<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic kubernetes plugin - monitoring definition, bug? in CN-Series Discussions</title>
    <link>https://live.paloaltonetworks.com/t5/cn-series-discussions/kubernetes-plugin-monitoring-definition-bug/m-p/587760#M27</link>
    <description>&lt;P&gt;Took me long, but now i think i found strange behavior of the Kubernetes plugin - monitoring definition and especially matching criteria in the dynamic address groups.&lt;/P&gt;
&lt;P&gt;TL/DR - having service in the kubernetes cluster, which does not have assigned ports - breaks the polling and watcher&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Starting point was : working cn series as well the possibility to use the data from monitoring definition - both: to create dynamic groups and to use them in policy. Suddenly after while the monitoring definition processes (polling and watcher) stopped function (stuck in initializing state). Long story short - discovered in the kubernetes plugin logs the following suspicious lines&lt;/P&gt;
&lt;LI-CODE lang="markup"&gt;2024-05-23 13:21:39.868 +0300 ERROR: k8s-ret: skipping service, no port information found 'NoneType' object is not iterable
2024-05-23 13:21:39.869 +0300 ERROR: k8s-ret: skipping service, no port information found 'NoneType' object is not iterable
&lt;/LI-CODE&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Looking at my kubernetes cluster - there were two distinct from the other services - without ports (PORT(S) column)&lt;/P&gt;
&lt;LI-CODE lang="markup"&gt;NAMESPACE         NAME                          TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)                  AGE
longhorn-system   longhorn-engine-manager       ClusterIP   None            &amp;lt;none&amp;gt;        &amp;lt;none&amp;gt;                   87d
longhorn-system   longhorn-replica-manager      ClusterIP   None            &amp;lt;none&amp;gt;        &amp;lt;none&amp;gt;                   87d
&lt;/LI-CODE&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Next - i just add some ports definition to them - as i want to see if that will make any difference to the panorama monitoring definition processes. And it does - the processes started in normal state (connected) &lt;BR /&gt;&lt;BR /&gt;&lt;/P&gt;
&lt;LI-CODE lang="markup"&gt;2024-05-23 13:24:20.327 +0300 DEBUG: k8s-ret: &amp;lt;labkube&amp;gt; Successfully retrieved Service information.&lt;/LI-CODE&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Still not sure for the root cause of this behavior, probably can be bug. As having services without ports seems just ok for me. &lt;BR /&gt;Hope that, this behavior is reproducible elsewhere and will be addressed?&lt;/P&gt;</description>
    <pubDate>Thu, 23 May 2024 10:52:30 GMT</pubDate>
    <dc:creator>YLesev</dc:creator>
    <dc:date>2024-05-23T10:52:30Z</dc:date>
    <item>
      <title>kubernetes plugin - monitoring definition, bug?</title>
      <link>https://live.paloaltonetworks.com/t5/cn-series-discussions/kubernetes-plugin-monitoring-definition-bug/m-p/587760#M27</link>
      <description>&lt;P&gt;Took me long, but now i think i found strange behavior of the Kubernetes plugin - monitoring definition and especially matching criteria in the dynamic address groups.&lt;/P&gt;
&lt;P&gt;TL/DR - having service in the kubernetes cluster, which does not have assigned ports - breaks the polling and watcher&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Starting point was : working cn series as well the possibility to use the data from monitoring definition - both: to create dynamic groups and to use them in policy. Suddenly after while the monitoring definition processes (polling and watcher) stopped function (stuck in initializing state). Long story short - discovered in the kubernetes plugin logs the following suspicious lines&lt;/P&gt;
&lt;LI-CODE lang="markup"&gt;2024-05-23 13:21:39.868 +0300 ERROR: k8s-ret: skipping service, no port information found 'NoneType' object is not iterable
2024-05-23 13:21:39.869 +0300 ERROR: k8s-ret: skipping service, no port information found 'NoneType' object is not iterable
&lt;/LI-CODE&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Looking at my kubernetes cluster - there were two distinct from the other services - without ports (PORT(S) column)&lt;/P&gt;
&lt;LI-CODE lang="markup"&gt;NAMESPACE         NAME                          TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)                  AGE
longhorn-system   longhorn-engine-manager       ClusterIP   None            &amp;lt;none&amp;gt;        &amp;lt;none&amp;gt;                   87d
longhorn-system   longhorn-replica-manager      ClusterIP   None            &amp;lt;none&amp;gt;        &amp;lt;none&amp;gt;                   87d
&lt;/LI-CODE&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Next - i just add some ports definition to them - as i want to see if that will make any difference to the panorama monitoring definition processes. And it does - the processes started in normal state (connected) &lt;BR /&gt;&lt;BR /&gt;&lt;/P&gt;
&lt;LI-CODE lang="markup"&gt;2024-05-23 13:24:20.327 +0300 DEBUG: k8s-ret: &amp;lt;labkube&amp;gt; Successfully retrieved Service information.&lt;/LI-CODE&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Still not sure for the root cause of this behavior, probably can be bug. As having services without ports seems just ok for me. &lt;BR /&gt;Hope that, this behavior is reproducible elsewhere and will be addressed?&lt;/P&gt;</description>
      <pubDate>Thu, 23 May 2024 10:52:30 GMT</pubDate>
      <guid>https://live.paloaltonetworks.com/t5/cn-series-discussions/kubernetes-plugin-monitoring-definition-bug/m-p/587760#M27</guid>
      <dc:creator>YLesev</dc:creator>
      <dc:date>2024-05-23T10:52:30Z</dc:date>
    </item>
    <item>
      <title>Re: kubernetes plugin - monitoring definition, bug?</title>
      <link>https://live.paloaltonetworks.com/t5/cn-series-discussions/kubernetes-plugin-monitoring-definition-bug/m-p/588277#M30</link>
      <description>&lt;P&gt;Hi &lt;SPAN style="background: var(--ck-color-mention-background); color: var(--ck-color-mention-text);"&gt;&lt;a href="https://live.paloaltonetworks.com/t5/user/viewprofilepage/user-id/373715045"&gt;@YLesev&lt;/a&gt;&lt;/SPAN&gt; ,&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Thanks for sharing your findings. I would recommend reaching out to support to share what you have.&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Wed, 29 May 2024 15:36:35 GMT</pubDate>
      <guid>https://live.paloaltonetworks.com/t5/cn-series-discussions/kubernetes-plugin-monitoring-definition-bug/m-p/588277#M30</guid>
      <dc:creator>JayGolf</dc:creator>
      <dc:date>2024-05-29T15:36:35Z</dc:date>
    </item>
  </channel>
</rss>

