn(host); + + if(unlikely(!host->rrdpush_sender_buffer || !host->rrdpush_sender_connected)) { + if(unlikely(!host->rrdpush_sender_error_shown)) + error("STREAM %s [send]: not ready - discarding collected metrics.", host->hostname); + + host->rrdpush_sender_error_shown = 1; + + rrdpush_buffer_unlock(host); + return; + } + else if(unlikely(host->rrdpush_sender_error_shown)) { + info("STREAM %s [send]: sending metrics...", host->hostname); + host->rrdpush_sender_error_shown = 0; + } + + if(need_to_send_chart_definition(st)) + rrdpush_send_chart_definition_nolock(st); + + rrdpush_send_chart_metrics_nolock(st); + + // signal the sender there are more data + if(host->rrdpush_sender_pipe[PIPE_WRITE] != -1 && write(host->rrdpush_sender_pipe[PIPE_WRITE], " ", 1) == -1) + error("STREAM %s [send]: cannot write to internal pipe", host->hostname); + + rrdpush_buffer_unlock(host); +} + +// ---------------------------------------------------------------------------- +// rrdpush sender thread + +static inline void rrdpush_sender_add_host_variable_to_buffer_nolock(RRDHOST *host, RRDVAR *rv) { + calculated_number *value = (calculated_number *)rv->value; + + buffer_sprintf( + host->rrdpush_sender_buffer + , "VARIABLE HOST %s = " CALCULATED_NUMBER_FORMAT "\n" + , rv->name + , *value + ); + + debug(D_STREAM, "RRDVAR pushed HOST VARIABLE %s = " CALCULATED_NUMBER_FORMAT, rv->name, *value); +} + +void rrdpush_sender_send_this_host_variable_now(RRDHOST *host, RRDVAR *rv) { + if(host->rrdpush_send_enabled && host->rrdpush_sender_spawn && host->rrdpush_sender_connected) { + rrdpush_buffer_lock(host); + rrdpush_sender_add_host_variable_to_buffer_nolock(host, rv); + rrdpush_buffer_unlock(host); + } +} + +static int rrdpush_sender_thread_custom_host_variables_callback(void *rrdvar_ptr, void *host_ptr) { + RRDVAR *rv = (RRDVAR *)rrdvar_ptr; + RRDHOST *host = (RRDHOST *)host_ptr; + + if(unlikely(rv->options & RRDVAR_OPTION_CUSTOM_HOST_VAR && rv->type == RRDVAR_TYPE_CALCULATED)) { + rrdpush_sender_add_host_variable_to_buffer_nolock(host, rv); + + // return 1, so that the traversal will return the number of variables sent + return 1; + } + + // returning a negative number will break the traversal + return 0; +} + +static void rrdpush_sender_thread_send_custom_host_variables(RRDHOST *host) { + int ret = rrdvar_callback_for_all_host_variables(host, rrdpush_sender_thread_custom_host_variables_callback, host); + (void)ret; + + debug(D_STREAM, "RRDVAR sent %d VARIABLES", ret); +} + +// resets all the chart, so that their definitions +// will be resent to the central netdata +static void rrdpush_sender_thread_reset_all_charts(RRDHOST *host) { + rrdhost_rdlock(host); + + RRDSET *st; + rrdset_foreach_read(st, host) { + rrdset_flag_clear(st, RRDSET_FLAG_UPSTREAM_EXPOSED); + + st->upstream_resync_time = 0; + + rrdset_rdlock(st); + + RRDDIM *rd; + rrddim_foreach_read(rd, st) + rd->exposed = 0; + + rrdset_unlock(st); + } + + rrdhost_unlock(host); +} + +static inline void rrdpush_sender_thread_data_flush(RRDHOST *host) { + rrdpush_buffer_lock(host); + + if(buffer_strlen(host->rrdpush_sender_buffer)) + error("STREAM %s [send]: discarding %zu bytes of metrics already in the buffer.", host->hostname, buffer_strlen(host->rrdpush_sender_buffer)); + + buffer_flush(host->rrdpush_sender_buffer); + + rrdpush_sender_thread_reset_all_charts(host); + rrdpush_sender_thread_send_custom_host_variables(host); + + rrdpush_buffer_unlock(host); +} + +void rrdpush_sender_thread_stop(RRDHOST *host) { + rrdpush_buffer_lock(host); + rrdhost_wrlock(host); + + netdata_thread_t thr = 0; + + if(host->rrdpush_sender_spawn) { + info("STREAM %s [send]: signaling sending thread to stop...", host->hostname); + + // signal the thread that we want to join it + host->rrdpush_sender_join = 1; + + // copy the thread id, so that we will be waiting for the right one + // even if a new one has been spawn + thr = host->rrdpush_sender_thread; + + // signal it to cancel + netdata_thread_cancel(host->rrdpush_sender_thread); + } + + rrdhost_unlock(host); + rrdpush_buffer_unlock(host); + + if(thr != 0) { + info("STREAM %s [send]: waiting for the sending thread to stop...", host->hostname); + void *result; + netdata_thread_join(thr, &result); + info("STREAM %s [send]: sending thread has exited.", host->hostname); + } +} + +static inline void rrdpush_sender_thread_close_socket(RRDHOST *host) { + host->rrdpush_sender_connected = 0; + + if(host->rrdpush_sender_socket != -1) { + close(host->rrdpush_sender_socket); + host->rrdpush_sender_socket = -1; + } +} + +static int rrdpush_sender_thread_connect_to_master(RRDHOST *host, int default_port, int timeout, size_t *reconnects_counter, char *connected_to, size_t connected_to_size) { + struct timeval tv = { + .tv_sec = timeout, + .tv_usec = 0 + }; + + // make sure the socket is closed + rrdpush_sender_thread_close_socket(host); + + debug(D_STREAM, "STREAM: Attempting to connect..."); + info("STREAM %s [send to %s]: connecting...", host->hostname, host->rrdpush_send_destination); + + host->rrdpush_sender_socket = connect_to_one_of( + host->rrdpush_send_destination + , default_port + , &tv + , reconnects_counter + , connected_to + , connected_to_size + ); + + if(unlikely(host->rrdpush_sender_socket == -1)) { + error("STREAM %s [send to %s]: failed to connect", host->hostname, host->rrdpush_send_destination); + return 0; + } + + info("STREAM %s [send to %s]: initializing communication...", host->hostname, connected_to); + + #define HTTP_HEADER_SIZE 8192 + char http[HTTP_HEADER_SIZE + 1]; + snprintfz(http, HTTP_HEADER_SIZE, + "STREAM key=%s&hostname=%s®istry_hostname=%s&machine_guid=%s&update_every=%d&os=%s&timezone=%s&tags=%s HTTP/1.1\r\n" + "User-Agent: %s/%s\r\n" + "Accept: */*\r\n\r\n" + , host->rrdpush_send_api_key + , host->hostname + , host->registry_hostname + , host->machine_guid + , default_rrd_update_every + , host->os + , host->timezone + , (host->tags)?host->tags:"" + , host->program_name + , host->program_version + ); + + if(send_timeout(host->rrdpush_sender_socket, http, strlen(http), 0, timeout) == -1) { + error("STREAM %s [send to %s]: failed to send HTTP header to remote netdata.", host->hostname, connected_to); + rrdpush_sender_thread_close_socket(host); + return 0; + } + + info("STREAM %s [send to %s]: waiting response from remote netdata...", host->hostname, connected_to); + + if(recv_timeout(host->rrdpush_sender_socket, http, HTTP_HEADER_SIZE, 0, timeout) == -1) { + error("STREAM %s [send to %s]: remote netdata does not respond.", host->hostname, connected_to); + rrdpush_sender_thread_close_socket(host); + return 0; + } + + if(strncmp(http, START_STREAMING_PROMPT, strlen(START_STREAMING_PROMPT)) != 0) { + error("STREAM %s [send to %s]: server is not replying properly (is it a netdata?).", host->hostname, connected_to); + rrdpush_sender_thread_close_socket(host); + return 0; + } + + info("STREAM %s [send to %s]: established communication - ready to send metrics...", host->hostname, connected_to); + + if(sock_setnonblock(host->rrdpush_sender_socket) < 0) + error("STREAM %s [send to %s]: cannot set non-blocking mode for socket.", host->hostname, connected_to); + + if(sock_enlarge_out(host->rrdpush_sender_socket) < 0) + error("STREAM %s [send to %s]: cannot enlarge the socket buffer.", host->hostname, connected_to); + + debug(D_STREAM, "STREAM: Connected on fd %d...", host->rrdpush_sender_socket); + + return 1; +} + +static void rrdpush_sender_thread_cleanup_callback(void *ptr) { + RRDHOST *host = (RRDHOST *)ptr; + + rrdpush_buffer_lock(host); + rrdhost_wrlock(host); + + info("STREAM %s [send]: sending thread cleans up...", host->hostname); + + rrdpush_sender_thread_close_socket(host); + + // close the pipe + if(host->rrdpush_sender_pipe[PIPE_READ] != -1) { + close(host->rrdpush_sender_pipe[PIPE_READ]); + host->rrdpush_sender_pipe[PIPE_READ] = -1; + } + + if(host->rrdpush_sender_pipe[PIPE_WRITE] != -1) { + close(host->rrdpush_sender_pipe[PIPE_WRITE]); + host->rrdpush_sender_pipe[PIPE_WRITE] = -1; + } + + buffer_free(host->rrdpush_sender_buffer); + host->rrdpush_sender_buffer = NULL; + + if(!host->rrdpush_sender_join) { + info("STREAM %s [send]: sending thread detaches itself.", host->hostname); + netdata_thread_detach(netdata_thread_self()); + } + + host->rrdpush_sender_spawn = 0; + + info("STREAM %s [send]: sending thread now exits.", host->hostname); + + rrdhost_unlock(host); + rrdpush_buffer_unlock(host); +} + +void *rrdpush_sender_thread(void *ptr) { + RRDHOST *host = (RRDHOST *)ptr; + + if(!host->rrdpush_send_enabled || !host->rrdpush_send_destination || !*host->rrdpush_send_destination || !host->rrdpush_send_api_key || !*host->rrdpush_send_api_key) { + error("STREAM %s [send]: thread created (task id %d), but host has streaming disabled.", host->hostname, gettid()); + return NULL; + } + + info("STREAM %s [send]: thread created (task id %d)", host->hostname, gettid()); + + int timeout = (int)appconfig_get_number(&stream_config, CONFIG_SECTION_STREAM, "timeout seconds", 60); + int default_port = (int)appconfig_get_number(&stream_config, CONFIG_SECTION_STREAM, "default port", 19999); + size_t max_size = (size_t)appconfig_get_number(&stream_config, CONFIG_SECTION_STREAM, "buffer size bytes", 1024 * 1024); + unsigned int reconnect_delay = (unsigned int)appconfig_get_number(&stream_config, CONFIG_SECTION_STREAM, "reconnect delay seconds", 5); + remote_clock_resync_iterations = (unsigned int)appconfig_get_number(&stream_config, CONFIG_SECTION_STREAM, "initial clock resync iterations", remote_clock_resync_iterations); + char connected_to[CONNECTED_TO_SIZE + 1] = ""; + + // initialize rrdpush globals + host->rrdpush_sender_buffer = buffer_create(1); + host->rrdpush_sender_connected = 0; + if(pipe(host->rrdpush_sender_pipe) == -1) fatal("STREAM %s [send]: cannot create required pipe.", host->hostname); + + // initialize local variables + size_t begin = 0; + size_t reconnects_counter = 0; + size_t sent_bytes = 0; + size_t sent_bytes_on_this_connection = 0; + size_t send_attempts = 0; + + + time_t last_sent_t = 0; + struct pollfd fds[2], *ifd, *ofd; + nfds_t fdmax; + + ifd = &fds[0]; + ofd = &fds[1]; + + size_t not_connected_loops = 0; + + netdata_thread_cleanup_push(rrdpush_sender_thread_cleanup_callback, host); + + for(; host->rrdpush_send_enabled && !netdata_exit ;) { + // check for outstanding cancellation requests + netdata_thread_testcancel(); + + // if we don't have socket open, lets wait a bit + if(unlikely(host->rrdpush_sender_socket == -1)) { + send_attempts = 0; + + if(not_connected_loops == 0 && sent_bytes_on_this_connection > 0) { + // fast re-connection on first disconnect + sleep_usec(USEC_PER_MS * 500); // milliseconds + } + else { + // slow re-connection on repeating errors + sleep_usec(USEC_PER_SEC * reconnect_delay); // seconds + } + + if(rrdpush_sender_thread_connect_to_master(host, default_port, timeout, &reconnects_counter, connected_to, CONNECTED_TO_SIZE)) { + last_sent_t = now_monotonic_sec(); + + // reset the buffer, to properly send charts and metrics + rrdpush_sender_thread_data_flush(host); + + // send from the beginning + begin = 0; + + // make sure the next reconnection will be immediate + not_connected_loops = 0; + + // reset the bytes we have sent for this session + sent_bytes_on_this_connection = 0; + + // let the data collection threads know we are ready + host->rrdpush_sender_connected = 1; + } + else { + // increase the failed connections counter + not_connected_loops++; + + // reset the number of bytes sent + sent_bytes_on_this_connection = 0; + } + + // loop through + continue; + } + else if(unlikely(now_monotonic_sec() - last_sent_t > timeout)) { + error("STREAM %s [send to %s]: could not send metrics for %d seconds - closing