Checkmk werks level1 December 2023

checkmk-werks-lvl1@lists.checkmk.com

1 participants
128 discussions

[2.3.0] Checkmk Werk 16284 created: host_tag_group: add help field to the response schema

by Checkmk werks level 1

Title: host_tag_group: add help field to the response schema Class: fix Compatible: compat Component: rest-api Date: 1702382101 Edition: cre Level: 1 Version: 2.3.0b1

10 months

[2.3.0] Checkmk Werk 16285 created: rules: match_on items must be at least of length 1

by Checkmk werks level 1

Title: rules: match_on items must be at least of length 1 Class: fix Compatible: compat Component: rest-api Date: 1702450037 Edition: cee Level: 1 Version: 2.3.0b1 This werk addresses an issue found by our QA team on the rules collection POST endpoint ...domain-types/rule/collections/all The match_on field requires a list of strings but didn't enforce any length. Previously, passing an empty string would provoke a crash. We now enforce a minimum string length of 1.

10 months

[2.3.0] Checkmk Werk 15025 created: User defined "PING IPv4/6" services are no longer overwritten on dual stack hosts

by Checkmk werks level 1

[//]: # (werk v2) # User defined "PING IPv4/6" services are no longer overwritten on dual stack hosts key | value ---------- | --- date | 2023-12-18T13:50:29+00:00 version | 2.3.0b1 class | fix edition | cre component | core level | 1 compatible | no For a dual stack host check mk will create *PING IPv4* and *Ping IPv6* services automatically. Those services are created by the core configuration and cannot be deactivated. With the CMC core two *Ping IPv4/6* services would appear on the host, the nagios core ignores the user defined service. The core configuration now only creates the respective PING service for the main interface of the host. If you have a user defined *Ping IPv4/6* service it will replace the old auto generated one with the next *Activate Changes*.

10 months

[2.2.0] Checkmk Werk 16146 created: Avoid crash upon failure to create RRD files

by Checkmk werks level 1

Title: Avoid crash upon failure to create RRD files Class: fix Compatible: compat Component: core Date: 1702630948 Edition: cre Level: 1 Version: 2.2.0p18 Failure to create an RRD file would generate a crash report. Now, we generate a log entry explaining the error, without crashing. This has no incidence on the monitoring but the data is not saved on disk anymore.

10 months

[2.3.0] Checkmk Werk 16274 deleted: Unusable services for "HPE Proliant Servers: Raid Status"

by Checkmk werks level 1

Werk 16274 was deleted. The following Werk is no longer relevant. Title: Unusable services for "HPE Proliant Servers: Raid Status" Class: fix Compatible: incomp Component: checks Date: 1702636533 Edition: cre Level: 1 Version: 2.3.0b1 This fixes some unusable services of the "HPE Proliant Servers: Raid Status" plugin. This werk only affecs you if you observe unknown "Logical device <ITEM>" services after the upgrade. In case you are affected please run a discovery on the affected hosts and fix the monitoring history as described below. We observed some devices to send <tt>"\x00"</tt> (the null-byte) as their name (OID .1.3.6.1.4.1.232.3.2.3.1.1.14). Not all components delt well with it, leading to unacknowlegable downtimes, non matching rules and the like. We now replace all null-bytes with <tt>"\\x00"</tt> (the literal containing the four characters backslash, 'x', 'zero', 'zero'). As of Checkmk 2.3, this should in fact no longer be necessary, but as we need a backportable solution, we change the item consistently in all versions. If this affects you, you might also want to fix the services in the monitring history by running C+: sed -i 's|\x00|\\x00|' var/check_mk/core/history var/check_mk/core/archive/* C-:

