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 /collectors/python.d.plugin/postgres/postgres.chart.py | |
parent | Initial commit. (diff) | |
download | netdata-upstream.tar.xz netdata-upstream.zip |
Adding upstream version 1.12.0.upstream/1.12.0upstream
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to '')
-rw-r--r-- | collectors/python.d.plugin/postgres/postgres.chart.py | 1092 |
1 files changed, 1092 insertions, 0 deletions
diff --git a/collectors/python.d.plugin/postgres/postgres.chart.py b/collectors/python.d.plugin/postgres/postgres.chart.py new file mode 100644 index 0000000..e988eec --- /dev/null +++ b/collectors/python.d.plugin/postgres/postgres.chart.py @@ -0,0 +1,1092 @@ +# -*- coding: utf-8 -*- +# Description: example netdata python.d module +# Authors: facetoe, dangtranhoang +# SPDX-License-Identifier: GPL-3.0-or-later + +from copy import deepcopy + +try: + import psycopg2 + from psycopg2 import extensions + from psycopg2.extras import DictCursor + from psycopg2 import OperationalError + PSYCOPG2 = True +except ImportError: + PSYCOPG2 = False + +from bases.FrameworkServices.SimpleService import SimpleService + + +DEFAULT_PORT = 5432 +DEFAULT_USER = 'postgres' +DEFAULT_CONNECT_TIMEOUT = 2 # seconds +DEFAULT_STATEMENT_TIMEOUT = 5000 # ms + + +WAL = 'WAL' +ARCHIVE = 'ARCHIVE' +BACKENDS = 'BACKENDS' +TABLE_STATS = 'TABLE_STATS' +INDEX_STATS = 'INDEX_STATS' +DATABASE = 'DATABASE' +BGWRITER = 'BGWRITER' +LOCKS = 'LOCKS' +DATABASES = 'DATABASES' +STANDBY = 'STANDBY' +REPLICATION_SLOT = 'REPLICATION_SLOT' +STANDBY_DELTA = 'STANDBY_DELTA' +REPSLOT_FILES = 'REPSLOT_FILES' +IF_SUPERUSER = 'IF_SUPERUSER' +SERVER_VERSION = 'SERVER_VERSION' +AUTOVACUUM = 'AUTOVACUUM' +DIFF_LSN = 'DIFF_LSN' +WAL_WRITES = 'WAL_WRITES' + +METRICS = { + DATABASE: [ + 'connections', + 'xact_commit', + 'xact_rollback', + 'blks_read', + 'blks_hit', + 'tup_returned', + 'tup_fetched', + 'tup_inserted', + 'tup_updated', + 'tup_deleted', + 'conflicts', + 'temp_files', + 'temp_bytes', + 'size' + ], + BACKENDS: [ + 'backends_active', + 'backends_idle' + ], + INDEX_STATS: [ + 'index_count', + 'index_size' + ], + TABLE_STATS: [ + 'table_size', + 'table_count' + ], + WAL: [ + 'written_wal', + 'recycled_wal', + 'total_wal' + ], + WAL_WRITES: [ + 'wal_writes' + ], + ARCHIVE: [ + 'ready_count', + 'done_count', + 'file_count' + ], + BGWRITER: [ + 'checkpoint_scheduled', + 'checkpoint_requested', + 'buffers_checkpoint', + 'buffers_clean', + 'maxwritten_clean', + 'buffers_backend', + 'buffers_alloc', + 'buffers_backend_fsync' + ], + LOCKS: [ + 'ExclusiveLock', + 'RowShareLock', + 'SIReadLock', + 'ShareUpdateExclusiveLock', + 'AccessExclusiveLock', + 'AccessShareLock', + 'ShareRowExclusiveLock', + 'ShareLock', + 'RowExclusiveLock' + ], + AUTOVACUUM: [ + 'analyze', + 'vacuum_analyze', + 'vacuum', + 'vacuum_freeze', + 'brin_summarize' + ], + STANDBY_DELTA: [ + 'sent_delta', + 'write_delta', + 'flush_delta', + 'replay_delta' + ], + REPSLOT_FILES: [ + 'replslot_wal_keep', + 'replslot_files' + ] +} + +NO_VERSION = 0 +DEFAULT = 'DEFAULT' +V96 = 'V96' +V10 = 'V10' +V11 = 'V11' + + +QUERY_WAL = { + DEFAULT: """ +SELECT + count(*) as total_wal, + count(*) FILTER (WHERE type = 'recycled') AS recycled_wal, + count(*) FILTER (WHERE type = 'written') AS written_wal +FROM + (SELECT + wal.name, + pg_walfile_name( + CASE pg_is_in_recovery() + WHEN true THEN NULL + ELSE pg_current_wal_lsn() + END ), + CASE + WHEN wal.name > pg_walfile_name( + CASE pg_is_in_recovery() + WHEN true THEN NULL + ELSE pg_current_wal_lsn() + END ) THEN 'recycled' + ELSE 'written' + END AS type + FROM pg_catalog.pg_ls_dir('pg_wal') AS wal(name) + WHERE name ~ '^[0-9A-F]{24}$' + ORDER BY + (pg_stat_file('pg_wal/'||name)).modification, + wal.name DESC) sub; +""", + V96: """ +SELECT + count(*) as total_wal, + count(*) FILTER (WHERE type = 'recycled') AS recycled_wal, + count(*) FILTER (WHERE type = 'written') AS written_wal +FROM + (SELECT + wal.name, + pg_xlogfile_name( + CASE pg_is_in_recovery() + WHEN true THEN NULL + ELSE pg_current_xlog_location() + END ), + CASE + WHEN wal.name > pg_xlogfile_name( + CASE pg_is_in_recovery() + WHEN true THEN NULL + ELSE pg_current_xlog_location() + END ) THEN 'recycled' + ELSE 'written' + END AS type + FROM pg_catalog.pg_ls_dir('pg_xlog') AS wal(name) + WHERE name ~ '^[0-9A-F]{24}$' + ORDER BY + (pg_stat_file('pg_xlog/'||name)).modification, + wal.name DESC) sub; +""", +} + +QUERY_ARCHIVE = { + DEFAULT: """ +SELECT + CAST(COUNT(*) AS INT) AS file_count, + CAST(COALESCE(SUM(CAST(archive_file ~ $r$\.ready$$r$ as INT)),0) AS INT) AS ready_count, + CAST(COALESCE(SUM(CAST(archive_file ~ $r$\.done$$r$ AS INT)),0) AS INT) AS done_count +FROM + pg_catalog.pg_ls_dir('pg_wal/archive_status') AS archive_files (archive_file); +""", + V96: """ +SELECT + CAST(COUNT(*) AS INT) AS file_count, + CAST(COALESCE(SUM(CAST(archive_file ~ $r$\.ready$$r$ as INT)),0) AS INT) AS ready_count, + CAST(COALESCE(SUM(CAST(archive_file ~ $r$\.done$$r$ AS INT)),0) AS INT) AS done_count +FROM + pg_catalog.pg_ls_dir('pg_xlog/archive_status') AS archive_files (archive_file); + +""", +} + +QUERY_BACKEND = { + DEFAULT: """ +SELECT + count(*) - (SELECT count(*) + FROM pg_stat_activity + WHERE state = 'idle') + AS backends_active, + (SELECT count(*) + FROM pg_stat_activity + WHERE state = 'idle') + AS backends_idle +FROM pg_stat_activity; +""", +} + +QUERY_TABLE_STATS = { + DEFAULT: """ +SELECT + ((sum(relpages) * 8) * 1024) AS table_size, + count(1) AS table_count +FROM pg_class +WHERE relkind IN ('r', 't'); +""", +} + +QUERY_INDEX_STATS = { + DEFAULT: """ +SELECT + ((sum(relpages) * 8) * 1024) AS index_size, + count(1) AS index_count +FROM pg_class +WHERE relkind = 'i'; +""", +} + +QUERY_DATABASE = { + DEFAULT: """ +SELECT + datname AS database_name, + numbackends AS connections, + xact_commit AS xact_commit, + xact_rollback AS xact_rollback, + blks_read AS blks_read, + blks_hit AS blks_hit, + tup_returned AS tup_returned, + tup_fetched AS tup_fetched, + tup_inserted AS tup_inserted, + tup_updated AS tup_updated, + tup_deleted AS tup_deleted, + conflicts AS conflicts, + pg_database_size(datname) AS size, + temp_files AS temp_files, + temp_bytes AS temp_bytes +FROM pg_stat_database +WHERE datname IN %(databases)s ; +""", +} + +QUERY_BGWRITER = { + DEFAULT: """ +SELECT + checkpoints_timed AS checkpoint_scheduled, + checkpoints_req AS checkpoint_requested, + buffers_checkpoint * current_setting('block_size')::numeric buffers_checkpoint, + buffers_clean * current_setting('block_size')::numeric buffers_clean, + maxwritten_clean, + buffers_backend * current_setting('block_size')::numeric buffers_backend, + buffers_alloc * current_setting('block_size')::numeric buffers_alloc, + buffers_backend_fsync +FROM pg_stat_bgwriter; +""", +} + +QUERY_LOCKS = { + DEFAULT: """ +SELECT + pg_database.datname as database_name, + mode, + count(mode) AS locks_count +FROM pg_locks +INNER JOIN pg_database + ON pg_database.oid = pg_locks.database +GROUP BY datname, mode +ORDER BY datname, mode; +""", +} + +QUERY_DATABASES = { + DEFAULT: """ +SELECT + datname +FROM pg_stat_database +WHERE + has_database_privilege( + (SELECT current_user), datname, 'connect') + AND NOT datname ~* '^template\d '; +""", +} + +QUERY_STANDBY = { + DEFAULT: """ +SELECT + application_name +FROM pg_stat_replication +WHERE application_name IS NOT NULL +GROUP BY application_name; +""", +} + +QUERY_REPLICATION_SLOT = { + DEFAULT: """ +SELECT slot_name +FROM pg_replication_slots; +""" +} + +QUERY_STANDBY_DELTA = { + DEFAULT: """ +SELECT + application_name, + pg_wal_lsn_diff( + CASE pg_is_in_recovery() + WHEN true THEN pg_last_wal_receive_lsn() + ELSE pg_current_wal_lsn() + END, + sent_lsn) AS sent_delta, + pg_wal_lsn_diff( + CASE pg_is_in_recovery() + WHEN true THEN pg_last_wal_receive_lsn() + ELSE pg_current_wal_lsn() + END, + write_lsn) AS write_delta, + pg_wal_lsn_diff( + CASE pg_is_in_recovery() + WHEN true THEN pg_last_wal_receive_lsn() + ELSE pg_current_wal_lsn() + END, + flush_lsn) AS flush_delta, + pg_wal_lsn_diff( + CASE pg_is_in_recovery() + WHEN true THEN pg_last_wal_receive_lsn() + ELSE pg_current_wal_lsn() + END, + replay_lsn) AS replay_delta +FROM pg_stat_replication +WHERE application_name IS NOT NULL; +""", + V96: """ +SELECT + application_name, + pg_xlog_location_diff( + CASE pg_is_in_recovery() + WHEN true THEN pg_last_xlog_receive_location() + ELSE pg_current_xlog_location() + END, + sent_location) AS sent_delta, + pg_xlog_location_diff( + CASE pg_is_in_recovery() + WHEN true THEN pg_last_xlog_receive_location() + ELSE pg_current_xlog_location() + END, + write_location) AS write_delta, + pg_xlog_location_diff( + CASE pg_is_in_recovery() + WHEN true THEN pg_last_xlog_receive_location() + ELSE pg_current_xlog_location() + END, + flush_location) AS flush_delta, + pg_xlog_location_diff( + CASE pg_is_in_recovery() + WHEN true THEN pg_last_xlog_receive_location() + ELSE pg_current_xlog_location() + END, + replay_location) AS replay_delta +FROM pg_stat_replication +WHERE application_name IS NOT NULL; +""", +} + +QUERY_REPSLOT_FILES = { + DEFAULT: """ +WITH wal_size AS ( + SELECT + setting::int AS val + FROM pg_settings + WHERE name = 'wal_segment_size' + ) +SELECT + slot_name, + slot_type, + replslot_wal_keep, + count(slot_file) AS replslot_files +FROM + (SELECT + slot.slot_name, + CASE + WHEN slot_file <> 'state' THEN 1 + END AS slot_file , + slot_type, + COALESCE ( + floor( + (pg_wal_lsn_diff(pg_current_wal_lsn (),slot.restart_lsn) + - (pg_walfile_name_offset (restart_lsn)).file_offset) / (s.val) + ),0) AS replslot_wal_keep + FROM pg_replication_slots slot + LEFT JOIN ( + SELECT + slot2.slot_name, + pg_ls_dir('pg_replslot/' || slot2.slot_name) AS slot_file + FROM pg_replication_slots slot2 + ) files (slot_name, slot_file) + ON slot.slot_name = files.slot_name + CROSS JOIN wal_size s + ) AS d +GROUP BY + slot_name, + slot_type, + replslot_wal_keep; +""", + V10: """ +WITH wal_size AS ( + SELECT + current_setting('wal_block_size')::INT * setting::INT AS val + FROM pg_settings + WHERE name = 'wal_segment_size' + ) +SELECT + slot_name, + slot_type, + replslot_wal_keep, + count(slot_file) AS replslot_files +FROM + (SELECT + slot.slot_name, + CASE + WHEN slot_file <> 'state' THEN 1 + END AS slot_file , + slot_type, + COALESCE ( + floor( + (pg_wal_lsn_diff(pg_current_wal_lsn (),slot.restart_lsn) + - (pg_walfile_name_offset (restart_lsn)).file_offset) / (s.val) + ),0) AS replslot_wal_keep + FROM pg_replication_slots slot + LEFT JOIN ( + SELECT + slot2.slot_name, + pg_ls_dir('pg_replslot/' || slot2.slot_name) AS slot_file + FROM pg_replication_slots slot2 + ) files (slot_name, slot_file) + ON slot.slot_name = files.slot_name + CROSS JOIN wal_size s + ) AS d +GROUP BY + slot_name, + slot_type, + replslot_wal_keep; +""", +} + +QUERY_SUPERUSER = { + DEFAULT: """ +SELECT current_setting('is_superuser') = 'on' AS is_superuser; +""", +} + +QUERY_SHOW_VERSION = { + DEFAULT: """ +SHOW server_version_num; +""", +} + +QUERY_AUTOVACUUM = { + DEFAULT: """ +SELECT + count(*) FILTER (WHERE query LIKE 'autovacuum: ANALYZE%%') AS analyze, + count(*) FILTER (WHERE query LIKE 'autovacuum: VACUUM ANALYZE%%') AS vacuum_analyze, + count(*) FILTER (WHERE query LIKE 'autovacuum: VACUUM%%' + AND query NOT LIKE 'autovacuum: VACUUM ANALYZE%%' + AND query NOT LIKE '%%to prevent wraparound%%') AS vacuum, + count(*) FILTER (WHERE query LIKE '%%to prevent wraparound%%') AS vacuum_freeze, + count(*) FILTER (WHERE query LIKE 'autovacuum: BRIN summarize%%') AS brin_summarize +FROM pg_stat_activity +WHERE query NOT LIKE '%%pg_stat_activity%%'; +""", +} + +QUERY_DIFF_LSN = { + DEFAULT: """ +SELECT + pg_wal_lsn_diff( + CASE pg_is_in_recovery() + WHEN true THEN pg_last_wal_receive_lsn() + ELSE pg_current_wal_lsn() + END, + '0/0') as wal_writes ; +""", + V96: """ +SELECT + pg_xlog_location_diff( + CASE pg_is_in_recovery() + WHEN true THEN pg_last_xlog_receive_location() + ELSE pg_current_xlog_location() + END, + '0/0') as wal_writes ; +""", +} + + +def query_factory(name, version=NO_VERSION): + if name == BACKENDS: + return QUERY_BACKEND[DEFAULT] + elif name == TABLE_STATS: + return QUERY_TABLE_STATS[DEFAULT] + elif name == INDEX_STATS: + return QUERY_INDEX_STATS[DEFAULT] + elif name == DATABASE: + return QUERY_DATABASE[DEFAULT] + elif name == BGWRITER: + return QUERY_BGWRITER[DEFAULT] + elif name == LOCKS: + return QUERY_LOCKS[DEFAULT] + elif name == DATABASES: + return QUERY_DATABASES[DEFAULT] + elif name == STANDBY: + return QUERY_STANDBY[DEFAULT] + elif name == REPLICATION_SLOT: + return QUERY_REPLICATION_SLOT[DEFAULT] + elif name == IF_SUPERUSER: + return QUERY_SUPERUSER[DEFAULT] + elif name == SERVER_VERSION: + return QUERY_SHOW_VERSION[DEFAULT] + elif name == AUTOVACUUM: + return QUERY_AUTOVACUUM[DEFAULT] + elif name == WAL: + if version < 100000: + return QUERY_WAL[V96] + return QUERY_WAL[DEFAULT] + elif name == ARCHIVE: + if version < 100000: + return QUERY_ARCHIVE[V96] + return QUERY_ARCHIVE[DEFAULT] + elif name == STANDBY_DELTA: + if version < 100000: + return QUERY_STANDBY_DELTA[V96] + return QUERY_STANDBY_DELTA[DEFAULT] + elif name == REPSLOT_FILES: + if version < 110000: + return QUERY_REPSLOT_FILES[V10] + return QUERY_REPSLOT_FILES[DEFAULT] + elif name == DIFF_LSN: + if version < 100000: + return QUERY_DIFF_LSN[V96] + return QUERY_DIFF_LSN[DEFAULT] + + raise ValueError('unknown query') + + +ORDER = [ + 'db_stat_temp_files', + 'db_stat_temp_bytes', + 'db_stat_blks', + 'db_stat_tuple_returned', + 'db_stat_tuple_write', + 'db_stat_transactions', + 'db_stat_connections', + 'database_size', + 'backend_process', + 'index_count', + 'index_size', + 'table_count', + 'table_size', + 'wal', + 'wal_writes', + 'archive_wal', + 'checkpointer', + 'stat_bgwriter_alloc', + 'stat_bgwriter_checkpoint', + 'stat_bgwriter_backend', + 'stat_bgwriter_backend_fsync', + 'stat_bgwriter_bgwriter', + 'stat_bgwriter_maxwritten', + 'replication_slot', + 'standby_delta', + 'autovacuum' +] + +CHARTS = { + 'db_stat_transactions': { + 'options': [None, 'Transactions on db', 'transactions/s', 'db statistics', 'postgres.db_stat_transactions', + 'line'], + 'lines': [ + ['xact_commit', 'committed', 'incremental'], + ['xact_rollback', 'rolled back', 'incremental'] + ] + }, + 'db_stat_connections': { + 'options': [None, 'Current connections to db', 'count', 'db statistics', 'postgres.db_stat_connections', + 'line'], + 'lines': [ + ['connections', 'connections', 'absolute'] + ] + }, + 'db_stat_blks': { + 'options': [None, 'Disk blocks reads from db', 'reads/s', 'db statistics', 'postgres.db_stat_blks', 'line'], + 'lines': [ + ['blks_read', 'disk', 'incremental'], + ['blks_hit', 