connection - we have sent %zu bytes on this connection via %zu send attempts.", host->hostname, connected_to, timeout, sent_bytes_on_this_connection, send_attempts); + rrdpush_sender_thread_close_socket(host); + } + + ifd->fd = host->rrdpush_sender_pipe[PIPE_READ]; + ifd->events = POLLIN; + ifd->revents = 0; + + ofd->fd = host->rrdpush_sender_socket; + ofd->revents = 0; + if(ofd->fd != -1 && begin < buffer_strlen(host->rrdpush_sender_buffer)) { + debug(D_STREAM, "STREAM: Requesting data output on streaming socket %d...", ofd->fd); + ofd->events = POLLOUT; + fdmax = 2; + send_attempts++; + } + else { + debug(D_STREAM, "STREAM: Not requesting data output on streaming socket %d (nothing to send now)...", ofd->fd); + ofd->events = 0; + fdmax = 1; + } + + debug(D_STREAM, "STREAM: Waiting for poll() events (current buffer length %zu bytes)...", buffer_strlen(host->rrdpush_sender_buffer)); + if(unlikely(netdata_exit)) break; + int retval = poll(fds, fdmax, 1000); + if(unlikely(netdata_exit)) break; + + if(unlikely(retval == -1)) { + debug(D_STREAM, "STREAM: poll() failed (current buffer length %zu bytes)...", buffer_strlen(host->rrdpush_sender_buffer)); + + if(errno == EAGAIN || errno == EINTR) { + debug(D_STREAM, "STREAM: poll() failed with EAGAIN or EINTR..."); + } + else { + error("STREAM %s [send to %s]: failed to poll(). Closing socket.", host->hostname, connected_to); + rrdpush_sender_thread_close_socket(host); + } + + continue; + } + else if(likely(retval)) { + if (ifd->revents & POLLIN || ifd->revents & POLLPRI) { + debug(D_STREAM, "STREAM: Data added to send buffer (current buffer length %zu bytes)...", buffer_strlen(host->rrdpush_sender_buffer)); + + char buffer[1000 + 1]; + if (read(host->rrdpush_sender_pipe[PIPE_READ], buffer, 1000) == -1) + error("STREAM %s [send to %s]: cannot read from internal pipe.", host->hostname, connected_to); + } + + if (ofd->revents & POLLOUT) { + if (begin < buffer_strlen(host->rrdpush_sender_buffer)) { + debug(D_STREAM, "STREAM: Sending data (current buffer length %zu bytes, begin = %zu)...", buffer_strlen(host->rrdpush_sender_buffer), begin); + + // BEGIN RRDPUSH LOCKED SESSION + + // during this session, data collectors + // will not be able to append data to our buffer + // but the socket is in non-blocking mode + // so, we will not block at send() + + netdata_thread_disable_cancelability(); + + debug(D_STREAM, "STREAM: Getting exclusive lock on host..."); + rrdpush_buffer_lock(host); + + debug(D_STREAM, "STREAM: Sending data, starting from %zu, size %zu...", begin, buffer_strlen(host->rrdpush_sender_buffer)); + ssize_t ret = send(host->rrdpush_sender_socket, &host->rrdpush_sender_buffer->buffer[begin], buffer_strlen(host->rrdpush_sender_buffer) - begin, MSG_DONTWAIT); + if (unlikely(ret == -1)) { + if (errno != EAGAIN && errno != EINTR && errno != EWOULDBLOCK) { + debug(D_STREAM, "STREAM: Send failed - closing socket..."); + error("STREAM %s [send to %s]: failed to send metrics - closing connection - we have sent %zu bytes on this connection.", host->hostname, connected_to, sent_bytes_on_this_connection); + rrdpush_sender_thread_close_socket(host); + } + else { + debug(D_STREAM, "STREAM: Send failed - will retry..."); + } + } + else if (likely(ret > 0)) { + // DEBUG - dump the string to see it + //char c = host->rrdpush_sender_buffer->buffer[begin + ret]; + //host->rrdpush_sender_buffer->buffer[begin + ret] = '\0'; + //debug(D_STREAM, "STREAM: sent from %zu to %zd:\n%s\n", begin, ret, &host->rrdpush_sender_buffer->buffer[begin]); + //host->rrdpush_sender_buffer->buffer[begin + ret] = c; + + sent_bytes_on_this_connection += ret; + sent_bytes += ret; + begin += ret; + + if (begin == buffer_strlen(host->rrdpush_sender_buffer)) { + // we send it all + + debug(D_STREAM, "STREAM: Sent %zd bytes (the whole buffer)...", ret); + buffer_flush(host->rrdpush_sender_buffer); + begin = 0; + } + else { + debug(D_STREAM, "STREAM: Sent %zd bytes (part of the data buffer)...", ret); + } + + last_sent_t = now_monotonic_sec(); + } + else { + debug(D_STREAM, "STREAM: send() returned %zd - closing the socket...", ret); + error("STREAM %s [send to %s]: failed to send metrics (send() returned %zd) - closing connection - we have sent %zu bytes on this connection.", + host->hostname, connected_to, ret, sent_bytes_on_this_connection); + rrdpush_sender_thread_close_socket(host); + } + + debug(D_STREAM, "STREAM: Releasing exclusive lock on host..."); + rrdpush_buffer_unlock(host); + + netdata_thread_enable_cancelability(); + + // END RRDPUSH LOCKED SESSION + } + else { + debug(D_STREAM, "STREAM: we have sent the entire buffer, but we received POLLOUT..."); + } + } + + if(host->rrdpush_sender_socket != -1) { + char *error = NULL; + + if (unlikely(ofd->revents & POLLERR)) + error = "socket reports errors (POLLERR)"; + + else if (unlikely(ofd->revents & POLLHUP)) + error = "connection closed by remote end (POLLHUP)"; + + else if (unlikely(ofd->revents & POLLNVAL)) + error = "connection is invalid (POLLNVAL)"; + + if(unlikely(error)) { + debug(D_STREAM, "STREAM: %s - closing socket...", error); + error("STREAM %s [send to %s]: %s - reopening socket - we have sent %zu bytes on this connection.", host->hostname, connected_to, error, sent_bytes_on_this_connection); + rrdpush_sender_thread_close_socket(host); + } + } + } + else { + debug(D_STREAM, "STREAM: poll() timed out."); + } + + // protection from overflow + if(buffer_strlen(host->rrdpush_sender_buffer) > max_size) { + debug(D_STREAM, "STREAM: Buffer is too big (%zu bytes), bigger than the max (%zu) - flushing it...", buffer_strlen(host->rrdpush_sender_buffer), max_size); + errno = 0; + error("STREAM %s [send to %s]: too many data pending - buffer is %zu bytes long, %zu unsent - we have sent %zu bytes in total, %zu on this connection. Closing connection to flush the