Stuck on an issue?
Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.
And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.
Metric Filter configuration producing unexpected results
See original GitHub issueIssue Description
This is a Bug Report
Description
I am trying to configure a metric filter alarm with a given filter pattern for the function under my service,
custom:
alerts:
topics:
alarm:
topic: arn:aws:sns:eu-west-1:704255512929:CloudWatch-ErrorLogs
function:
- name: AnAlarm
metric: AnError
treatMissingData: notBreaching
threshold: 0
statistic: Sum
period: 60
evaluationPeriods: 1
comparisonOperator: GreaterThanThreshold
pattern: '[ERROR]'
I was expecting this to create a single metric filter (ERROR) linked a single alarm that is then linked to an SNS Topic (CloudWatch-ErrorLogs), however what is actually happening is 2 metric filters are being created, one with a suffix ALERT and another with a suffix OK. Both of these are then linked to the same metric and the same alert.
From reviewing the code it appears there is code specific to create 2 metric filters, one for the Alert and one for the OK and I wondered why this was the case? Would it be possible to support a way of configuring it to exclude the OK metric filters?
Additional Data
- Serverless Framework Version: 1.26.0:
Issue Analytics
- State:
- Created 5 years ago
- Reactions:5
- Comments:11
Top Results From Across the Web
Top Related Medium Post
Top Related StackOverflow Question
Troubleshoot Live Code
Top Related Reddit Thread
Top Related Hackernoon Post
Top Related Tweet
Top Related Dev.to Post
Top Related Hashnode Post
It still seems to be problem. Have we done the fix?
I should do that - I’ve just been lazy (and I’ve never made a pull request before so I have to figure out how to do that 😃. Let me see if I can get that together in the next week or so…