ID: 14313
Title: hp_proliant_raid: illogical rebuild progress percentage
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 2.2.0i1
The check "hp_proliant_raid" shows the progress percentage of rebuild tasks for
raids that are currently rebuilding. In some cases, this percentage was shown
as 4294967295%. The number "4294967295" is returned by the device in the event
that the progress of the rebuild task cannot be determined. The check now shows
an appropriate message instead.
ID: 14312
Title: hp_proliant_raid: new item names
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 2.2.0i1
The check "hp_proliant_raid" showed an empty item name in the event that a name
for the raid had not been set. This is problematic as all items must have an
item name in the new Check API. The item name has been amended to include both
the raid's name and its index. As a result, services need to be rediscovered on
the host.
ID: 14401
Title: crash in legacy plugins with empty details
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 2.2.0i1
Some legacy plugins can print a newline indicating details to come and not follow up with any details.
This lead to a crash with the message "ValueError ('details' must be non-empty str or None, got )".
This has been fixed. No user interaction required.
ID: 14421
Title: <tt>mssql_datafiles/transactionlogs</tt>: Correct maximum size in unlimited cases
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 2.2.0i1
The check plugins <tt>mssql_datafiles/transactionlogs</tt> did not handle unlimited instances
correctly. Specifically, the maximum size only took the space still available on the filesystem
into account, but not the current size of the file. This resulted in too low thresholds in case
of percentage levels.
ID: 14486
Title: 'Host groups' view: Add coloring for host states
Component: Multisite
Level: 1
Class: Bug fix
Version: 2.2.0i1
In the view "Host groups" host state cells were not colored according to their state. This is fixed, so now whenever there's at least one occurence of the given host state for the given host group, the table cell is colored. This behavior now matches that of the service state cells again.
ID: 14460
Title: Fix error on rule search
Component: Multisite
Level: 1
Class: Bug fix
Version: 2.2.0i1
Searching for e.g. "Group" or "Ineffective" could lead to an error message like
"The selected element 'None' is not longer available.".
ID: 13925
Title: Drop dashboard "Checkmk server"
Component: Multisite
Level: 1
Class: New feature
Version: 2.2.0i1
The "Checkmk server" dashboard is dropped. More informational overviews for Checkmk servers are available with the "Checkmk dashboard", as well as the linux specific dashboards "Linux hosts" and "Single Linux host".
ID: 13935
Title: outbound validation in REST API when attributes got deleted
Component: REST API
Level: 1
Class: Bug fix
Version: 2.2.0i1
There can be a case where attributes are stored on a host or folder
but the definition of these attributes were deleted. This would then
result in an error on outbound validation.
ID: 14374
Title: Prometheus special agent configuration: Crash when updating from Checkmk 2.0 to 2.1
Component: Setup
Level: 1
Class: Bug fix
Version: 2.2.0i1
When updating from Checkmk 2.0 to 2.1, there was a crash related to the configuration of the Prometheus special agent:
C+:
2022-06-24 12:59:27 - Updating site '...' from version 2.0.0p26.cee to 2.1.0p3.cee...
...
-| 6/25 Rewriting rulesets...
...
-| ERROR: Failed to transform rule: (Ruleset: special_agents:prometheus, ...)
-| + "Rewriting rulesets" failed
-| Traceback (most recent call last):
...
-| AssertionError
C-:
After the update, the user interface displayed the following error message when attempting to edit the affected Prometheus rules:
"Unable to read current options of this rule. Falling back to default values. When saving this rule now, your previous settings will be overwritten. Problem was: Prometheus connection option: Value must be a tuple with two elements"