data.", host->hostname, connected_to, host->rrdpush_sender_buffer->len, host->rrdpush_sender_buffer->len - begin, sent_bytes, sent_bytes_on_this_connection); + rrdpush_sender_thread_close_socket(host); + } + } + + netdata_thread_cleanup_pop(1); + return NULL; +} + + +// ---------------------------------------------------------------------------- +// rrdpush receiver thread + +static void log_stream_connection(const char *client_ip, const char *client_port, const char *api_key, const char *machine_guid, const char *host, const char *msg) { + log_access("STREAM: %d '[%s]:%s' '%s' host '%s' api key '%s' machine guid '%s'", gettid(), client_ip, client_port, msg, host, api_key, machine_guid); +} + +static RRDPUSH_MULTIPLE_CONNECTIONS_STRATEGY get_multiple_connections_strategy(struct config *c, const char *section, const char *name, RRDPUSH_MULTIPLE_CONNECTIONS_STRATEGY def) { + char *value; + switch(def) { + default: + case RRDPUSH_MULTIPLE_CONNECTIONS_ALLOW: + value = "allow"; + break; + + case RRDPUSH_MULTIPLE_CONNECTIONS_DENY_NEW: + value = "deny"; + break; + } + + value = appconfig_get(c, section, name, value); + + RRDPUSH_MULTIPLE_CONNECTIONS_STRATEGY ret = def; + + if(strcasecmp(value, "allow") == 0 || strcasecmp(value, "permit") == 0 || strcasecmp(value, "accept") == 0) + ret = RRDPUSH_MULTIPLE_CONNECTIONS_ALLOW; + + else if(strcasecmp(value, "deny") == 0 || strcasecmp(value, "reject") == 0 || strcasecmp(value, "block") == 0) + ret = RRDPUSH_MULTIPLE_CONNECTIONS_DENY_NEW; + + else + error("Invalid stream config value at section [%s], setting '%s', value '%s'", section, name, value); + + return ret; +} + +static int rrdpush_receive(int fd + , const char *key + , const char *hostname + , const char *registry_hostname + , const char *machine_guid + , const char *os + , const char *timezone + , const char *tags + , const char *program_name + , const char *program_version + , int update_every + , char *client_ip + , char *client_port +) { + RRDHOST *host; + int history = default_rrd_history_entries; + RRD_MEMORY_MODE mode = default_rrd_memory_mode; + int health_enabled = default_health_enabled; + int rrdpush_enabled = default_rrdpush_enabled; + char *rrdpush_destination = default_rrdpush_destination; + char *rrdpush_api_key = default_rrdpush_api_key; + char *rrdpush_send_charts_matching = default_rrdpush_send_charts_matching; + time_t alarms_delay = 60; + RRDPUSH_MULTIPLE_CONNECTIONS_STRATEGY rrdpush_multiple_connections_strategy = RRDPUSH_MULTIPLE_CONNECTIONS_ALLOW; + + update_every = (int)appconfig_get_number(&stream_config, machine_guid, "update every", update_every); + if(update_every < 0) update_every = 1; + + history = (int)appconfig_get_number(&stream_config, key, "default history", history); + history = (int)appconfig_get_number(&stream_config, machine_guid, "history", history); + if(history < 5) history = 5; + + mode = rrd_memory_mode_id(appconfig_get(&stream_config, key, "default memory mode", rrd_memory_mode_name(mode))); + mode = rrd_memory_mode_id(appconfig_get(&stream_config, machine_guid, "memory mode", rrd_memory_mode_name(mode))); + + health_enabled = appconfig_get_boolean_ondemand(&stream_config, key, "health enabled by default", health_enabled); + health_enabled = appconfig_get_boolean_ondemand(&stream_config, machine_guid, "health enabled", health_enabled); + + alarms_delay = appconfig_get_number(&stream_config, key, "default postpone alarms on connect seconds", alarms_delay); + alarms_delay = appconfig_get_number(&stream_config, machine_guid, "postpone alarms on connect seconds", alarms_delay); + + rrdpush_enabled = appconfig_get_boolean(&stream_config, key, "default proxy enabled", rrdpush_enabled); + rrdpush_enabled = appconfig_get_boolean(&stream_config, machine_guid, "proxy enabled", rrdpush_enabled); + + rrdpush_destination = appconfig_get(&stream_config, key, "default proxy destination", rrdpush_destination); + rrdpush_destination = appconfig_get(&stream_config, machine_guid, "proxy destination", rrdpush_destination); + + rrdpush_api_key = appconfig_get(&stream_config, key, "default proxy api key", rrdpush_api_key); + rrdpush_api_key = appconfig_get(&stream_config, machine_guid, "proxy api key", rrdpush_api_key); + + rrdpush_multiple_connections_strategy = get_multiple_connections_strategy(&stream_config, key, "multiple connections", rrdpush_multiple_connections_strategy); + rrdpush_multiple_connections_strategy = get_multiple_connections_strategy(&stream_config, machine_guid, "multiple connections", rrdpush_multiple_connections_strategy); + + rrdpush_send_charts_matching = appconfig_get(&stream_config, key, "default proxy send charts matching", rrdpush_send_charts_matching); + rrdpush_send_charts_matching = appconfig_get(&stream_config, machine_guid, "proxy send charts matching", rrdpush_send_charts_matching); + + tags = appconfig_set_default(&stream_config, machine_guid, "host tags", (tags)?tags:""); + if(tags && !*tags) tags = NULL; + + if(!strcmp(machine_guid, "localhost")) + host = localhost; + else + host = rrdhost_find_or_create( + hostname + , registry_hostname + , machine_guid + , os + , timezone + , tags + , program_name + , program_version + , update_every + , history + , mode + , (unsigned int)(health_enabled != CONFIG_BOOLEAN_NO) + , (unsigned int)(rrdpush_enabled && rrdpush_destination && *rrdpush_destination && rrdpush_api_key && *rrdpush_api_key) + , rrdpush_destination + , rrdpush_api_key + , rrdpush_send_charts_matching + ); + + if(!host) { + close(fd); + log_stream_connection(client_ip, client_port, key, machine_guid, hostname, "FAILED - CANNOT ACQUIRE HOST"); + error("STREAM %s [receive from [%s]:%s]: failed to find/create host structure.", hostname, client_ip, client_port); + return 1; + } + +#ifdef NETDATA_INTERNAL_CHECKS + info("STREAM %s [receive from [%s]:%s]: client willing to stream metrics for host '%s' with machine_guid '%s': update every = %d, history = %ld, memory mode = %s, health %s, tags '%s'" + , hostname + , client_ip + , client_port + , host->hostname + , host->machine_guid + , host->rrd_update_every + , host->rrd_history_entries + , rrd_memory_mode_name(host->rrd_memory_mode) + , (health_enabled == CONFIG_BOOLEAN_NO)?"disabled":((health_enabled == CONFIG_BOOLEAN_YES)?"enabled":"auto") + , host->tags?host->tags:"" + ); +#endif // NETDATA_INTERNAL_CHECKS + + struct plugind cd = { + .enabled = 1, + .update_every = default_rrd_update_every, + .pid = 0, + .serial_failures = 0, + .successful_collections = 0, + .obsolete = 0, + .started_t = now_realtime_sec(), + .next = NULL, + }; + + // put the client IP and port into the buffers used by plugins.d + snprintfz(cd.id, CONFIG_MAX_NAME, "%s:%s", client_ip, client_port); + snprintfz(cd.filename, FILENAME_MAX, "%s:%s", client_ip, client_port); + snprintfz(cd.fullfilename, FILENAME_MAX, "%s:%s", client_ip, client_port); + snprintfz(cd.cmd, PLUGINSD_CMD_MAX, "%s:%s", client_ip, client_port); + + info("STREAM %s [receive from [%s]:%s]: initializing communication...", host->hostname, client_ip, client_port); + if(send_timeout(fd, START_STREAMING_PROMPT, strlen(START_STREAMING_PROMPT), 0, 60) != strlen(START_STREAMING_PROMPT)) { + log_stream_connection(client_ip, client_port, key, host->machine_guid, host->hostname, "FAILED - CANNOT REPLY"); + error("STREAM %s [receive from [%s]:%s]: cannot send ready command.", host->hostname, client_ip, client_port); + close(fd); + return 0; + } + + // remove the non-blocking flag from the socket + if(sock_delnonblock(fd) < 0) + error("STREAM %s [receive from [%s]:%s]: cannot remove the non-blocking flag from socket %d", host->hostname, client_ip, client_port, fd); + + // convert the socket to a FILE * + FILE *fp = fdopen(fd, "r"); + if(!fp) { + log_stream_connection(client_ip, client_port, key, host->machine_guid, host->hostname, "FAILED - SOCKET ERROR"); + error("STREAM %s [receive from [%s]:%s]: failed to get a FILE for FD %d.", host->hostname, client_ip, client_port, fd); + close(fd); + return 0; + } + + rrdhost_wrlock(host); + if(host->connected_senders > 0) { + switch(rrdpush_multiple_connections_strategy) { + case RRDPUSH_MULTIPLE_CONNECTIONS_ALLOW: + info("STREAM %s [receive from [%s]:%s]: multiple streaming connections for the same host detected. If multiple netdata are pushing metrics for the same charts, at the same time, the result is unexpected.", host->hostname, client_ip, client_port); + break; + + case RRDPUSH_MULTIPLE_CONNECTIONS_DENY_NEW: + rrdhost_unlock(host); + log_stream_connection(client_ip, client_port, key, host->machine_guid, host->hostname, "REJECTED - ALREADY CONNECTED"); + info("STREAM %s [receive from [%s]:%s]: multiple streaming connections for the same host detected. Rejecting new connection.", host->hostname, client_ip, client_port); + fclose(fp); + return 0; + } + } + + rrdhost_flag_clear(host, RRDHOST_FLAG_ORPHAN); + host->connected_senders++; + host->senders_disconnected_time = 0; + if(health_enabled != CONFIG_BOOLEAN_NO) { + if(alarms_delay > 0) { + host->health_delay_up_to = now_realtime_sec() + alarms_delay; + info("Postponing health checks for %ld seconds, on host '%s', because it was just connected." + , alarms_delay + , host->hostname + ); + } + } + rrdhost_unlock(host); + + // call the plugins.d processor to receive the metrics + info("STREAM %s [receive from [%s]:%s]: receiving metrics...", host->hostname, client_ip, client_port); + log_stream_connection(client_ip, client_port, key, host->machine_guid, host->hostname, "CONNECTED"); + + size_t count = pluginsd_process(host, &cd, fp, 1); + + log_stream_connection(client_ip, client_port, key, host->machine_guid, host->hostname, "DISCONNECTED"); + error("STREAM %s [receive from [%s]:%s]: disconnected (completed %zu updates).", host->hostname, client_ip, client_port, count); + + rrdhost_wrlock(host); + host->senders_disconnected_time = now_realtime_sec(); + host->connected_senders--; + if(!host->connected_senders) { + rrdhost_flag_set(host, RRDHOST_FLAG_ORPHAN); + if(health_enabled == CONFIG_BOOLEAN_AUTO) + host->health_enabled = 0; + } + rrdhost_unlock(host); + + if(host->connected_senders == 0) + rrdpush_sender_thread_stop(host); + + // cleanup + fclose(fp); + + return (int)count; +} + +struct rrdpush_thread { + int fd; + char *key; + char *hostname; + char *registry_hostname; + char *machine_guid; + char *os; + char *timezone; + char *tags; + char *client_ip; + char *client_port; + char *program_name; + char *program_version; + int update_every; +}; + +static void rrdpush_receiver_thread_cleanup(void *ptr) { + static __thread int executed = 0; + if(!executed) { + executed = 1; + struct rrdpush_thread *rpt = (struct rrdpush_thread *) ptr; + + info("STREAM %s [receive from [%s]:%s]: receive thread ended (task id %d)", rpt->hostname, rpt->client_ip, rpt->client_port, gettid()); + + freez(rpt->key); + freez(rpt->hostname); + freez(rpt->registry_hostname); + freez(rpt->machine_guid); + freez(rpt->os); + freez(rpt->timezone); + freez(rpt->tags); + freez(rpt->client_ip); + freez(rpt->client_port); + freez(rpt->program_name); + freez(rpt->program_version); + freez(rpt); + } +} + +static void *rrdpush_receiver_thread(void *ptr) { + netdata_thread_cleanup_push(rrdpush_receiver_thread_cleanup, ptr); + + struct rrdpush_thread *rpt = (struct rrdpush_thread *)ptr; + info("STREAM %s [%s]:%s: receive thread created (task id %d)", rpt->hostname, rpt->client_ip, rpt->client_port, gettid()); + + rrdpush_receive( + rpt->fd + , rpt->key + , rpt->hostname + , rpt->registry_hostname + , rpt->machine_guid + , rpt->os + , rpt->timezone + , rpt->tags + , rpt->program_name + , rpt->program_version + , rpt->update_every + , rpt->client_ip + , rpt->client_port + ); + + netdata_thread_cleanup_pop(1); + return NULL; +} + +static void rrdpush_sender_thread_spawn(RRDHOST *host) { + rrdhost_wrlock(host); + + if(!host->rrdpush_sender_spawn) { + char tag[NETDATA_THREAD_TAG_MAX + 1]; + snprintfz(tag, NETDATA_THREAD_TAG_MAX, "STREAM_SENDER[%s]", host->hostname); + + if(netdata_thread_create(&host->rrdpush_sender_thread, tag, NETDATA_THREAD_OPTION_JOINABLE, rrdpush_sender_thread, (void *) host)) + error("STREAM %s [send]: failed to create new thread for client.", host->hostname); + else + host->rrdpush_sender_spawn = 1; + } + + rrdhost_unlock(host); +} + +int rrdpush_receiver_permission_denied(struct web_client *w) { + // we always respond with the same message and error code + // to prevent an attacker from gaining info about the error + buffer_flush(w->response.data); + buffer_sprintf(w->response.data, "You are not permitted to access this. Check the logs for more info."); + return 401; +} + +int rrdpush_receiver_too_busy_now(struct web_client *w) { + // we always respond with the same message and error code + // to prevent an attacker from gaining info about the error + buffer_flush(w->response.data); + buffer_sprintf(w->response.data, "The server is too busy now to accept this request. Try later."); + return 503; +} + +int rrdpush_receiver_thread_spawn(RRDHOST *host, struct web_client *w, char *url) { + (void)host; + + info("clients wants to STREAM metrics."); + + char *key = NULL, *hostname = NULL, *registry_hostname = NULL, *machine_guid = NULL, *os = "unknown", *timezone = "unknown", *tags = NULL; + int update_every = default_rrd_update_every; + char buf[GUID_LEN + 1]; + + while(url) { + char *value = mystrsep(&url, "&"); + if(!value || !*value) continue; + + char *name = mystrsep(&value, "="); + if(!name || !*name) continue; + if(!value || !*value) continue; + + if(!strcmp(name, "key")) + key = value; + else if(!strcmp(name, "hostname")) + hostname = value; + else if(!strcmp(name, "registry_hostname")) + registry_hostname = value; + else if(!strcmp(name, "machine_guid")) + machine_guid = value; + else if(!strcmp(name, "update_every")) + update_every = (int)strtoul(value, NULL, 0); + else if(!strcmp(name, "os")) + os = value; + else if(!strcmp(name, "timezone")) + timezone = value; + else if(!strcmp(name, "tags")) + tags = value; + else + info("STREAM [receive from [%s]:%s]: request has parameter '%s' = '%s', which is not used.", w->client_ip, w->client_port, key, value); + } + + if(!key || !*key) { + log_stream_connection(w->client_ip, w->client_port, (key && *key)?key:"-", (machine_guid && *machine_guid)?machine_guid:"-", (hostname && *hostname)?hostname:"-", "ACCESS DENIED - NO KEY"); + error("STREAM [receive from [%s]:%s]: request without an API key. Forbidding access.", w->client_ip, w->client_port); + return rrdpush_receiver_permission_denied(w); + } + + if(!hostname || !*hostname) { + log_stream_connection(w->client_ip, w->client_port, (key && *key)?key:"-", (machine_guid && *machine_guid)?machine_guid:"-", (hostname && *hostname)?hostname:"-", "ACCESS DENIED - NO HOSTNAME"); + error("STREAM [receive from [%s]:%s]: request without a hostname. Forbidding access.", w->client_ip, w->client_port); + return rrdpush_receiver_permission_denied(w); + } + + if(!machine_guid || !*machine_guid) { + log_stream_connection(w->client_ip, w->client_port, (key && *key)?key:"-", (machine_guid && *machine_guid)?machine_guid:"-", (hostname && *hostname)?hostname:"-", "ACCESS DENIED - NO MACHINE GUID"); + error("STREAM [receive from [%s]:%s]: request without a machine GUID. Forbidding access.", w->client_ip, w->client_port); + return rrdpush_receiver_permission_denied(w); + } + + if(regenerate_guid(key, buf) == -1) { + log_stream_connection(w->client_ip, w->client_port, (key && *key)?key:"-", (machine_guid && *machine_guid)?machine_guid:"-", (hostname && *hostname)?hostname:"-", "ACCESS DENIED - INVALID KEY"); + error("STREAM [receive from [%s]:%s]: API key '%s' is not valid GUID (use the command uuidgen to generate one). Forbidding access.", w->client_ip, w->client_port, key); + return rrdpush_receiver_permission_denied(w); + } + + if(regenerate_guid(machine_guid, buf) == -1) { + log_stream_connection(w->client_ip, w->client_port, (key && *key)?key:"-", (machine_guid && *machine_guid)?machine_guid:"-", (hostname && *hostname)?hostname:"-", "ACCESS DENIED - INVALID MACHINE GUID"); + error("STREAM [receive from [%s]:%s]: machine GUID '%s' is not GUID. Forbidding access.", w->client_ip, w->client_port, machine_guid); + return rrdpush_receiver_permission_denied(w); + } + + if(!appconfig_get_boolean(&stream_config, key, "enabled", 0)) { + log_stream_connection(w->client_ip, w->client_port, (key && *key)?key:"-", (machine_guid && *machine_guid)?machine_guid:"-", (hostname && *hostname)?hostname:"-", "ACCESS DENIED - KEY NOT ENABLED"); + error("STREAM [receive from [%s]:%s]: API key '%s' is not allowed. Forbidding access.", w->client_ip, w->client_port, key); + return rrdpush_receiver_permission_denied(w); + } + + { + SIMPLE_PATTERN *key_allow_from = simple_pattern_create(appconfig_get(&stream_config, key, "allow from", "*"), NULL, SIMPLE_PATTERN_EXACT); + if(key_allow_from) { + if(!simple_pattern_matches(key_allow_from, w->client_ip)) { + simple_pattern_free(key_allow_from); + log_stream_connection(w->client_ip, w->client_port, (key && *key)?key:"-", (machine_guid && *machine_guid)?machine_guid:"-", (hostname && *hostname) ? hostname : "-", "ACCESS DENIED - KEY NOT ALLOWED FROM THIS IP"); + error("STREAM [receive from [%s]:%s]: API key '%s' is not permitted from this IP. Forbidding access.", w->client_ip, w->client_port, key); + return rrdpush_receiver_permission_denied(w); + } + simple_pattern_free(key_allow_from); + } + } + + if(!appconfig_get_boolean(&stream_config, machine_guid, "enabled", 1)) { + log_stream_connection(w->client_ip, w->client_port, (key && *key)?key:"-", (machine_guid && *machine_guid)?machine_guid:"-", (hostname && *hostname)?hostname:"-", "ACCESS