'cache', 'incremental'] + ] + }, + 'db_stat_tuple_returned': { + 'options': [None, 'Tuples returned from db', 'tuples/s', 'db statistics', 'postgres.db_stat_tuple_returned', + 'line'], + 'lines': [ + ['tup_returned', 'sequential', 'incremental'], + ['tup_fetched', 'bitmap', 'incremental'] + ] + }, + 'db_stat_tuple_write': { + 'options': [None, 'Tuples written to db', 'writes/s', 'db statistics', 'postgres.db_stat_tuple_write', 'line'], + 'lines': [ + ['tup_inserted', 'inserted', 'incremental'], + ['tup_updated', 'updated', 'incremental'], + ['tup_deleted', 'deleted', 'incremental'], + ['conflicts', 'conflicts', 'incremental'] + ] + }, + 'db_stat_temp_bytes': { + 'options': [None, 'Temp files written to disk', 'KiB/s', 'db statistics', 'postgres.db_stat_temp_bytes', + 'line'], + 'lines': [ + ['temp_bytes', 'size', 'incremental', 1, 1024] + ] + }, + 'db_stat_temp_files': { + 'options': [None, 'Temp files written to disk', 'files', 'db statistics', 'postgres.db_stat_temp_files', + 'line'], + 'lines': [ + ['temp_files', 'files', 'incremental'] + ] + }, + 'database_size': { + 'options': [None, 'Database size', 'MiB', 'database size', 'postgres.db_size', 'stacked'], + 'lines': [ + ] + }, + 'backend_process': { + 'options': [None, 'Current Backend Processes', 'processes', 'backend processes', 'postgres.backend_process', + 'line'], + 'lines': [ + ['backends_active', 'active', 'absolute'], + ['backends_idle', 'idle', 'absolute'] + ] + }, + 'index_count': { + 'options': [None, 'Total indexes', 'index', 'indexes', 'postgres.index_count', 'line'], + 'lines': [ + ['index_count', 'total', 'absolute'] + ] + }, + 'index_size': { + 'options': [None, 'Indexes size', 'MiB', 'indexes', 'postgres.index_size', 'line'], + 'lines': [ + ['index_size', 'size', 'absolute', 1, 1024 * 1024] + ] + }, + 'table_count': { + 'options': [None, 'Total Tables', 'tables', 'tables', 'postgres.table_count', 'line'], + 'lines': [ + ['table_count', 'total', 'absolute'] + ] + }, + 'table_size': { + 'options': [None, 'Tables size', 'MiB', 'tables', 'postgres.table_size', 'line'], + 'lines': [ + ['table_size', 'size', 'absolute', 1, 1024 * 1024] + ] + }, + 'wal': { + 'options': [None, 'Write-Ahead Logs', 'files', 'wal', 'postgres.wal', 'line'], + 'lines': [ + ['written_wal', 'written', 'absolute'], + ['recycled_wal', 'recycled', 'absolute'], + ['total_wal', 'total', 'absolute'] + ] + }, + 'wal_writes': { + 'options': [None, 'Write-Ahead Logs', 'KiB/s', 'wal_writes', 'postgres.wal_writes', 'line'], + 'lines': [ + ['wal_writes', 'writes', 'incremental', 1, 1024] + ] + }, + 'archive_wal': { + 'options': [None, 'Archive Write-Ahead Logs', 'files/s', 'archive wal', 'postgres.archive_wal', 'line'], + 'lines': [ + ['file_count', 'total', 'incremental'], + ['ready_count', 'ready', 'incremental'], + ['done_count', 'done', 'incremental'] + ] + }, + 'checkpointer': { + 'options': [None, 'Checkpoints', 'writes', 'checkpointer', 'postgres.checkpointer', 'line'], + 'lines': [ + ['checkpoint_scheduled', 'scheduled', 'incremental'], + ['checkpoint_requested', 'requested', 'incremental'] + ] + }, + 'stat_bgwriter_alloc': { + 'options': [None, 'Buffers allocated', 'KiB/s', 'bgwriter', 'postgres.stat_bgwriter_alloc', 'line'], + 'lines': [ + ['buffers_alloc', 'alloc', 'incremental', 1, 1024] + ] + }, + 'stat_bgwriter_checkpoint': { + 'options': [None, 'Buffers written during checkpoints', 'KiB/s', 'bgwriter', + 'postgres.stat_bgwriter_checkpoint', 'line'], + 'lines': [ + ['buffers_checkpoint', 'checkpoint', 'incremental', 1, 1024] + ] + }, + 'stat_bgwriter_backend': { + 'options': [None, 'Buffers written directly by a backend', 'KiB/s', 'bgwriter', + 