We are experiencing what seems to be a significant bug.
The inventory CSV upload job froze yesterday and now we cannot submit a new one because the API claims “job already processing”.
Indeed it is still “processing” although this Status output was taken today, i.e. this is the exact same result we had almost 24H ago:
Has anyone experienced that? This is the second time. We had to abort the first test job that froze for a week (we didn’t pay attention to the status).
This time we’ll let it as it is in case that helps for debugging.
Has anyone else experienced that?
Are we doing something wrong?
Is there a bug in the ShipHero backend?
Thank you for your help! This is a big blocker for our project.
Yes, I just confirm it’s the same issue our engineering team is trying to fix.
Hopefully we’ll have it resolved in the next few days, but I’ll let you know as soon as I have an update.
Thanks for the patience on this!
Tom
We are trying to determine what exactly is the issue occurring here, so that we can attempt to create a workaround.
We are aware that we are passing in inventory changes for some SKUs that do not exist in the particular Shiphero warehouse. Will your fix prevent the job from error/freeze in these cases? (i.e. skip/ignore SKUs we don’t have)
Is the issue that the job count is too high? Would reducing size and submitting multiple jobs help?
Any insight you can provide will be extremely helpful.
Hi @dan
Unfortunately, we don’t have a successful workaround at the moment because it’s not entirely related to invalid SKUs or the number of lines processed.
Making smaller uploads will help, but we can’t guarantee it will succeed all of the time.
I know this has gone on way too long, and we currently have an engineering team dedicated to fixing this.
I will let you know as soon as I have an update or possible workaround.
Yesterday we started a small inventory sync of only 1000 items (to avoid the known issue reported here) and this morning it was still in “pending” status. Sometime this morning it started and went to processing but with 0 in all the *_count fields. and it is still frozen at this stage. Hopefully it’s just a one of.
We have now a mechanism to give up after a certain amount of time but it’s still worrying.