DENIED - MACHINE GUID NOT ENABLED"); + error("STREAM [receive from [%s]:%s]: machine GUID '%s' is not allowed. Forbidding access.", w->client_ip, w->client_port, machine_guid); + return rrdpush_receiver_permission_denied(w); + } + + { + SIMPLE_PATTERN *machine_allow_from = simple_pattern_create(appconfig_get(&stream_config, machine_guid, "allow from", "*"), NULL, SIMPLE_PATTERN_EXACT); + if(machine_allow_from) { + if(!simple_pattern_matches(machine_allow_from, w->client_ip)) { + simple_pattern_free(machine_allow_from); + log_stream_connection(w->client_ip, w->client_port, (key && *key)?key:"-", (machine_guid && *machine_guid)?machine_guid:"-", (hostname && *hostname) ? hostname : "-", "ACCESS DENIED - MACHINE GUID NOT ALLOWED FROM THIS IP"); + error("STREAM [receive from [%s]:%s]: Machine GUID '%s' is not permitted from this IP. Forbidding access.", w->client_ip, w->client_port, machine_guid); + return rrdpush_receiver_permission_denied(w); + } + simple_pattern_free(machine_allow_from); + } + } + + if(unlikely(web_client_streaming_rate_t > 0)) { + static netdata_mutex_t stream_rate_mutex = NETDATA_MUTEX_INITIALIZER; + static volatile time_t last_stream_accepted_t = 0; + + netdata_mutex_lock(&stream_rate_mutex); + time_t now = now_realtime_sec(); + + if(unlikely(last_stream_accepted_t == 0)) + last_stream_accepted_t = now; + + if(now - last_stream_accepted_t < web_client_streaming_rate_t) { + netdata_mutex_unlock(&stream_rate_mutex); + error("STREAM [receive from [%s]:%s]: too busy to accept new streaming request. Will be allowed in %ld secs.", w->client_ip, w->client_port, (long)(web_client_streaming_rate_t - (now - last_stream_accepted_t))); + return rrdpush_receiver_too_busy_now(w); + } + + last_stream_accepted_t = now; + netdata_mutex_unlock(&stream_rate_mutex); + } + + struct rrdpush_thread *rpt = callocz(1, sizeof(struct rrdpush_thread)); + rpt->fd = w->ifd; + rpt->key = strdupz(key); + rpt->hostname = strdupz(hostname); + rpt->registry_hostname = strdupz((registry_hostname && *registry_hostname)?registry_hostname:hostname); + rpt->machine_guid = strdupz(machine_guid); + rpt->os = strdupz(os); + rpt->timezone = strdupz(timezone); + rpt->tags = (tags)?strdupz(tags):NULL; + rpt->client_ip = strdupz(w->client_ip); + rpt->client_port = strdupz(w->client_port); + rpt->update_every = update_every; + + if(w->user_agent && w->user_agent[0]) { + char *t = strchr(w->user_agent, '/'); + if(t && *t) { + *t = '\0'; + t++; + } + + rpt->program_name = strdupz(w->user_agent); + if(t && *t) rpt->program_version = strdupz(t); + } + + netdata_thread_t thread; + + debug(D_SYSTEM, "starting STREAM receive thread."); + + char tag[FILENAME_MAX + 1]; + snprintfz(tag, FILENAME_MAX, "STREAM_RECEIVER[%s,[%s]:%s]", rpt->hostname, w->client_ip, w->client_port); + + if(netdata_thread_create(&thread, tag, NETDATA_THREAD_OPTION_DEFAULT, rrdpush_receiver_thread, (void *)rpt)) + error("Failed to create new STREAM receive thread for client."); + + // prevent the caller from closing the streaming socket + if(web_server_mode == WEB_SERVER_MODE_STATIC_THREADED) { + web_client_flag_set(w, WEB_CLIENT_FLAG_DONT_CLOSE_SOCKET); + } + else { + if(w->ifd == w->ofd) + w->ifd = w->ofd = -1; + else + w->ifd = -1; + } + + buffer_flush(w->response.data); + return 200; +} diff --git a/streaming/rrdpush.h b/streaming/rrdpush.h new file mode 100644 index 0000000..7bf3db9 --- /dev/null +++ b/streaming/rrdpush.h @@ -0,0 +1,25 @@ +// SPDX-License-Identifier: GPL-3.0-or-later + +#ifndef NETDATA_RRDPUSH_H +#define NETDATA_RRDPUSH_H 1 + +#include "web/server/web_client.h" +#include "daemon/common.h" + +extern unsigned int default_rrdpush_enabled; +extern char *default_rrdpush_destination; +extern char *default_rrdpush_api_key; +extern char *default_rrdpush_send_charts_matching; +extern unsigned int remote_clock_resync_iterations; + +extern int rrdpush_init(); +extern void rrdset_done_push(RRDSET *st); +extern void rrdset_push_chart_definition_now(RRDSET *st); +extern void *rrdpush_sender_thread(void *ptr); + +extern int rrdpush_receiver_thread_spawn(RRDHOST *host, struct web_client *w, char *url); +extern void rrdpush_sender_thread_stop(RRDHOST *host); + +extern void rrdpush_sender_send_this_host_variable_now(RRDHOST *host, RRDVAR *rv); + +#endif //NETDATA_RRDPUSH_H diff --git a/streaming/stream.conf b/streaming/stream.conf new file mode 100644 index 0000000..493eba3 --- /dev/null +++ b/streaming/stream.conf @@ -0,0 +1,191 @@ +# netdata configuration for aggregating data from remote hosts +# +# API keys authorize a pair of sending-receiving netdata servers. +# Once their communication is authorized, they can exchange metrics for any +# number of hosts. +# +# You can generate API keys, with the linux command: uuidgen + + +# ----------------------------------------------------------------------------- +# 1. ON SLAVE NETDATA - THE ONE THAT WILL BE SENDING METRICS + +[stream] + # Enable this on slaves, to have them send metrics. + enabled = no + + # Where is the receiving netdata? + # A space separated list of: + # + # [PROTOCOL:]HOST[%INTERFACE][:PORT] + # + # If many are given, the first available will get the metrics. + # + # PROTOCOL = tcp, udp, or unix (only tcp and unix are supported by masters) + # HOST = an IPv4, IPv6 IP, or a hostname, or a unix domain socket path. + # IPv6 IPs should be given with brackets [ip:address] + # INTERFACE = the network interface to use (only for IPv6) + # PORT = the port number or service name (/etc/services) + # + # This communication is not HTTP (it cannot be proxied by web proxies). + destination = + + # The API_KEY to use (as the sender) + api key = + + # The timeout to connect and send metrics + timeout seconds = 60 + + # If the destination line above does not specify a port, use this + default port = 19999 + + # filter the charts to be streamed + # netdata SIMPLE PATTERN: + # - space separated list of patterns (use \ to include spaces in patterns) + # - use * as wildcard, any number of times within each pattern + # - prefix a pattern with ! for a negative match (ie not stream the charts it matches) + # - the order of patterns is important (left to right) + # To send all except a few, use: !this !that * (ie append a wildcard pattern) + send charts matching = * + + # The buffer to use for sending metrics. + # 1MB is good for 10-20 seconds of data, so increase this if you expect latencies. + # The buffer is flushed on reconnects (this will not prevent gaps at the charts). + buffer size bytes = 1048576 + + # If the connection fails, or it disconnects, + # retry after that many seconds. + reconnect delay seconds = 5 + + # Sync the clock of the charts for that many iterations, when starting. + initial clock resync iterations = 60 + + +# ----------------------------------------------------------------------------- +# 2. ON MASTER NETDATA - THE ONE THAT WILL BE RECEIVING METRICS + +# You can have one API key per slave, +# or the same API key for all slaves. +# +# netdata searches for options in this order: +# +# a) master netdata settings (netdata.conf) +# b) [stream] section (above) +# c) [API_KEY] section (below, settings for the API key) +# d) [MACHINE_GUID] section (below, settings for each machine) +# +# You can combine the above (the more specific setting will be used). + +# API key authentication +# If the key is not listed here, it will not be able to push metrics. + +# [API_KEY] is [YOUR-API-KEY], i.e [11111111-2222-3333-4444-555555555555] +[API_KEY] + # Default settings for this API key + + # You can disable the API key, by setting this to: no + # The default (for unknown API keys) is: no + enabled = no + + # A list of simple patterns matching the IPs of the servers that + # will be pushing metrics using this API key. + # The metrics are received via the API port, so the same IPs + # should also be matched at netdata.conf [web].allow connections from + allow from = * + + # The default history in entries, for all hosts using this API key. + # You can also set it per host below. + # If you don't set it here, the history size of the central netdata + # will be used. + default history = 3600 + + # The default memory mode to be used for all hosts using this API key. + # You can also set it per host below. + # If you don't set it here, the memory mode of netdata.conf will be used. + # Valid modes: + # save save on exit, load on start + # map like swap (continuously syncing to disks - you need SSD) + # ram keep it in RAM, don't touch the disk + # none no database at all (use this on headless proxies) + default memory mode = ram + + # Shall we enable health monitoring for the hosts using this API key? + # 3 possible values: + # yes enable alarms + # no do not enable alarms + # auto enable alarms, only when the sending netdata is connected + # You can also set it per host, below. + # The default is taken from [health].enabled of netdata.conf + health enabled by default = auto + + # postpone alarms for a short period after the sender is connected + default postpone alarms on connect seconds = 60 + + # allow or deny multiple connections for the same host? + # If you are sure all your netdata have their own machine GUID, + # set this to 'allow', since it allows faster reconnects. + # When set to 'deny', new connections for a host will not be + # accepted until an existing connection is cleared. + multiple connections = allow + + # need to route metrics differently? set these. + # the defaults are the ones at the [stream] section (above) + #default proxy enabled = yes | no + #default proxy destination = IP:PORT IP:PORT ... + #default proxy api key = API_KEY + #default proxy send charts matching = * + + +# ----------------------------------------------------------------------------- +# 3. PER SENDING HOST SETTINGS, ON MASTER NETDATA +# THIS IS OPTIONAL - YOU DON'T HAVE TO CONFIGURE IT + +# This section exists to give you finer control of the master settings for each +# slave host, when the same API key is used by many netdata slaves / proxies. +# +# Each netdata has a unique GUID - generated the first time netdata starts. +# You can find it at /var/lib/netdata/registry/netdata.public.unique.id +# (at the slave). +# +# The host sending data will have one. If the host is not ephemeral, +# you can give settings for each sending host here. + +[MACHINE_GUID] + # enable this host: yes | no + # When disabled, the master will not receive metrics for this host. + # THIS IS NOT A SECURITY MECHANISM - AN ATTACKER CAN SET ANY OTHER GUID. + # Use only the API key for security. + enabled = no + + # A list of simple patterns matching the IPs of the servers that + # will be pushing metrics using this MACHINE GUID. + # The metrics are received via the API port, so the same IPs + # should also be matched at netdata.conf [web].allow connections from + # and at stream.conf [API_KEY].allow from + allow from = * + + # The number of entries in the database + history = 3600 + + # The memory mode of the database: save | map | ram | none + memory mode = save + + # Health / alarms control: yes | no | auto + health enabled = yes + + # postpone alarms when the sender connects + postpone alarms on connect seconds = 60 + + # allow or deny multiple connections for the same host? + # If you are sure all your netdata have their own machine GUID, + # set this to 'allow', since it allows faster reconnects. + # When set to 'deny', new connections for a host will not be + # accepted until an existing connection is cleared. + multiple connections = allow + + # need to route metrics differently? + # the defaults are the ones at the [API KEY] section + #proxy enabled = yes | no + #proxy destination = IP:PORT IP:PORT ... + #proxy api key = API_KEY + #proxy send charts matching = * |