19 Feb 2021 03:39 AM
Hi,
I have configure the string in logs to be monitor as metric (log Event) and created the custom alert with threshold of 1 count for 1min to 3min default.
The above static threshold will alerts for each error string occurrence in logs?
Also once the metric is created and added to custom chart it will detect/display the historical (older) log errors?
Solved! Go to Solution.
22 Feb 2021 01:21 AM
Also do we need to need to select the particular log process in scope? if we not selected the scope then what will be the issue?
22 Feb 2021 07:23 AM
Hello @Mayana K.
The good practice is to limit the scope, and especially when you are using the calculated metrics.
Regards,
Babar
22 Feb 2021 07:41 AM
Actually the issue we are facing is, we have configure a log monitoring error with "X" parameter by creating a custom metric.
When ever the error occurs for X parameter, The dynatrace generate a problem with "X error" and "Y error" as well, but we don`t have any reference of "Y error" in logs.
The below steps has performed to configure a alert
But dynatrace sends a problem with both X and Y description
22 Feb 2021 08:52 AM
Hello @Mayana K.
Did you try to limit the scope?
Are both problems coming from the same logs?
Regards,
Babar
22 Feb 2021 08:57 AM
Yes limited the scope.
And both the error are generating in same problem and coming from the same logs.
23 Feb 2021 07:13 AM
We got the fix.
X error = 'system error failed'
Y error = 'system error failure'
I was using the query to filter the logs as 'system error failed' . This will matching the both X error and Y error.
X error = 'system error failed'
Y error = 'system error failure'
seems this is why we got the alert for both x and Y errors.
I have updated the query as "'system error failed'" then it exactly matching the parameter.
19 Feb 2021 09:58 AM
Concerning historic occurances: No, it counts from the time you defined the pattern.