Icinga 2 v2.11.7

Issues and PRs
Blogpost

Version 2.11.7 fixes several issues to improve the reliability of the cluster functionality.

Bugfixes

• Fix a connection leak with misconfigured agents #8482
• Properly sync changes of config objects in global zones done via the API #8473 #8457
• Prevent other clients from being disconnected when replaying the cluster log takes very long #8475
• Avoid duplicate connections between endpoints #8399
• Ignore incoming config object updates for unknown zones #8459
• Check timestamps before removing files in config sync #8486

Enhancements

• Include HTTP status codes in log #8454

Icinga 2 v2.12.1

Issues and PRs
Blogpost

This version fixes several crashes, deadlocks and excessive check latencies. It also addresses several bugs regarding IDO, API, notifications and checks.

Bugfixes

• Core
  • Fix crashes during config update #8348 #8345
  • Fix crash while removing a downtime #8228
  • Ensure the daemon doesn't get killed by logrotate #8170
  • Fix hangup during shutdown #8211
  • Fix a deadlock in Icinga DB #8168
  • Clean up zombie processes during reload #8376
  • Reduce check latency #8276
• IDO
  • Prevent unnecessary IDO updates #8327 #8320
  • Commit IDO MySQL transactions earlier #8349
  • Make sure to insert IDO program status #8330
  • Improve IDO queue stats logging #8271 #8328 #8379
• Misc
  • Ensure API connections are closed properly #8293
  • Prevent unnecessary notifications #8299
  • Don't skip null values of command arguments #8174
  • Fix Windows .exe version #8234
  • Reset Icinga check warning after successful config update #8189

Icinga 2.11.6

Issues and PRs
Blogpost

Version 2.11.6 fixes several crashes, prevents unnecessary notifications and addresses several bugs in IDO and the API.

Bugfixes

• Crashes
  • Fix crashes during config update #8337 #8308
  • Fix crash while removing a downtime #8226
  • Ensure the daemon doesn't get killed by logrotate #8227
• IDO
  • Prevent unnecessary IDO updates #8316 #8305
  • Commit IDO MySQL transactions earlier #8298
  • Make sure to insert IDO program status #8291
  • Improve IDO queue stats logging #8270 #8325 #8378
• API
  • Ensure API connections are closed properly #8292
  • Fix open connections when agent waits for CA approval #8230
  • Close connections without successful TLS handshakes within 10s #8224
• Misc
  • Prevent unnecessary notifications #8300
  • Fix Windows .exe version #8235
  • Reset Icinga check warning after successful config update #8225

Icinga 2.12.0

Issues and PRs
Blogpost
Upgrading docs

Thanks to all contributors:
Ant1x, azthec, baurmatt, bootc, Foxeronie, ggzengel, islander, joni1993, KAMI911, mcktr, MichalMMac, sebastic, sthen, unki, vigiroux, wopfel

Breaking changes

• Deprecate Windows plugins in favor of our
  PowerShell plugins #8071
• Deprecate Livestatus #8051
• Refuse acknowledging an already acknowledged checkable #7695
• Config lexer: complain on EOF in heredocs, i.e. {{{abc<EOF> #7541

Enhancements

• Core
  • Implement new database backend: Icinga DB #7571
  • Re-send notifications previously suppressed by their time periods #7816
• API
  • Host/Service: Add acknowledgement_last_change and next_update attributes #7881 #7534
  • Improve error message for POST queries #7681
  • /v1/actions/remove-comment: let users specify themselves #7646
  • /v1/actions/remove-downtime: let users specify themselves #7645
  • /v1/config/stages: Add 'activate' parameter #7535
• CLI
  • Add pki verify command for better TLS certificate troubleshooting #7843
  • Add OpenSSL version to 'Build' section in --version #7833
  • Improve experience with 'Node Setup for Agents/Satellite' #7835
• DSL
  • Add get_template() and get_templates() #7632
  • MacroProcessor::ResolveArguments(): skip null argument values #7567
  • Fix crash due to dependency apply rule with ignore_on_error and non-existing parent #7538
  • Introduce ternary operator (x ? y : z) #7442
  • LegacyTimePeriod: support specifying seconds #7439
  • Add support for Lambda Closures (() use(x) => x and () use(x) => { return x }) #7417
• ITL
  • Add notemp parameter to oracle health #7748
  • Add extended checks options to snmp-interface command template #7602
  • Add file age check for Windows command definition #7540
• Docs
  • Development: Update debugging instructions #7867
  • Add new API clients #7859
  • Clarify CRITICAL vs. UNKNOWN #7665
  • Explicitly explain how to disable freshness checks #7664
  • Update installation for RHEL/CentOS 8 and SLES 15 #7640
  • Add Powershell example to validate the certificate #7603
• Misc
  • Don't send event::Heartbeat to unauthenticated peers #7747
  • OpenTsdbWriter: Add custom tag support #7357

