ID: 6838
Title: skype.data_proxy, wmi_webservices: Fix failing inventory for missing items
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 1.6.0i1
No longer report fail in service discovery if no items are
listed in wmi tables.
ID: 6862
Title: Fixed Handle Leak in Windows Agent
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 1.6.0i1
To prevent handle leaks in Windows Agent for certain versions of Windows OS
provided perf_reader.exe utility in ~/share/doc/check_mk/treasures/windows directory.
Short description can be found ~/share/doc/check_mk/treasures/windows/readme.txt
ID: 6679
Title: Introducing BI datasource program and get_bi_aggregations API call
Component: Checks & agents
Level: 1
Class: New feature
Version: 1.6.0i1
This werks introduces the
<ul>
<li>Thew new Web API interface get_bi_aggregations</li>
<li>A datasource program, which uses this interface</li>
<li>Two checks bi_aggregation and bi_aggregation_connection, which evaluate the data</li>
</ul>
Monitoring of BI aggregations was already available with the active check <tt>Check state of BI aggregation</tt>.
This active check could only monitor one aggregation at a time and furthermore drastically slowed the Web GUI when
configured for multiple hosts. In some rare instances, when configured for hundreds of hosts it could also lock down
the GUI by occupying all available apache workers.
The purpose of this datasource program is to fetch all data in one call and distribute it to specific hosts via piggyback.
The use of the BI datasource program is the new recommended way when monitoring BI aggregations.
<h3>New API Interface: <tt>get_bi_aggregations</tt></h3>
Calling this API interface without any extra options returns a dictionary, which includes
a list of all aggregation and some additional information about unresponsive sites and unavailable aggregations
You can specify a filter dictionary which currently may have two different kind of filter types.
For example
<tt>"filter": {"names": ["Aggr HostA", "Aggr HostB"]}</tt><br>
<tt>"filter": {"groups": ["AggregationGroup1", "AggregationGroup2"]}</tt><br>
<tt>"filter": {"names": ["Aggr HostA"], "groups": ["AggregationGroup1"]}</tt>
You can find more information in our official documentation soon.
<h3>New datasource program: <tt>Check BI Aggregations via Web Interface</tt></h3>
The datasource program is configurable in WATO. It is able to contact multiple monitoring sites at the same time.
Each site can be configured indepedently. The configuration offers the following options
<ul>
<li>Filter fetched aggregations by regex and aggregation groups</li>
<li>Assign fetched aggregations to specific hosts via piggyback, for example all hosts which are affected by this aggregations</li>
<li>Rewrite of aggregation states, e.g. State if aggregation is in downtime, State if aggregation is acknowledged</li>
</ul>
<h3>New checks</h3>
There are two new checks.
<tt>bi_aggregation</tt> simply displays the state of the aggregation. It is not configurable
<tt>bi_aggregation_connection</tt> shows connections problems with the monitoring site(s).
ID: 6678
Title: The hostname painter now has configurable background colors
Component: Multisite
Level: 1
Class: New feature
Version: 1.6.0i1
In addition you can also configure that the background color changes if the host is in downtime.
The following coloring options are now available
<ul>
<li>Colorize background if host is up</li>
<li>Colorize background if host is down</li>
<li>Colorize background if host is unreachable</li>
<li>Colorize background if host is pending</li>
<li>Colorize background if host is in downtime</li>
</ul>
ID: 6855
Title: Locally installed notification plugins were not immediately available
Component: Notifications
Level: 1
Class: Bug fix
Version: 1.6.0i1
When installing a new notification plugin to local/share/check_mk/notifications
this script was not immediately visible in the configuration GUI in all
previous 1.5 releases.
Workaround: Restarting the site apache made the script available in previous
versions.
ID: 6416
Title: mknotifyd: set a default timeout if no value is given for a method
Component: Notifications
Level: 1
Class: Bug fix
Version: 1.6.0i1
If the option "Timing settings for specific methods" was used for a notification
method and no explicit timeout was given an exception was raised in the mknotifyd
if a timeout occured. As a result notification processes that run in a timeout
are not killed, the list of notifications that are in progress are not updated
as long as the timed out process exists and the output of the timed out process
is discarded.
The issue is now fixed and the default timeout of 60s is set if no timeout
is specified in the global settings.
If you use the option "Timing settings for specific methods" in versions prior
to 1.5.0p8 please set a timeout explicitlyto circumvent this issue.
ID: 6634
Title: cmciii: Fix incorrect values due to ignored unit value
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 1.6.0i1
In some cases, power or energy values would be off by a factor 1000
due to the unit reported via SNMP not being queried. This has been fixed.
ID: 6854
Title: Host diagnose was not using SNMPv3 credentials correctly
Component: WATO
Level: 1
Class: Bug fix
Version: 1.6.0i1
The host diagnose page was not respecting the SNMPv3 credentials provided by
the user on the host diagnose page. It was doing a SNMPv1/v2 test instead
of SNMPv3 instead of using v3.