# Kibana stack monitoring all nodes reported to be master **URL:** \ **Category:** Kibana\ **Tags:** elastic-stack-monitoring\ **Created:** [July 25, 2022, 2:47pm UTC](https://discuss.elastic.co/t/kibana-stack-monitoring-all-nodes-reported-to-be-master/310549 "2022-07-25T14:47:34Z")\ **Posts on this page:** 6\ **Page:** 1
**Author:** ![Thuunder7](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/thuunder7/32/97482_2.png) [@Thuunder7](https://discuss.elastic.co/u/Thuunder7)\ **Post date:** [July 25, 2022, 2:47pm UTC](https://discuss.elastic.co/t/kibana-stack-monitoring-all-nodes-reported-to-be-master/310549/1 "2022-07-25T14:47:34Z")
Hello, I recently upgraded my beats, from 7.16.3 to 8.2.3 and after upgrading, for some reason metricbeat is reporting that all elasticsearch nodes are **master**. ![image](https://us1.discourse-cdn.com/elastic/original/3X/4/f/4f75a58556748358501ab3e51ae17af625c0b1b0.png) Additionally metricbeat is also reporting a wrong elasticsearch version. ![image](https://us1.discourse-cdn.com/elastic/original/3X/2/d/2d0d5d25de2afc6e2ea3488678a63a07291d7c0c.png) Is this a bug? Regards ---
**Author:** ![cheiligers](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/cheiligers/32/73114_2.png) [@cheiligers](https://discuss.elastic.co/u/cheiligers)\ **Post date:** [July 25, 2022, 8:31pm UTC](https://discuss.elastic.co/t/kibana-stack-monitoring-all-nodes-reported-to-be-master/310549/2 "2022-07-25T20:31:35Z")
@Thuunder7 did you upgrade your whole stack to 8.2.3? what version is your Elasticsearch and Kibana running on? The metricbeat UI in Kibana shows the data stored in your `metrics-*` elasticsearch indices and if the documents report the same data as what you're seeing in the UI, then there might be something else going on. ---
**Author:** ![Thuunder7](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/thuunder7/32/97482_2.png) [@Thuunder7](https://discuss.elastic.co/u/Thuunder7)\ **Post date:** [July 27, 2022, 6:31pm UTC](https://discuss.elastic.co/t/kibana-stack-monitoring-all-nodes-reported-to-be-master/310549/3 "2022-07-27T18:31:36Z")
Yes i did upgrade whole stack to 8.2.3. First i upgraded elasticsearch nodes and kibana but i gave two weeks until i decided to upgrade metricbeat and logstash. During those 2 weeks with metricbeat being on version 7.16.3 everything looked fine on Kibana stack monitoring, but when i upgraded metricbeat this "issue" appeared. I looked into metrics-\* indices but i only have these two: ![image](https://us1.discourse-cdn.com/elastic/original/3X/4/9/49802764612952dbccdb405fe9784040d237189b.png) And as you can see they both have no documents. ---
**Author:** ![cheiligers](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/cheiligers/32/73114_2.png) [@cheiligers](https://discuss.elastic.co/u/cheiligers)\ **Post date:** [July 27, 2022, 7:02pm UTC](https://discuss.elastic.co/t/kibana-stack-monitoring-all-nodes-reported-to-be-master/310549/4 "2022-07-27T19:02:19Z")
@Thuunder7 I've asked the monitoring team to try and help here. ---
**Author:** ![miltonhultgren](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/miltonhultgren/32/100401_2.png) [@miltonhultgren](https://discuss.elastic.co/u/miltonhultgren)\ **Post date:** [July 28, 2022, 6:24am UTC](https://discuss.elastic.co/t/kibana-stack-monitoring-all-nodes-reported-to-be-master/310549/5 "2022-07-28T06:24:26Z")
@Thuunder7 We've fixed these two bugs in Kibana 8.3! You can find the related PRs below: > > > fixes #126741 > > When using external collection, we were trying to read the Elas…ticsearch version from \`elasticsearch.version\`. > However, according to the \[module docs\](https://www.elastic.co/guide/en/beats/metricbeat/current/exported-fields-elasticsearch.html) the right field is \`elasticsearch.cluster.stats.state.version\` (which is aliased for \`cluster\_state.version\`). > > But, it seems the Metricbeat module does not actually populate this field, and it might be because the ES APIs aren't so clear about it. > > \`elasticsearch.version\` is populated from the results of \`/\_cluster/state\` which reports \`version\` as a simple integer. > > The response from \`/\_cluster/stats\` reports the version under both \`indices\` and \`nodes\`, both as lists. > > The response from \`/\` reports it under \`version.number\` but with the build type included (8.3.0-SNAPSHOT). > > The version shows up in a few places but not as clearly as the old internal collection documents. > As a side question, what is the version of a distributed system? A lot of the version fields in the new API response are arrays since it might be the state during a rollover to contain multiple versions. > > In this PR, I'm resorting to using the first item in \`elasticsearch.cluster.stats.nodes.versions\` as the version to display in the overview. > > > Fixes #126557 > > The Metricbeat documents always contain the field \`master\` but …the code only checked for existence of the field, not whether it was true or false. > > https://www.elastic.co/guide/en/beats/metricbeat/current/exported-fields-elasticsearch.html (search for elasticsearch.node.master) ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [August 25, 2022, 6:24am UTC](https://discuss.elastic.co/t/kibana-stack-monitoring-all-nodes-reported-to-be-master/310549/6 "2022-08-25T06:24:31Z")
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.