Open rhettg opened 10 years ago
I'd appreciate any suggestions on how to further debug this.
We're seeing bad frame reads, too but in /librabbitmq/__init__.py
, in drain_events
.
This looks very similar to #30
There is a bug at https://github.com/celery/librabbitmq/blob/master/Modules/_librabbitmq/connection.c#L1343 where an empty (0 byte) body always triggers an exception, even if it was deliberately sent as empty.
I have a patch that fixes this issue in my env in pull request #59.
Any news?
That patch #59 has been merged. Try upgrading to 1.6.x and see if you still have the issue.
I believe you missed this merge when releasing 1.6.1:
https://github.com/celery/librabbitmq/commits/v1.6.1 < contains no commit abbd949
Can you please release 1.6.2?
:+1: could you please release 1.6.2 ?
@ask could you release a new version with the fix for this?
I have the last version and still getting this Bad Frame Read issue, anyone fixed it ?
1.6.1 hasn't got this fix IIRC. It's in master and the 1.6 branch though, so checking out either should get around the problem.
Hello,
Any news about 1.6.2?
I get "bad frame read" exception in 1.6.1. maybe I should use other amqp lib instead of librabbitmq.
It's been a couple years since I issued a patch for this bug. https://github.com/celery/librabbitmq/pull/59#issuecomment-266296125
Just upgraded to librabbitmq 1.5.2 and celery 3.1.12
I'm getting a failure during one of my tasks that involves a large chord.
This seems like it might be related to #30