ID: 15960
Title: permissions: allow guest users Host read access on monitoring endpoints
Component: REST API
Level: 1
Class: Bug fix
Version: 2.3.0b1
After some security updates to the REST-API, guest users were
no longer able to view host data on monitoring endpoints due
to insufficient permissions. There is a differentiation between
'monitoring' and 'setup' endpoints with regards to permissions
which was previously not taken into account.
This werk addresses this issue by allowing monitoring read
access for guest users.
ID: 13270
Title: agent_ibmsvc: option -i/--modules not recognized without --accept-any-hostkey
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 2.2.0p8
The information regarding 'Retrieve information about...' provided for the `ibmsvc` special agent
would be ignored if 'Accept any SSH Host Key' was unchecked.
This change makes the information about modules be used unconditionally.
ID: 15258
Title: Removed StatsGroupBy header from Livestatus protocol
Component: Livestatus
Level: 1
Class: New feature
Version: 2.3.0b1
The StatsGroupBy header has been deprecated for almost 13 years now,
just use the Columns instead.
ID: 15928
Title: Fix visibility of cloned view, dashboard and report extensions
Component: Multisite
Level: 1
Class: Bug fix
Version: 2.3.0b1
Werk #15135 introduced the possibility to share views, dashboards and report
definitions.
If you cloned an installed extension and changed the configuration of the
clone, opening the view, dashboard or report always showed the configuration
of the extension.
This has been fixed, please review your shared views, dashboards and reports if
you used extensions.
ID: 15969
Title: cisco_temperature: Correct warning threshold for environment temperature
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 2.3.0b1
Currently only the threshold specified by OID 1.3.6.1.4.1.9.9.13.1.3.1.4 (ciscoEnvMonTemperatureThreshold) is considered when checking the temperature of a Cisco switch. As only one threshold is given here, it is used as both WARN and CRIT threshold, causing confusion when the device itself reports a warning state based on a warning threshold not given via this OID.
Now the upper thresholds specified in OID 1.3.6.1.4.1.9.9.91.1.2.1.1 (entSensorThresholdTable) will be used as WARN and CRIT thresholds if an entry for the relevant switch can be found.
For this, the threshold relation and severity will be used to determine if the threshold can be applied in Checkmk, otherwise the threshold will be ignored.
If no thresholds from the entSensorThresholdTable apply, the ciscoEnvMonTemperatureThreshold is used as a fallback.
ID: 15903
Title: cisco_temperature, cisco_temperature_dom: Improved check state and summary if sensor status not ok
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 2.3.0b1
The cisco_temperature and cisco_temperature_dom checks currently only consider sensors with the state "ok". If a sensor becomes "unavailable" or "non-operational", the check becomes "UNKNOWN" with the summary "Item not found in monitoring data".
With this werk, the summary will now report the status of the sensor instead.
For the "unavailable" sensor status, the check state will still change to "UNKNOWN", for "non-operational" status it will be CRIT.
The discovery is not affected by this and will still only discover "ok" sensors.