ID: 13856
Title: Pipe symbol in name of the NIC doesn't lead to crash
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 2.2.0i1
In Windows names of a network card may contain a wide variety of the
symbols including the pipe symbol. If the pipe symbol is presented in
the name, then WATO crashes.
With this release, the windows_if.ps1 plugin replaces pipe symbol, i.e.
'|' with slash symbol, i.e. '/', thus eliminating the problem.
Ticket CMK-10291
ID: 14066
Title: KUBE kube_node_container_count: adjustment to properly handle configured levels
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 2.2.0i1
Previously, configuring any levels for the kube_node_container_count check
resulted in an error because the check was unable to process the specified
format. This werk fixes this behaviour.
ID: 13802
Title: KUBE: kube_collector_info shows number of Nodes with collector
Component: Checks & agents
Level: 1
Class: New feature
Version: 2.2.0i1
With this werk, the service uses the Kubernetes API to report on the status of
DaemonSets belonging to node collectors. For each node collector the following
are reported
- the number of Nodes with a Pod, which is available and desired
- the number of Nodes, on which a Pod is desired
The DaemonSets are identified by the labels "node-collector=machine-sections"
and "node-collector=container-metrics". The service reports an OK state regardless of
the data process outcome and the API data.
ID: 13800
Title: KUBE: update section name kube_collectors_metadata_v1
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 2.2.0i1
This change affects users whom used the Check API to write their own Kubernetes check. The name of
the section kube_collectors_metadata_v1, resp. parsed section name kube_collectors_metadata, has
been changed to kube_collector_metadata_v1, resp. kube_collector_metatdata. They are now in line
with the documentation.
ID: 13804
Title: KUBE: add cmk/kubernetes/label/ to host label keys
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 2.2.0i1
This fix affects users who use Kubernetes host labels.
Previously, labels which are set by Kubernetes would be mapped one-to-one to Checkmk labels. This
would mean that setting labels in Kubernetes could overwrite Checkmk labels, which should not be
overwritten. For instance, setting the label <tt>cmk/os_family=something</tt> would incorrectly set the
corresponding Checkmk label to <tt>cmk/os_family:something</tt>. With this werk, Checkmk labels are
prefixed with <tt>cmk/kubernetes/label/</tt>. For example, Checkmk now sets the aforementioned label to
<tt>cmk/kubernetes/label/cmk/os_family:something</tt>.
User-defined rules with old host labels need to be updated.
ID: 13300
Title: BI special agent: no longer hangs during agent baking
Component: BI
Level: 1
Class: Bug fix
Version: 2.2.0i1
In the previous version the BI special agent required the configuration lock to compute the aggregation status.
Long running processes, like the agent baking or host renaming, could hold the configuration lock for a considerable
amount of time, hereby cauing the BI special agent to run into a timeout.
With this change, the agent no longer needs a locked configuration to compute its data.
ID: 13880
Title: Fix reset of automatic disk space cleanup settings on update
Component: Setup
Level: 1
Class: Bug fix
Version: 2.2.0i1
Since version 2.0.0 the options of the global setting "Automatic disk space
cleanup" were resetted to default on updating Checkmk.
If you modified the options in the past, please have a look if the current
options fit your needs and adjust them, if necessary.
ID: 13299
Title: BI datasource program: Fixed aggregation assignment to affected hosts
Component: BI
Level: 1
Class: Bug fix
Version: 2.2.0i1
The datasource option <tt>Assign to the affected hosts</tt> was broken and has been fixed.
ID: 14037
Title: oracle_performance: Wrong monitoring state when using thresholds
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 2.2.0i1
After setting up Oracle Performance Discovery rule and a threshold
in Oracle Performance rule the service would go to UNKNOWN and no
useful data was shown.
Now, the thresholds are applied and service values shown.