'postgres.stat_bgwriter_backend', 'line'], + 'lines': [ + ['buffers_backend', 'backend', 'incremental', 1, 1024] + ] + }, + 'stat_bgwriter_backend_fsync': { + 'options': [None, 'Fsync by backend', 'times', 'bgwriter', 'postgres.stat_bgwriter_backend_fsync', 'line'], + 'lines': [ + ['buffers_backend_fsync', 'backend fsync', 'incremental'] + ] + }, + 'stat_bgwriter_bgwriter': { + 'options': [None, 'Buffers written by the background writer', 'KiB/s', 'bgwriter', + 'postgres.bgwriter_bgwriter', 'line'], + 'lines': [ + ['buffers_clean', 'clean', 'incremental', 1, 1024] + ] + }, + 'stat_bgwriter_maxwritten': { + 'options': [None, 'Too many buffers written', 'times', 'bgwriter', 'postgres.stat_bgwriter_maxwritten', + 'line'], + 'lines': [ + ['maxwritten_clean', 'maxwritten', 'incremental'] + ] + }, + 'autovacuum': { + 'options': [None, 'Autovacuum workers', 'workers', 'autovacuum', 'postgres.autovacuum', 'line'], + 'lines': [ + ['analyze', 'analyze', 'absolute'], + ['vacuum', 'vacuum', 'absolute'], + ['vacuum_analyze', 'vacuum analyze', 'absolute'], + ['vacuum_freeze', 'vacuum freeze', 'absolute'], + ['brin_summarize', 'brin summarize', 'absolute'] + ] + }, + 'standby_delta': { + 'options': [None, 'Standby delta', 'KiB', 'replication delta', 'postgres.standby_delta', 'line'], + 'lines': [ + ['sent_delta', 'sent delta', 'absolute', 1, 1024], + ['write_delta', 'write delta', 'absolute', 1, 1024], + ['flush_delta', 'flush delta', 'absolute', 1, 1024], + ['replay_delta', 'replay delta', 'absolute', 1, 1024] + ] + }, + 'replication_slot': { + 'options': [None, 'Replication slot files', 'files', 'replication slot', 'postgres.replication_slot', 'line'], + 'lines': [ + ['replslot_wal_keep', 'wal keeped', 'absolute'], + ['replslot_files', 'pg_replslot files', 'absolute'] + ] + } +} + + +class Service(SimpleService): + def __init__(self, configuration=None, name=None): + SimpleService.__init__(self, configuration=configuration, name=name) + self.order = list(ORDER) + self.definitions = deepcopy(CHARTS) + self.do_table_stats = configuration.pop('table_stats', False) + self.do_index_stats = configuration.pop('index_stats', False) + self.databases_to_poll = configuration.pop('database_poll', None) + self.statement_timeout = configuration.pop('statement_timeout', DEFAULT_STATEMENT_TIMEOUT) + self.configuration = configuration + self.conn = None + self.server_version = None + self.is_superuser = False + self.alive = False + self.databases = list() + self.secondaries = list() + self.replication_slots = list() + self.queries = dict() + self.data = dict() + + def reconnect(self): + return self.connect() + + def connect(self): + if self.conn: + self.conn.close() + self.conn = None + + try: + params = dict( + host=None, + port=DEFAULT_PORT, + database=None, + user=DEFAULT_USER, + password=None, + connect_timeout=DEFAULT_CONNECT_TIMEOUT, + options='-c statement_timeout={0}'.format(self.statement_timeout), + ) + params.update(self.configuration) + + self.conn = psycopg2.connect(**params) + self.conn.set_isolation_level(extensions.ISOLATION_LEVEL_AUTOCOMMIT) + self.conn.set_session(readonly=True) + except OperationalError as error: + self.error(error) + self.alive = False + else: + self.alive = True + + return self.alive + + def check(self): + if not PSYCOPG2: + self.error("'python-psycopg2' package is needed to use postgres module") + return False + + if not self.connect(): + self.error('failed to connect to {0}'.format(hide_password(self.configuration))) + return False + + try: + self.check_queries() + except Exception as error: + self.error(error) + return False + + self.populate_queries() + self.create_dynamic_charts() + + return