Support Questions
Find answers, ask questions, and share your expertise
Announcements
Alert: Welcome to the Unified Cloudera Community. Former HCC members be sure to read and learn how to activate your account here.

Report or alert based CPU and Memory usage

Highlighted

Report or alert based CPU and Memory usage

Contributor

Hello,

 

Is there any way we can configure Cloudera Manager to send an email if utilization of CPU or Memory of a particular daemon / node (for eg. Impala daemon) crosses a specific limit (for eg. 80%)?

 

Please advise.

 

Thanks

snm1523

1 REPLY 1

Re: Report or alert based CPU and Memory usage

Cloudera Employee

Hi,

Have you tried using mail triggers.Please go through the steps you will find a way to configure alerts for CPU usage based on triggers configured.

Mail alert triggers could be configured as seen below at [1].

For example for creating a custom trigger on Host CPU Usage you could follow these steps:
1. Login to Cloudera Manager
2. Navigate to Hosts
3. Select the host you would like to monitor and/or:
3.1. During trigger creation you can set it to affect all hosts if needed
4. Under "Details" on the left side, there is the "Health Tests" section
5. Select the "Create Trigger" button on the top-right of the "Health Tests" section
6. Name the Trigger
7. Edit the "Expression" to test: Last "cpu_percent" > 80 => Mark as concerning
8. Select if trigger should affect all hosts or only the host listed.
9. Verify the Chart on the "Preview" section on the right side of the page

Creating a custom trigger from a present Chart could be done with following Step 1-3 above, then:
4. In the "Charts" section on the right there are charts from which you can create triggers with the following steps:
4.1. Select the small gear icon in the top-right corner of the chart (appears on mouseover) e.g. "Host CPU Usage"
4.2. Select "Create Trigger" from the drop-down menu
4.3. You should modify the trigger as needed.
4.4. Select if trigger should affect all hosts or only the host listed.

 

[1]
http://www.cloudera.com/content/www/en-us/documentation/enterprise/5-14-x/topics/cm_dg_triggers.html

[2]
http://www.cloudera.com/content/www/en-us/documentation/enterprise/5-14-x/topics/cm_ag_alerts.html


setting type of Alert:

1. You don't need to restart anything for the triggers to work. (Not hosts, not CM, no restart needed.) When you created the trigger it should have by default set it to enabled.
2. Checking the host details you can also check all its triggers configured, if you suspect it's disabled or if you want to change the trigger later.
3. If you do not receive the alerts as expected you might need to review the following configuration settings:
3.1. Depending on how you set the trigger to change the health state: "Concerning or "Bad", you should set the alert threshold at:
Event Server configuration -> Health Alert Threshold -> Concerning OR Bad
(by default it doesn't alert on concerning, we are sorry if this caused a confusion)
3.2. If you still don't receive the alerts as intended, please check the following steps at [1].

 Enabling Alerts on triggers:
1. Login to Cloudera Manager
2. Navigate to Administration/Alerts
3. Select "Hosts" in "Alert Type" list on the left
4. Select the small "wrench" icon next to Hosts to edit the settings
5. Search for "Enable Health Alerts for This Host" and
6. enable the alerting on health tests