ID: 11204
Title: Fixed CSV export of host and service labels
Component: Multisite
Level: 1
Class: Bug fix
Version: 1.7.0i1
Previously, host and service labels in CSV exports were in JSON format. From
now on, labels will be separated by commas for such exports.
ID: 11266
Title: ups_out_load, ups_power: Fix discovery of output lines with zero load resp. power
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 1.7.0i1
Previously output lines were discovered in the check plugins {{ups_out_load}},
{{ups_power}} if and only if they had a value greater than zero.
Now we use the voltage value in order to decide whether a line is connected or
not.
ID: 11363
Title: if64, if64adm: Monitoring of the admin status <tt>ifAdminStatus</tt> for switch ports
Component: Checks & agents
Level: 1
Class: New feature
Version: 1.7.0i1
The 64-bit version of the interface check for switch ports can be configured
to also monitor the admin status <tt>ifAdminStatus</tt>. If this option is
activated, checkmk uses the check plugin <tt>if64adm</tt> instead of
<tt>if64</tt>. Up to now, only an <tt>ifAdminStatus</tt> of 2 was reported by
setting the port state to <i>9 - admin down</i>. With this werk, the admin
status is now monitored completely separately from the port state. Consequently,
the port state <i>9 - admin down</i> is not available any more.
Previously, the monitoring of the admin state was activated implicitly by configuring
checkmk to discover ports in the state <i>9 - admin down</i> (rule "Network Interface
and Switch Port Discovery"). Now, this is handled in a separate rule: "SNMP Interface
check: Monitor ifAdminStatus (use 'if64adm' instead of 'if64')". If and only if this
rule is activated, the admin status will be monitored, independently of the
configuration in "Network Interface and Switch Port Discovery".
Furthermore, since the admin status is now reported separately from the port state,
it can also be used to configure 1) the discovery of network interfaces (rule "Network
Interface and Switch Port Discovery"), 2) the monitoring state (rule "Network interfaces
and switch ports"). For example, you can make checkmk discover only interfaces with
an admin status of 2 or you can set the monitoring state of all interfaces with an
amdin status of 1 to WARNING. Note that these configuration options only have an
effect if the admin status is actually monitored, i.e., if the new rule mentioned
above is activated.
This change is incompatible for all users who are currently monitoring the admin
status using <tt>if64adm</tt>. These users should follow these steps:
<ul>
<li>Activate the monitoring of the admin status using the new rule "SNMP Interface
check: Monitor ifAdminStatus (use 'if64adm' instead of 'if64')".
</li>
<li>Check the discovery configuration (rule "Network Interface and Switch Port
Discovery"). As mentioned above, the admin state can now be used to further refine the
discovery of network interfaces. On the other hand, the port state <i>9 - admin down</i>
is not available any more, so users might have to adjust their configurations.</li>
<li>Re-discover the services of affected hosts.</li>
<li>Check the monitoring configuration (rule "Network interfaces and switch ports").
Analogously to the discovery configuration, the admin state can now be used to set
the monitoring state of network interfaces and the port state <i>9 - admin down</i>
is not available any more.</li>
</ul>
ID: 11265
Title: Discovery Page: Do not show last job failures if they are fixed
Component: WATO
Level: 1
Class: Bug fix
Version: 1.7.0i1
Previously failures of last discovery jobs were displayed in a red message box again.
This happened in the following situation: If a hostname was not resolveable the
discovery page displayed {{Failed to lookup IP address of HOST via DNS}}. After
an IP address was configured this message was shown again on the discovery page
but also a discovered services table. This behaviour is misleading.
Now the last error message is put in the current job details but will not be
displayed as a critical message again.
ID: 11260
Title: Fix crash in EC configuration (regression since 1.6.0p14)
Component: Event Console
Level: 1
Class: Bug fix
Version: 1.7.0i1
When there were open events in the Event Console that were created by a rule
that was removed from the Event Console, this resulted in an exception on the
Event Console configuration page (IndexError (list index out of range)).