True + + def get_data(self): + if not self.alive and not self.reconnect(): + return None + + try: + cursor = self.conn.cursor(cursor_factory=DictCursor) + + self.data.update(zero_lock_types(self.databases)) + + for query, metrics in self.queries.items(): + self.query_stats(cursor, query, metrics) + + except OperationalError: + self.alive = False + return None + + cursor.close() + + return self.data + + def query_stats(self, cursor, query, metrics): + cursor.execute(query, dict(databases=tuple(self.databases))) + + for row in cursor: + for metric in metrics: + # databases + if 'database_name' in row: + dimension_id = '_'.join([row['database_name'], metric]) + # secondaries + elif 'application_name' in row: + dimension_id = '_'.join([row['application_name'], metric]) + # replication slots + elif 'slot_name' in row: + dimension_id = '_'.join([row['slot_name'], metric]) + # other + else: + dimension_id = metric + + if metric in row: + if row[metric] is not None: + self.data[dimension_id] = int(row[metric]) + elif 'locks_count' in row: + if metric == row['mode']: + self.data[dimension_id] = row['locks_count'] + + def check_queries(self): + cursor = self.conn.cursor() + + self.server_version = detect_server_version(cursor, query_factory(SERVER_VERSION)) + self.debug('server version: {0}'.format(self.server_version)) + + self.is_superuser = check_if_superuser(cursor, query_factory(IF_SUPERUSER)) + self.debug('superuser: {0}'.format(self.is_superuser)) + + self.databases = discover(cursor, query_factory(DATABASES)) + self.debug('discovered databases {0}'.format(self.databases)) + if self.databases_to_poll: + to_poll = self.databases_to_poll.split() + self.databases = [db for db in self.databases if db in to_poll] or self.databases + + self.secondaries = discover(cursor, query_factory(STANDBY)) + self.debug('discovered secondaries: {0}'.format(self.secondaries)) + + if self.server_version >= 94000: + self.replication_slots = discover(cursor, query_factory(REPLICATION_SLOT)) + self.debug('discovered replication slots: {0}'.format(self.replication_slots)) + + cursor.close() + + def populate_queries(self): + self.queries[query_factory(DATABASE)] = METRICS[DATABASE] + self.queries[query_factory(BACKENDS)] = METRICS[BACKENDS] + self.queries[query_factory(LOCKS)] = METRICS[LOCKS] + self.queries[query_factory(BGWRITER)] = METRICS[BGWRITER] + self.queries[query_factory(DIFF_LSN, self.server_version)] = METRICS[WAL_WRITES] + self.queries[query_factory(STANDBY_DELTA, self.server_version)] = METRICS[STANDBY_DELTA] + + if self.do_index_stats: + self.queries[query_factory(INDEX_STATS)] = METRICS[INDEX_STATS] + if self.do_table_stats: + self.queries[query_factory(TABLE_STATS)] = METRICS[TABLE_STATS] + + if self.is_superuser: + self.queries[query_factory(ARCHIVE, self.server_version)] = METRICS[ARCHIVE] + + if self.server_version >= 90400: + self.queries[query_factory(WAL, self.server_version)] = METRICS[WAL] + + if self.server_version >= 100000: + self.queries[query_factory(REPSLOT_FILES, self.server_version)] = METRICS[REPSLOT_FILES] + + if self.server_version >= 90400: + self.queries[query_factory(AUTOVACUUM)] = METRICS[AUTOVACUUM] + + def create_dynamic_charts(self): + for database_name in self.databases[::-1]: + dim = [ + database_name + '_size', + database_name, + 'absolute', + 1, + 1024 * 1024, + ] + self.definitions['database_size']['lines'].append(dim) + for chart_name in [name for name in self.order if name.startswith('db_stat')]: + add_database_stat_chart( + order=self.order, + definitions=self.definitions, + name=chart_name, + database_name=database_name, + ) + add_database_lock_chart( + order=self.order, + definitions=self.definitions, + database_name=database_name, + ) + + for