Bugfixes

• Core
  • Fix JSON-RPC crashes #7532 #7737
  • Fix zone definitions in zones #7546
  • Fix deadlock during start on OpenBSD #7739
  • Consider PENDING not a problem #7685
  • Fix zombie processes after reload #7606
  • Don't wait for checks to finish during reload #7894
• Cluster
  • Fix segfault during heartbeat timeout with clients not yet signed #7970
  • Make the config update process mutually exclusive (Prevents file system race conditions) #7936
  • Fix check_timeout not being forwarded to agent command endpoints #7861
  • Config sync: Use a more friendly message when configs are equal and don't need a reload #7811
  • Fix open connections when agent waits for CA approval #7686
  • Consider a JsonRpcConnection alive on a single byte of TLS payload, not only on a whole message #7836
  • Send JsonRpcConnection heartbeat every 20s instead of 10s #8102
  • Use JsonRpcConnection heartbeat only to update connection liveness (m_Seen) #8142
  • Fix TLS context not being updated on signed certificate messages on agents #7654
• API
  • Close connections w/o successful TLS handshakes after 10s #7809
  • Handle permission exceptions soon enough, returning 404 #7528
• SELinux
  • Fix safe-reload #7858
  • Allow direct SMTP notifications #7749
• Windows
  • Terminate check processes with UNKNOWN state on timeout #7788
  • Ensure that log replay files are properly renamed #7767
• Metrics
  • Graphite/OpenTSDB: Ensure that reconnect failure is detected #7765
  • Always send 0 as value for thresholds #7696
• Scripts
  • Fix notification scripts to stay compatible with Dash #7706
  • Fix bash line continuation in mail-host-notification.sh #7701
  • Fix notification scripts string comparison #7647
  • Service and host mail-notifications: Add line-breaks to very long output #6822
  • Set correct UTF-8 email subject header (RFC1342) #6369
• Misc
  • DSL: Fix segfault due to passing null as custom function to Array#{sort,map,reduce,filter,any,all}() #8053
  • CLI: pki save-cert: allow to specify --key and --cert for backwards compatibility #7995
  • Catch exception when trusted cert is not readable during node setup on agent/satellite #7838
  • CheckCommand ssl: Fix wrong parameter -N #7741
  • Code quality fixes
  • Small documentation fixes

v2.11.5

Issues and PRs
Blogpost

Version 2.11.5 fixes file system race conditions
in the config update process occurring in large HA environments
and improves the cluster connection liveness mechanisms.

Bugfixes

• Make the config update process mutually exclusive (Prevents file system race conditions) #8093
• Consider a JsonRpcConnection alive on a single byte of TLS payload, not only on a whole message #8094
• Send JsonRpcConnection heartbeat every 20s instead of 10s #8103
• Use JsonRpcConnection heartbeat only to update connection liveness (m_Seen) #8097

v2.11.4

Issues and PRs
Blogpost

Version 2.11.4 fixes a crash during a heartbeat timeout with clients not yet signed. It also resolves
an issue with endpoints not reconnecting after a reload/deploy, which caused a lot of UNKNOWN states.

Bugfixes

• Cluster
  • Fix segfault during heartbeat timeout with clients not yet signed #7997
  • Fix endpoints not reconnecting after reload (UNKNOWN hosts/services after reload) #8043
• Setup
  • Fix exception on trusted cert not readable during node setup #8044
  • prepare-dirs: Only set permissions during directory creation #8046
• DSL
  • Fix segfault on missing compare function in Array functions (sort, map, reduce, filter, any, all) #8054

v2.12.0-rc1

Changes

Issue and PRs

Notes

Release blogpost: https://icinga.com/2020/03/16/releasing-icinga-db-v1-0-rc1/
Upgrading docs: https://icinga.com/docs/icinga2/snapshot/doc/16-upgrading-icinga-2/#upgrading-to-v212

