Ceph : Issueshttps://tracker.ceph.com/https://tracker.ceph.com/favicon.ico2021-10-15T11:12:04ZCeph
Redmine Dashboard - Feature #52945 (New): mgr/dashboard: improve CephFS grafanahttps://tracker.ceph.com/issues/529452021-10-15T11:12:04ZErnesto Puerta
<p>Feedback from Ben England:<br /><pre>
MDS Performance shows absolute numbers, not rates. It should be showing reads/writes per sec and client requests/sec.
Additionally it should also display a graph of total files and total directories. This would allow people to determine average file size and average files/directory, important statistics in determining metadata density vs data.
...
Finally, I'd like to see a graph showing metadata cache hit rates. Without metadata cache hits, additional reads are required for each Ceph I/O operation.
</pre></p> Dashboard - Cleanup #48384 (New): mgr/dashboard: remove GRAFANA_API_URL and validation through ba...https://tracker.ceph.com/issues/483842020-11-26T18:59:08ZErnesto Puerta
There's this option <code>GRAFANA_API_URL</code> which is used for:
<ul>
<li>Pushing Grafana dashboards to Grafana. This is legacy/non-required functionality, as now Grafana Dashboards come embedded within the containerized Grafana</li>
<li>Allowing the front-end to detect whether Grafana is up and a specific Grafana Dashboard exists and provide some specific/fancy error messages (info boxes)</li>
</ul>
<p>Given the second behaviour is a little convoluted (for the front-end to guess whether it can ping the same URL an iframe is going to import, a REST API is triggered to the backend and returned to the frontend) and partially incorrect (the backend could reach out Grafana, while the front-end couldn't). An alternative would be to directly interact and guess Grafana status from the front-end via <code>iframe</code> (either by <code>onload()</code> or a similar approach).</p>
<p><img src="https://tracker.ceph.com/attachments/download/5267/Screenshot%20from%202020-11-26%2019-52-14.png" alt="" /></p> Dashboard - Bug #47997 (Resolved): mgr/dashboard: OSD disk performance statistics not working in ...https://tracker.ceph.com/issues/479972020-10-26T11:22:51ZRobert Groenewald
<p>When we want to view the disk statistics we get the following message :</p>
<pre>found duplicate series for the match group {device=\"dm-6\", instance=\"ceph_cluster\"} on the right hand-side of the operation: [{__name__=\"ceph_disk_occupation\", ceph_daemon=\"osd.41\", db_device=\"/dev/dm-33\", device=\"dm-6\", exported_instance=\"*OSD HOST NAME*\", instance=\"ceph_cluster\", job=\"ceph\"}, {__name__=\"ceph_disk_occupation\", ceph_daemon=\"osd.50\", db_device=\"/dev/dm-7\", device=\"dm-6\", exported_instance=\"imx-ceph-am11-osd-02\", instance=\"ceph_cluster\", job=\"ceph\"}];many-to-many matching not allowed: matching labels must be unique on one side</pre> Dashboard - Bug #47334 (Resolved): mgr/dashboard: many-to-many matching not allowed: matching lab...https://tracker.ceph.com/issues/473342020-09-07T10:27:18ZPatrick Seidensal
<p>In the "Host Details" Grafana dashboard, errors for example on "OSD Disk Performance Statistics" tab are seen:</p>
<p>many-to-many matching not allowed: matching labels must be unique on one side</p>
<pre>
Request details
Url api/datasources/proxy/1/api/v1/query_range?query=(irate(node_disk_writes_completed%7Binstance%3D~%22(neptun).*%22%7D%5B5m%5D)%20or%20irate(node_disk_writes_completed_total%7Binstance%3D~%22(neptun).*%22%7D%5B5m%5D))%20*%20on(instance%2C%20device)%20group_left(ceph_daemon)%20label_replace(label_replace(ceph_disk_occupation%2C%20%22device%22%2C%20%22%241%22%2C%20%22device%22%2C%20%22%2Fdev%2F(.*)%22)%2C%20%22instance%22%2C%20%22%241%22%2C%20%22instance%22%2C%20%22(%5B%5E%3A.%5D*).*%22)&start=1597239675&end=1597243290&step=15
Method GET
X-Grafana-Org-Id 1
Accept application/json, text/plain, */*
</pre> Dashboard - Bug #44784 (Resolved): mgr/dashboard: Some Grafana panels in Host overview, Host deta...https://tracker.ceph.com/issues/447842020-03-27T11:32:04ZKristoffer Gronlund
