When there are no hits, i get: UNKNOWN: No metric found (Are the filters properly set?)
Can't find any option to get OK instead of UNKNOWN when the filter is empty.
Page 1 / 1
Hi,
Did you try --zeroed option ?
Hi,
Same result…
Hi,
I have the same problem. The --zeroed option doesn't seem to work. I have the latest version :
Hi,
the --zeroed option should not be the solution. I guess there has been a change in Azure’s API that require the plugin to be updated.
Can you send the output when using the --debug option? Warning: sensitive data may be present, make sure you’re not sharing private IP addresses or anything.
The output :
UNKNOWN: No metric found (Are the filters properly set?) == Info: Uses proxy env variable no_proxy == 'localhost, 127.0.0.1, XXXXXXXX' == Info: Trying xxxx:xxxx... == Info: Connected to xxxxx (xxxxx) port xxxx (#0) == Info: allocate connect buffer! == Info: Establish HTTP proxy tunnel to management.azure.com:443 => Send header: CONNECT management.azure.com:443 HTTP/1.1 Host: management.azure.com:443 Proxy-Connection: Keep-Alive
@Mike there seems to be no average value in the returned JSON:
n { "id": "/subscriptions/xxxxxxxx/resourceGroups/xxxxxx/providers/Microsoft.CognitiveServices/accounts/xxxxx/providers/Microsoft.Insights/metrics/Latency", "type": "Microsoft.Insights/metrics", "name": { "value": "Latency", "localizedValue": "Latency" }, "displayDescription": "Latency in milliseconds. Do not use for Azure OpenAI service.", "unit": "MilliSeconds", "timeseries": " { "metadatavalues": "], "data": " { "timeStamp": "2024-11-19T15:11:00Z" }, { "timeStamp": "2024-11-19T15:26:00Z" } ] } ], "errorCode": "Success" } ]
Maybe the resource on the base of which the data is produced was stopped or unavailable between 15:11 and 15:26?
Maybe the resource on the base of which the data is produced was stopped or unavailable between 15:11 and 15:26?
That's what I'm having trouble with, the resource stops, but that's normal operation. Sometimes I get more than a hundred alerts in unknown state when everything is fine.
What do you mean by “that's normal operation”?
If there are planned outages, then you should either customize the check period for these services, or declared scheduled downtimes.
The “resources” switch off when they are not being used, which is normal.... but unpredictable.