ID: 1066
Title: Implemented Dashboard Designer
Component: Multisite
Level: 3
Class: New Feature
Version: 1.2.5i6
This change implements the long awaited Dashboard Designer. With this feature users
can edit existing dashboards, create new ones and provide their dashboards to other
users just like it can be done with the views.
Another nice feature is the formalized context we introduced for views and dashboards.
With this concept it is possible to create a host specific dashboard once and open
it for several different hosts. You can create a "Oracle Database dashboard" once
which can show you the state of a single database server. The name of the database
server (context: hostname) of your choice can be provided when accessing the dashboard
to "instanciate" such a dashboard.
Dashboards edited by users are saved in <tt>var/check_mk/web/USER/user_dashboards.mk</tt>
just like edited views are now saved in <tt>var/check_mk/web/USER/user_views.mk</tt>. The
views saved by a user before this release which are stored in <tt>var/check_mk/web/USER/views.mk</tt>
will be migrated automatically once the user edits it's first view. From this time the
new edited views will be read/written from/to the new file.
For details take a look at the <a href="/checkmk_multisite_dashboards.html">dashboard documentation</a>.
ID: 1065
Title: heartbeat_crm / heartbeat_crm.resources: Rewrote checks / formalized parameters
Component: Checks & Agents
Level: 1
Class: New Feature
Version: 1.2.5i6
Both checks heartbeat_crm and heartbeat_crm.resources nailed down the resources / DC to
the node which held the resources / DC during inventory. This was not fitting most users
requirements and has been changed. By default the node currently holding the resources / DC
does not matter anymore as long as they are held by at least one node.
If you like to change this behaviour back to the old default, you need to configure the
inventory rule <tt>Heartbeat CRM Inventory</tt> and re-run an inventory on the affected
hosts.
Already inventorized checks should continue working as they did before. The change only
affects newly inventorized services.
Two new WATO rules <tt>Heartbeat CRM general status</tt> and <tt>Heartbeat CRM resource status</tt>
have been added to make the parameters of these checks configurable.
ID: 1089
Title: Snapshot restore: fixed exception during exception handling...
Component: WATO
Level: 1
Class: Bug Fix
Version: 1.2.5i6
If there were an error during the extracting of a backup domain, the backup process
crashed entirely instead of continueing with the next backup domain.
ID: 0185
Title: knuerr_rms_humidity, knuerr_rms_temp: Two new Checks to Monitor the Temperature and the Humidity on Knürr RMS Devices
Component: Checks & Agents
Level: 1
Class: New Feature
Version: 1.2.5i6
ID: 1064
Title: Fixed rare issue with WATO communication in distributed setups (different OS versions)
Component: WATO
Level: 1
Class: Bug Fix
Version: 1.2.5i6
In distributed setups where the python versions differ, especially the remote site is older
than Python 2.5 and the central site is newer than this, it might happen that the communication
between these both sites is not possible anymore since werk #0984 (2014-06-23 / 1.2.5i4).
This commit makes the communication work again with older python versions. This has been tested
with Python 2.4. It might also work with python 2.3 - but is untested yet.
ID: 1051
Title: tcp_conn_stats: fix missing performance data
Component: Checks & Agents
Level: 1
Class: Bug Fix
Version: 1.2.5i6
Due to a recent code rework the TCP Connections check had output invalid
performance data. This has been fixed.
ID: 0184
Title: knuerr_sensors: New Check to monitor Sensors on a Knürr RMS Device
Component: Checks & Agents
Level: 1
Class: New Feature
Version: 1.2.5i5
ID: 1076
Title: Fix wrong percentual host availability > 100% when excluding downtimes
Component: Reporting & Availability
Level: 2
Class: Bug Fix
Version: 1.2.5i5
When you chose to exclude scheduled downtimes from the "100%" then hosts with
actual downtime periods could get an availability for more then 100%. This has
been fixed. The Settings <i>ignore</i> and <i>honor</i> were not affected. Also
not the display of absolute time values.