# Alerting on OS metrics **URL:** \ **Category:** PMM 2.x\ **Created:** [April 8, 2022, 2:16pm UTC](https://forums.percona.com/t/alerting-on-os-metrics/15168 "2022-04-08T14:16:37Z")\ **Posts on this page:** 7\ **Page:** 1
**Author:** ![Mikem](https://avatars.discourse-cdn.com/v4/letter/m/41988e/32.png) [@Mikem](https://forums.percona.com/u/Mikem)\ **Post date:** [April 8, 2022, 2:16pm UTC](https://forums.percona.com/t/alerting-on-os-metrics/15168/1 "2022-04-08T14:16:37Z")
I have monitors for cluster metrics configured but want to understand how best to set up alerts for Operating System metrics like disk space and status of particular services. ---
**Author:** ![Vadim\_Yalovets](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/vadim_yalovets/32/43_2.png) [@Vadim\_Yalovets](https://forums.percona.com/u/Vadim_Yalovets)\ **Post date:** [April 8, 2022, 4:10pm UTC](https://forums.percona.com/t/alerting-on-os-metrics/15168/2 "2022-04-08T16:10:48Z")
Hi Mikem, You may find in our documentation [an instruction](https://www.percona.com/doc/percona-monitoring-and-management/2.x/using/alerting.html#add-an-alert-rule) for adding an alert rule. Also you will need to find a required metric and create a formula. e.g. node\_filesystem\_free\_bytes You may check some formulas examples in already provided rules. e.g. Node high CPU load ![Screenshot 2022-04-08 at 18.59.55](https://us1.discourse-cdn.com/flex019/uploads/percona1/original/2X/b/b3c2f06412fa3b394dc2e543b93d9c9fb5fbd596.png) ---
**Author:** ![Mikem](https://avatars.discourse-cdn.com/v4/letter/m/41988e/32.png) [@Mikem](https://forums.percona.com/u/Mikem)\ **Post date:** [April 8, 2022, 6:05pm UTC](https://forums.percona.com/t/alerting-on-os-metrics/15168/3 "2022-04-08T18:05:54Z")
Thank you. That is helpful. Can you point me to the documentation that outlines how/where to set an alert to a specific host? I am struggling with that a bit. Thank you. ---
**Author:** ![steve.hoffman](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/steve.hoffman/32/66_2.png) [@steve.hoffman](https://forums.percona.com/u/steve.hoffman)\ **Post date:** [April 8, 2022, 7:00pm UTC](https://forums.percona.com/t/alerting-on-os-metrics/15168/4 "2022-04-08T19:00:58Z")
you can filter in the query itself `node_filesystem_free_bytes{node_name="server.domain.com"} > 400000000000` ---
**Author:** ![Mikem](https://avatars.discourse-cdn.com/v4/letter/m/41988e/32.png) [@Mikem](https://forums.percona.com/u/Mikem)\ **Post date:** [April 8, 2022, 8:02pm UTC](https://forums.percona.com/t/alerting-on-os-metrics/15168/5 "2022-04-08T20:02:15Z")
I have some alerts configured like that but I am puzzled by one alert I set up months ago and do not remember the details. I created an alert using the pmm\_mysql\_down template but cannot find where I configured this association with the node. ![image](https://us1.discourse-cdn.com/flex019/uploads/percona1/original/2X/8/84dd7979d9e2c9879b77440b01557cf5457fb9db.png) ---
**Author:** ![steve.hoffman](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/steve.hoffman/32/66_2.png) [@steve.hoffman](https://forums.percona.com/u/steve.hoffman)\ **Post date:** [April 8, 2022, 8:08pm UTC](https://forums.percona.com/t/alerting-on-os-metrics/15168/6 "2022-04-08T20:08:39Z")
check the alert rule that generated that alert and see if you added a label filter using maybe service\_name? ---
**Author:** ![Mikem](https://avatars.discourse-cdn.com/v4/letter/m/41988e/32.png) [@Mikem](https://forums.percona.com/u/Mikem)\ **Post date:** [April 11, 2022, 2:01pm UTC](https://forums.percona.com/t/alerting-on-os-metrics/15168/7 "2022-04-11T14:01:08Z")
Steve, I don’t see a filter on the Alert Rule. Would this rule be tied to all hosts monitored by PMM? I am struggling to understand the relationship between Alert Rules and monitored hosts. I have configured alerts with Dashboard Panels using Database parameters, i.e. wsrep\_cluster\_size but that seems different.