However, this message was rather a consequence error. The original error
message was found in var/log/mkeventd.log. There you could find messages of
this kind:
C+:
2020-08-27 14:19:42,913 [40] [cmk.mkeventd.StatusServer] Error handling client : 'my_rule'
Traceback (most recent call last):
File "/omd/sites/stable/lib/python/cmk/ec/main.py", line 3031, in serve
File "/omd/sites/stable/lib/python/cmk/ec/main.py", line 3071, in handle_client
File "/omd/sites/stable/lib/python/cmk/ec/main.py", line 3092, in _answer_query
File "/omd/sites/stable/lib/python/cmk/ec/main.py", line 2753, in query
File "/omd/sites/stable/lib/python/cmk/ec/main.py", line 2871, in _enumerate
File "/omd/sites/stable/lib/python/cmk/ec/main.py", line 736, in get_status
row += self._add_event_limit_status()
File "/omd/sites/stable/lib/python/cmk/ec/main.py", line 766, in _add_event_limit_status
self.get_rules_with_active_event_limit(),
File "/omd/sites/stable/lib/python/cmk/ec/main.py", line 1784, in get_rules_with_active_event_limit
if num_events >= self._get_rule_event_limit(rule_id)[0]:
File "/omd/sites/stable/lib/python/cmk/ec/main.py", line 1885, in _get_rule_event_limit
rule = self._rule_by_id.get(rule_id)
KeyError: 'my_rule'
C-:
ID: 11244
Title: Check_MK Discovery service: Contact SNMP devices for real if 'Perform a full SNMP scan' is enabled
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 1.7.0i1
There are two options {{Perform a full SNMP scan always, detect new check types}}
and {{Just rely on existing check files, detect new items only}} especially for
SNMP devices in the ruleset {{Periodic service discovery}}. The idea behind the
first option is contacting SNMP devices for real and possibly find new check
types. This option had no effect since Checkmk version 1.5.0 and cached data
was always used. The result was that the {{Check_MK Discovery}} service might
not have found new check typs on SNMP devices.
Now if this option is set Checkmk really contacts SNMP devices and may find new
check types and services.
Please note: The execution time of the {{Check_MK Discovery}} service may take
more time as before on hosts with SNMP data sources. In this case you can
increase the regular and retry check interval of this service using the ruleset
{{Normal check interval for service checks}} and {{Retry check interval for service checks}}
in order to prevent timeouts.
ID: 11243
Title: Bulk Dicovery: Align caching options with discovery page
Component: WATO
Level: 1
Class: Bug fix
Version: 1.7.0i1
On the {{Bulk Discovery}} WATO page you could set the options
{{Use cached data if present}} and {{Do full SNMP scan for SNMP devices}}.
The latter one had no effect since Checkmk version 1.5.0. The idea behind the
option {{Do a full SNMP scan}} was contacting the SNMP device and ignoring
existing caches. This behaviour is implemented behind the option
{{Do a full service scan}} on the {{WATO Discovery}} page.
Moreover the {{Bulk Discovery}} page was inconsistent with the {{WATO Discovery}}
page - independent of above options. Thus we removed both options
{{Use cached data if present}} and {{Do full SNMP scan for SNMP devices}} and
add the option {{Do a full service scan}} in order to adapt the behaviour to
the discovery page. Now both pages work the same way.
Please note: With this new option a service scan of SNMP devices may take more
time as before because now this device will be contacted for real.
ID: 11242
Title: Status of the Check_MK services: Individual, per data source configured parameters had no effect
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 1.7.0i1
Within the ruleset {{Status of the Check_MK services}} overall parameters
and/or individual, per data source parameters can be configured. The latter
ones, if configured, overwrite the overall parameters. This has been fixed.
ID: 11460
Title: Windows agent service sets correct access rights in ProgramData directory
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 1.7.0i1
Previously, a standard user can write in ProgramData/checkmk/agent directory,
thereby getting a possibility to modify sensitive information.
With this ix the above mentioned vulnerability is eliminated: a standard user
has a right to read and execute.
ID: 10314
Title: Remove deprecated check_multi plugin
Component: Site Management
Level: 1
Class: New feature
Version: 1.7.0i1
The check_multi plugin is deprecated and not maintained anymore and therefor removed from the Checkmk package.
If you are still using this plugin you can install it manually in your local hierarchy.