Daemon Monitoring

Server daemon configuration (monitoring,log)

log-file
Item Value
Description Write messages to log file per default messages are output to stdout.
Type string
Example "/var/log/replication-manager.log"
log-file (2.0), logfile (0.6)
Item Value
Description Write messages to log file per default messages are output to stdout
Type string
Example /var/log/replication-manager.log
log-level
Item Value
Description Log verbosity level, og level >3 are very verbose and print the server monitoring workflow only to use in debugging
Type int
Example 1 to 7
log-syslog (2.0)
Item Value
Description duplicate stdout to local UDP syslog port
Type boolean
Example true
log-rotate-max-age (2.1)
Item Value
Description Rotate after this number of days
Type integer
Default 7
log-rotate-max-backup (2.1)
Item Value
Description Keep this number of log files
Type integer
Default 7
log-rotate-max-size (2.1)
Item Value
Description Rotate after this size in Mb
Type integer
Default 5
log-sql-in-monitoring (2.1)
Item Value
Description Log all SQL queries send to backend for monitoring servers
Type boolean
Default false

The sql_general.log can be found under each cluster directory and can be use to track what was trigger to backend during failover, rejoin step

verbose (2.1)
Item Value
Description Add more debugging information
Type int
Example 1
memprofile
Item Value
Description Write a memory profile to a file readable by pprof.
Type string
Default /tmp/repmgr.mprof
monitoring-datadir (2.0), working-directory (1.1)
Item Value
Description Path to write temporary and persistent files
Type string
Example /var/lib/replication-manager
monitoring-sharedir (2.0), share-directory (1.1)
Item Value
Description Path to the share files provided with packaging with no write access
Type string
Default Value /usr/share/replication-manager
monitoring-basedir (2.0)
Item Value
Description Path to a basedir where a data and share directory can be found, used mostly for developer or collocation of the product with a tar.gz deployment.
Type string
Default Value /usr/local/replication-manger
monitoring-save-config (2.1)
Item Value
Description Save configuration changes to /clusterd
Type bool
Default Value false
monitoring-ticker (2.0), read-timeout (1.1)
Item Value
Description Monitoring interval in seconds.
Type integer
Default Value 2
monitoring-write-heartbeat (2.0)
Item Value
Description Inject heartbeat into proxy or via external VIP. Without this option long inactive write period and database expire log days can make switchover to failed without founding last binlog position
Type bool
Default Value 2
monitoring-write-heartbeat-credential (2.0)
Item Value
Description Credential DB user to inject heartbeat into proxy or via external VIP.
Type bool
Default Value 2
title
Item Value
Description An explicit description of the managed cluster
Type string
Default Value CRM production MariaDB cluster
monitoring-schema-change (2.0)
Item Value
Description Monitor schema change
Type bool
Default Value true

This parameter is needed for alerting on schema change and for sharding proxy to push DDL change

monitoring-schema-change-script (2.1)
Item Value
Description Monitor schema change external script
Type string
Default Value ""
monitoring-performance-schema (2.1)
Item Value
Description Monitor performance schema
Type bool
Default Value true

This enable to get slow queries templating from performance schema

monitoring-processlist (2.1)
Item Value
Description Enable capture 50 longest process via processlist
Type bool
Default Value true
monitoring-queries (2.1)
Item Value
Description Monitor long queries
Type bool
Default Value true

Currently unused

monitoring-query-rules (2.1)
Item Value
Description Monitor query routing from proxies
Type bool
Default Value true

Enable to build a consolidated view of multiple cluster proxies query rules

monitoring-capture (2.1)
Item Value
Description Enable capture on internal error for 5 monitor loops
Type bool
Default Value true

The capture files are found in the cluster datadir

Capture are show slave status, show processlist, show global status, show innodb status

monitoring-capture-trigger (2.1)
Item Value
Description Enable capture on internal error for 5 monitor loops
Type string
Default Value "ERR00076,ERR00041"

Trigger to capture on max connections threshold and slave delay

monitoring-capture-file-keep (2.1)
Item Value
Description Purge capture file keep that number of them
Type integer
Default Value 5

Deprecated

daemon (2.0)
Item Value
Description Tell to monitor in post 2.0 release
Type integer
Default Value 1