application_name in self.secondaries[::-1]: + add_replication_delta_chart( + order=self.order, + definitions=self.definitions, + name='standby_delta', + application_name=application_name, + ) + + for slot_name in self.replication_slots[::-1]: + add_replication_slot_chart( + order=self.order, + definitions=self.definitions, + name='replication_slot', + slot_name=slot_name, + ) + + +def discover(cursor, query): + cursor.execute(query) + result = list() + for v in [value[0] for value in cursor]: + if v not in result: + result.append(v) + return result + + +def check_if_superuser(cursor, query): + cursor.execute(query) + return cursor.fetchone()[0] + + +def detect_server_version(cursor, query): + cursor.execute(query) + return int(cursor.fetchone()[0]) + + +def zero_lock_types(databases): + result = dict() + for database in databases: + for lock_type in METRICS['LOCKS']: + key = '_'.join([database, lock_type]) + result[key] = 0 + + return result + + +def hide_password(config): + return dict((k, v if k != 'password' else '*****') for k, v in config.items()) + + +def add_database_lock_chart(order, definitions, database_name): + def create_lines(database): + result = list() + for lock_type in METRICS['LOCKS']: + dimension_id = '_'.join([database, lock_type]) + result.append([dimension_id, lock_type, 'absolute']) + return result + + chart_name = database_name + '_locks' + order.insert(-1, chart_name) + definitions[chart_name] = { + 'options': + [None, 'Locks on db: ' + database_name, 'locks', 'db ' + database_name, 'postgres.db_locks', 'line'], + 'lines': create_lines(database_name) + } + + +def add_database_stat_chart(order, definitions, name, database_name): + def create_lines(database, lines): + result = list() + for line in lines: + new_line = ['_'.join([database, line[0]])] + line[1:] + result.append(new_line) + return result + + chart_template = CHARTS[name] + chart_name = '_'.join([database_name, name]) + order.insert(0, chart_name) + name, title, units, _, context, chart_type = chart_template['options'] + definitions[chart_name] = { + 'options': [name, title + ': ' + database_name, units, 'db ' + database_name, context, chart_type], + 'lines': create_lines(database_name, chart_template['lines'])} + + +def add_replication_delta_chart(order, definitions, name, application_name): + def create_lines(standby, lines): + result = list() + for line in lines: + new_line = ['_'.join([standby, line[0]])] + line[1:] + result.append(new_line) + return result + + chart_template = CHARTS[name] + chart_name = '_'.join([application_name, name]) + position = order.index('database_size') + order.insert(position, chart_name) + name, title, units, _, context, chart_type = chart_template['options'] + definitions[chart_name] = { + 'options': [name, title + ': ' + application_name, units, 'replication delta', context, chart_type], + 'lines': create_lines(application_name, chart_template['lines'])} + + +def add_replication_slot_chart(order, definitions, name, slot_name): + def create_lines(slot, lines): + result = list() + for line in lines: + new_line = ['_'.join([slot, line[0]])] + line[1:] + result.append(new_line) + return result + + chart_template = CHARTS[name] + chart_name = '_'.join([slot_name, name]) + position = order.index('database_size') + order.insert(position, chart_name) + name, title, units, _, context, chart_type = chart_template['options'] + definitions[chart_name] = { + 'options': [name, title + ': ' + slot_name, units, 'replication slot files', context, chart_type], + 'lines': create_lines(slot_name, chart_template['lines'])} |