Thanks to all contributors: Ant1x, azthec, baurmatt, bootc, Foxeronie, ggzengel, islander, joni1993, KAMI911, mcktr, MichalMMac, sebastic, sthen, unki, vigiroux, wopfel,

Breaking changes

• Refuse acknowledging an already acknowledged checkable #7695
• Config lexer: complain on EOF in heredocs, i.e. {{{abc<EOF> #7541

Enhancements

• Core
  • Implement new database backend: Icinga DB #7571
• API
  • Host/Service: Add acknowledgement_last_change and next_update attributes #7881 #7534
  • Improve error message for POST queries #7681
  • /v1/actions/remove-comment: let users specify themselves #7646
  • /v1/actions/remove-downtime: let users specify themselves #7645
  • /v1/config/stages: Add 'activate' parameter #7535
• CLI
  • Add pki verify command for better TLS certificate troubleshooting #7843
  • Add OpenSSL version to 'Build' section in --version #7833
  • Improve experience with 'Node Setup for Agents/Satellite' #7835
• DSL
  • Add get_template() and get_templates() #7632
  • MacroProcessor::ResolveArguments(): skip null argument values #7567
  • Fix crash due to dependency apply rule with ignore_on_error and non-existing parent #7538
  • Introduce ternary operator (x ? y : z) #7442
  • LegacyTimePeriod: support specifying seconds #7439
  • Add support for Lambda Closures (() use(x) => x and () use(x) => { return x }) #7417
• ITL
  • Add notemp parameter to oracle health #7748
  • Add extended checks options to snmp-interface command template #7602
  • Add file age check for Windows command definition #7540
• Docs
  • Development: Update debugging instructions #7867
  • Add new API clients #7859
  • Clarify CRITICAL vs. UNKNOWN #7665
  • Explicitly explain how to disable freshness checks #7664
  • Update installation for RHEL/CentOS 8 and SLES 15 #7640
  • Add Powershell example to validate the certificate #7603
• Misc
  • Don't send event::Heartbeat to unauthenticated peers #7747
  • OpenTsdbWriter: Add custom tag support #7357

Bugfixes

• Core
  • Fix JSON-RPC crashes #7532 #7737
  • Fix zone definitions in zones #7546
  • Fix deadlock during start on OpenBSD #7739
  • Consider PENDING not a problem #7685
  • Fix zombie processes after reload #7606
• Cluster
  • Fix check_timeout not being forwarded to agent command endpoints #7861
  • Config sync: Use a more friendly message when configs are equal and don't need a reload #7811
  • Fix open connections when agent waits for CA approval #7686
  • Fix TLS context not being updated on signed certificate messages on agents #7654
• API
  • Close connections w/o successful TLS handshakes after 10s #7809
  • Handle permission exceptions soon enough, returning 404 #7528
• SELinux
  • Fix safe-reload #7858
  • Allow direct SMTP notifications #7749
• Windows
  • Terminate check processes with UNKNOWN state on timeout #7788
  • Ensure that log replay files are properly renamed #7767
• Metrics
  • Graphite/OpenTSDB: Ensure that reconnect failure is detected #7765
  • Always send 0 as value for thresholds #7696
• Scripts
  • Fix notification scripts to stay compatible with Dash #7706
  • Fix bash line continuation in mail-host-notification.sh #7701
  • Fix notification scripts string comparison #7647
  • Service and host mail-notifications: Add line-breaks to very long output #6822
  • Set correct UTF-8 email subject header (RFC1342) #6369
• Misc
  • Catch exception when trusted cert is not readable during node setup on agent/satellite #7838
  • CheckCommand ssl: Fix wrong parameter -N #7741
  • Code quality fixes
  • Small documentation fixes

v2.11.3

Issues and PRs
Blogpost

This release addresses a critical crash regarding our cluster JSON-RPC connections. This mainly affects users who run large environments and use HA zones. In most cases updating all masters should resolve the problem, but we still recommend updating all Icinga 2 instances.

Bugfixes

• Cluster
  • Fix JSON-RPC crashes (#7532) in large environments: #7846 #7848 #7849

v2.11.2

https://icinga.com/2019/10/24/icinga-2-11-2/

v2.11.1

https://icinga.com/2019/10/17/icinga-2-11-1-release/