<p>Some Grafana dashboards are broken due to mismatches between FQDN and hostname in instance labels between the node exporter and the ceph exporter.<br />Others seem to have broken when devices can be created on LVM and not just raw devices.</p> Dashboard - Feature #43526 (New): mgr/dashboard: add grafana support by default to pre-defined roleshttps://tracker.ceph.com/issues/435262020-01-08T19:44:22ZErnesto Puerta
<p>Currently, pre-defined roles ({block,cephfs,rgw,...}-manager) don't display Grafana dashboards. It makes sense to enable them by default. Per-component permissions would also allow only to access Grafana dashboards for that specific component (nevertheless, Grafana dashboards can be easily accessed by anyone).</p> Dashboard - Feature #43101 (New): mgr/dashboard: Add a link to open the current embedded Grafana ...https://tracker.ceph.com/issues/431012019-12-03T10:30:46ZLenz Grimmer
<p>Sometimes it would be useful to be able to see a certain Grafana dashboard in it's "native" form, without being embedded in the Ceph Dashboard. It would be nice to add a link to the controls above the embedded Grafana dashboards named "Open Dashboard in Grafana", which opens that particular Grafana Dashboard on the actual Grafana host in a new browser tab/window.</p> Dashboard - Cleanup #41586 (New): mgr/dashboard: remove Grafana rolehttps://tracker.ceph.com/issues/415862019-08-30T09:16:04ZErnesto Puerta
As it's neither useful nor necessary:
<ul>
<li>Useless: roles are meant to control access, but Grafana service is external to dashboard, so there's no way to enforce these policies (any user can access Grafana dashboards independently).</li>
<li>Unnecessary: Grafana is just a infrastructure component, while roles are meant to control access to Ceph workflows (RBD, RGW, etc.). Hence every Grafana Dashboard should be hidden/displayed based on the specific Ceph workflow where it's embedded in.</li>
</ul>
<p>Related to <a class="external" href="https://github.com/ceph/ceph/pull/29989#issuecomment-526514531">https://github.com/ceph/ceph/pull/29989#issuecomment-526514531</a></p> Dashboard - Bug #40008 (Resolved): mgr/dashboard: Need a method to check references to Grafana da...https://tracker.ceph.com/issues/400082019-05-23T04:46:19ZKiefer Chang
<p>Ceph Dashboard embeds Grafana dashboards by specifying <strong>uid</strong> in <cd-grafana> components.<br />If a Grafana dashboard is updated with uid changed, uid property in corresponding <cd-grafana> component should also be updated.<br />We need a method to ensure these references are correct, or at least exist.</p>
<p>In the short term, a script can be added to check a <cd-grafana> always refer to a existing Grafana dashboard. <br />In the long term, we can add some e2e tests to verify correctness of embedded dashboards.</p> Dashboard - Feature #38662 (New): mgr/dashboard: Display helpful message if Grafana dashboards ar...https://tracker.ceph.com/issues/386622019-03-11T13:23:07ZLenz Grimmer
<p>From the <a href="http://docs.ceph.com/docs/master/mgr/dashboard/#enabling-the-embedding-of-grafana-dashboards" class="external">dashboard documentation</a>:</p>
<blockquote>
<p>Ceph Dashboard embeds the Grafana dashboards via iframe HTML elements. If Grafana is configured without SSL/TLS support, most browsers will block the embedding of insecure content into a secured web page, if the SSL support in the dashboard has been enabled (which is the default configuration). If you can’t see the embedded Grafana dashboards after enabling them as outlined above, check your browser’s documentation on how to unblock mixed content. Alternatively, consider enabling SSL/TLS support in Grafana.</p>
</blockquote>
<p>It would probably make sense for the dashboard to display a similar message, if the setting <code>grafana-api-url</code> begins with <code>http://</code> and SSL is enabled on the dashboard.<br />It should be possible to disable/hide this message, if the user can or does not want to change the Grafana SSL configuration (and disables the security precaution in the browser instead).</p>