10 months

[2.3.0] Checkmk Werk 16274 created: Unusable services for "HPE Proliant Servers: Raid Status"

by Checkmk werks level 1

[//]: # (werk v2) # Unusable services for "HPE Proliant Servers: Raid Status" key | value --- | --- compatible | no version | 2.3.0b1 date | 2023-12-15T10:35:33+00:00 level | 1 class | fix component | checks edition | cre This fixes some unusable services of the "HPE Proliant Servers: Raid Status" plugin. This werk only affects you if you observe unknown "Logical device <ITEM>" services after the upgrade. In case you are affected please run a discovery on the affected hosts and fix the monitoring history as described below. We observed some devices to send `""` (the null-byte) as their name (`OID .1.3.6.1.4.1.232.3.2.3.1.1.14`). Not all components delt well with it, leading to unacknowlegable downtimes, non matching rules and the like. We now replace all null-bytes with `"\x00"` (the literal containing the four characters backslash, 'x', 'zero', 'zero'). As of Checkmk 2.3, this should in fact no longer be necessary, but as we need a backportable solution, we change the item consistently in all versions. If this affects you, you might also want to fix the services in the monitoring history by running ``` sed -i 's||\x00|' var/check_mk/core/history var/check_mk/core/archive/* ```

10 months

[2.3.0] Checkmk Werk 16146 created: Avoid crash upon failure to create RRD files

by Checkmk werks level 1

Title: Avoid crash upon failure to create RRD files Class: fix Compatible: compat Component: core Date: 1702630948 Edition: cre Level: 1 Version: 2.3.0b1 Failure to create an RRD file would generate a crash report. Now, we generate a log entry explaining the error, without crashing. This has no incidence on the monitoring but the data is not saved on disk anymore.

10 months

[2.1.0] Checkmk Werk 16274 created: Unusable services for "HPE Proliant Servers: Raid Status"

by Checkmk werks level 1

Title: Unusable services for "HPE Proliant Servers: Raid Status" Class: fix Compatible: incomp Component: checks Date: 1702636533 Edition: cre Level: 1 Version: 2.1.0p38 This fixes some unusable services of the "HPE Proliant Servers: Raid Status" plugin. This werk only affecs you if you observe unknown "Logical device <ITEM>" services after the upgrade. In case you are affected please run a discovery on the affected hosts and fix the monitoring history as described below. We observed some devices to send <tt>"\x00"</tt> (the null-byte) as their name (OID .1.3.6.1.4.1.232.3.2.3.1.1.14). Not all components delt well with it, leading to unacknowlegable downtimes, non matching rules and the like. We now replace all null-bytes with <tt>"\x00"</tt> (the literal containing the four characters backslash, 'x', 'zero', 'zero'). As of Checkmk 2.3, this should in fact no longer be necessary, but as we need a backportable solution, we change the item consistently in all versions. If this affects you, you might also want to fix the services in the monitring history by running C+: sed -i 's|\x00|\\x00|' var/check_mk/core/history var/check_mk/core/archive/* C-:

10 months

[2.1.0] Checkmk Werk 16275 created: Crash of Checkmk Agent service if MRPE plugins are used

by Checkmk werks level 1

Title: Crash of Checkmk Agent service if MRPE plugins are used Class: fix Compatible: compat Component: checks Date: 1702646733 Edition: cre Level: 1 Version: 2.1.0p38

10 months

[2.2.0] Checkmk Werk 16274 created: Unusable services for "HPE Proliant Servers: Raid Status"

by Checkmk werks level 1

Title: Unusable services for "HPE Proliant Servers: Raid Status" Class: fix Compatible: incomp Component: checks Date: 1702636533 Edition: cre Level: 1 Version: 2.2.0p18 This fixes some unusable services of the "HPE Proliant Servers: Raid Status" plugin. This werk only affecs you if you observe unknown "Logical device <ITEM>" services after the upgrade. In case you are affected please run a discovery on the affected hosts and fix the monitoring history as described below. We observed some devices to send <tt>"\x00"</tt> (the null-byte) as their name (OID .1.3.6.1.4.1.232.3.2.3.1.1.14). Not all components delt well with it, leading to unacknowlegable downtimes, non matching rules and the like. We now replace all null-bytes with <tt>"\x00"</tt> (the literal containing the four characters backslash, 'x', 'zero', 'zero'). As of Checkmk 2.3, this should in fact no longer be necessary, but as we need a backportable solution, we change the item consistently in all versions. If this affects you, you might also want to fix the services in the monitring history by running C+: sed -i 's|\x00|\\x00|' var/check_mk/core/history var/check_mk/core/archive/* C-:

10 months

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

Checkmk werks level1 December 2023