ID: 6662
Title: Timespecific check parameters: Changed computation algorithm to allow more flexible configurations
Component: WATO
Level: 2
Class: New feature
Version: 1.6.0i1
Werk #4840 introduced timespecific parameters, which allowed timedependant check parameter
configuration through timeperiods. Customer feedback has shown that these configuration were
not flexible enough. Thats why we've changed the computation algorithm.
The new algorithm only affects dictionary based check parameters, tuple based check parameters remain unchanged.
H3:Ruleset example old configuration (deprecated with this werk)
<ul>
<li>Use parameters {"warn": 90} on saturday, default {"warn": 75, "crit": 80}</li>
<li>Use parameters {"warn": 85, "crit": 92"} on weekend, default {"warn": 70, "crit": 75, "bandwidth": 1000}</li>
<li>Use parameters {"check_mode": "auto"} (no timespecific rule)</li>
</ul>
The old mechanism created the following parameters
<ul>
<li>Monday till Friday: {"warn": 75, "crit": 80"}</li>
<li>Saturday: {"warn": 90}</li>
<li>Sunday: {"warn": 75, "crit": 80}</li>
</ul>
As you can see, only the parameters from the first rule were taken into account.
H3:Ruleset example new configuration
<ul>
<li>Use parameters {"warn": 90} on saturday, default {"warn": 75, "crit": 80}</li>
<li>Use parameters {"warn": 85, "crit": 92"} on weekend, default {"warn": 70, "crit": 75, "bandwidth": 1000}</li>
<li>Use parameters {"check_mode": "auto"} (no timespecific rule)</li>
</ul>
The new mechanism now creates the following parameters
<ul>
<li>Monday till Friday: {"warn": 75, "crit": 80", "check_mode": "auto"}</li>
<li>Saturday: {"warn": 90, "crit": 80, "bandwith": 1000, "check_mode": "auto"}</li>
<li>Sunday: {"warn": 85, "crit": 92, "bandwith": 1000, "check_mode": "auto"}</li>
</ul>
As you can see, only the parameters from the first rule were taken into account.
The rule evaluation now starts at the bottom rule and then progresses to the top rule.
For each rule, first of all the default parameters are applied. Afterwards the timespecific
component(s) are applied. If a single rule have more than one timespecific parameter,
these parameters are merged, too.
For example:
<ul>
<li>Use parameters {"warn": 75, "crit": 80, "check_mode": "auto"} on Saturday 10-14h, Use parameters {"special_parameter": True} on Saturday 12-14h</li>
<li>Use parameters {"warn": 85, "crit": 92"} on weekend, default {"warn": 70, "crit": 75, "bandwidth": 1000}</li>
<li>Use parameters {"check_mode": "auto"} (no timespecific rule)</li>
</ul>
Result
<ul>
<li>Saturday 11:00: {"warn": 75, "crit": 80, "bandwidth": 1000, "check_mode": "auto"}</li>
<li>Saturday 13:00: {"warn": 75, "crit": 80, "bandwidth": 1000, "check_mode": "auto", "special_parameter": True}</li>
</ul>
ID: 6573
Title: Azure Monitoring
Component: Checks & agents
Level: 1
Class: New feature
Version: 1.6.0i1
This werk adds a new datasource program to monitor Microsoft Azure cloud environments.
ID: 6570
Title: azure_sites: New Check to monitor azure web servers
Component: Checks & agents
Level: 1
Class: New feature
Version: 1.6.0i1
This werk adds a check to monitor IIS web servers in the Azure cloud.
ID: 6661
Title: Periodic service discovery/cronjob: Changed loglevel of informational message
Component: WATO
Level: 1
Class: Bug fix
Version: 1.6.0i1
The discover-marked-hosts cronjob no longer receives a <tt>Timeout of 120 seconds reached ...</tt> message,
in case the discovery takes longer than expected. This logmessage is now only shown in higher log level.
ID: 6640
Title: Faster link lookup to Check_MK host or service pages
Component: Notifications
Level: 1
Class: Bug fix
Version: 1.6.0i1
Context variables HOSTURL and SERVICEURL now include the site key, this
enables a faster lookup and visualization of pages since Check_MK does not
need to resolve to which site the host belongs. This becomes especialy
relevant in distributed environments with multiple monitoring sites.
ID: 6639
Title: slack: New notification plugin for slack
Component: Notifications
Level: 1
Class: New feature
Version: 1.6.0i1
It is now possible to for Check_MK to send notifications about state
changes of hosts and services into a Slack channel.
Slack needs to be first configured to recieve messages from Check_MK. This
can be done by setting up a Webhook <a
href=\"https://my.slack.com/services/new/incoming-webhook/\"
target=\"_blank\"> here </a>, where you specify which channel wil recieve
the notifications. In Check_MK, under WATO->Notifications, create a new
notification rule. And select under Notification Method Slack. Copy the URL
that Slack gives you after completing the configuration into the Slack
Webhook-URL field. Optionally if you want sent messages to include
hyperlinks to Check_MK, enable the field URL prefix for links to Check_MK.
ID: 6603
Title: CRE: Fixed Check_MK service crash if a check plugin is unknown to the check context
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 1.6.0i1
A host has discovered a set of check plugin names which are triggered by the
Check_MK service. If the agent type of this host changes without performing a
rediscovery afterwards and the configuration changes are activated then there
may be new or vanished check plugins which are unknown to the current check
context. This leads to a KeyError of the Check_MK service.
Now if a check plugin is unknown to the current check context Check_MK ignores
this check plugin. Aside from that the Check_MK Discovery service reports about
vanished or unmonitored serices.