Closed timbussmann closed 4 years ago
ping @Particular/ravendb-persistence-maintainers
What’s interesting is that the failure occurs locally only, and not on TC
I ran the test over a dozen of times on my machine using the latest versions and I did not run into any issues. While walking through the timeout query code I saw though that every query that we issue can effectively run up to the query timeout that is defined on the database or server which by default is 300 seconds
https://ravendb.net/docs/article-page/4.2/csharp/server/configuration/database-configuration
Every query in the timeout seam including the GetNextChunk
could theoretically take up to 300 seconds and therefore hang/block for that specific amount of time. But like I said I wasn't able to reproduce it.
We are closing this for now. Should it come up again we have the necessary breadcrumbs here to further investigate.
I've noticed that the delayed delivery acceptance tests randomly fail when running them locally. I noticed this during switching over to the acceptance testing transport, but according to @mauroservienti it is also reproducible when using MSMQ transport.
The tests fail with a timeout, never delivering the delayed message. By adding some more logging, I was able to pinpoint the issue down to
GetCleanupChunk
which sometimes blocks when called initially, causing the timeout query to never complete.Here's the full Debug level log from such a failing test:
notice the part around:
The later entry is an additional log statement I added. There is another log statement which is added once
GetCleanupChunk
is done. It doesn't show up in the log file on failed tests.