CloudWatch api "stalling"

[pwmcity]

After installing the aws TA, i was able to collect some cloudwatch data, but all I get now is this:

/var/log/splunk/aws_cloudwatch.log

    ...
   2015-05-22 16:27:15,112 INFO pid=4461 tid=QueryWorkerThread-2 file=aws_cloudwatch.py:_main_work_loop:582 | Queried 631 metrics with 631 failures totaling 0 statistics in 13.279s before stalling.
    2015-05-22 16:27:15,122 INFO pid=4461 tid=QueryWorkerThread-1 file=aws_cloudwatch.py:_main_work_loop:582 | Queried 638 metrics with 638 failures totaling 0 statistics in 13.282s before stalling.
    2015-05-22 16:27:15,137 INFO pid=4461 tid=QueryWorkerThread-4 file=aws_cloudwatch.py:_main_work_loop:582 | Queried 538 metrics with 538 failures totaling 0 statistics in 13.307s before stalling.
    2015-05-22 16:27:15,139 INFO pid=4461 tid=QueryWorkerThread-3 file=aws_cloudwatch.py:_main_work_loop:582 | Queried 601 metrics with 601 failures totaling 0 statistics in 13.304s before stalling.

Ideas?

[klowknowz · Accepted solution]

Its actually a bad error description. If you put it in debug you will see:

'No Query Result, put back to wait queue'

That is what it triggered the failure. This is not a failure just that data was not returned for the period stated. Should be more like a warning or "No Data found".

[jcoates_splunk]

well, the data is failing to be a recognizable CloudWatch message stating there is a metric to download...

[jcoates_splunk]

hi, could you open a support ticket please?