Closed bkirz closed 11 years ago
Is there a plan for merging this? This and the related PR in Qless would be helpful for launch.
I don't think it's ready yet. I mentioned this to @benkirzhner in person and forgot to mention it, but one problem here is that the failure message is only preserved if the client passes the message through to retry
. This is reasonable when the job is in the running state (and qless does not yet have the failure message) but in the common case where the job is in the failed state (and already has the failed message) and the job is retried on the Qless UI, I think qless-core should take care of preserving the failure message, since it already has it.
Sorry for all the confusion / headache / bad timing. This functionality has been ported into master
now. Great idea, by the way :-)
It would also be nice for the last failure to be preserved when the retry button is used on the Qless UI. In this case, the UI isn't passing the failure message through, but rather, the failure message is already in redis, and should be preserved.
Also, it would be really nice if the job history included the failure type (the full message would be bloated, I think). This would make it easy to see if the final failure was the same as all of the previous ones, or if there were different failures.