Re: ShardSplitTest flakiness investigation

2023-04-11 Thread Ishan Chattopadhyaya
Hi Alex, Thank you for your investigation. I will be able to take a deeper look at it by Thursday or Friday. This fix is potentially very important for my client, Fullstory, so thank you for your work on this on their behalf too. Regards, Ishan On Mon, 10 Apr, 2023, 23:54 Alex Deparvu, wrote: >

Re: ShardSplitTest flakiness investigation

2023-04-10 Thread Alex Deparvu
Hi, Gentle reminder for anyone interested to take a look at the PR. I have added a lot of detail to this failure (and a fix proposal) and would appreciate some feedback, or at least let me know if you are planning on taking a look at some point in the near future. This would help both with

Re: ShardSplitTest flakiness investigation

2023-04-03 Thread Alex Deparvu
An update on the failing test. The failure comes from documents having null version field from the add operations running in parallel with the split shard operation. Digging into this a bit more (and running the test a very large number of times) I think I identified 3 problematic cases that can

Re: ShardSplitTest flakiness investigation

2023-03-30 Thread Kevin Risden
I don't have any helpful pointers as to why this might be happening - but I do want to say thanks for digging in and finding out the cause as well as finding some related old jiras. Its helpful to even just make small steps forward. Kevin Risden On Wed, Mar 29, 2023 at 1:21 PM Alex Deparvu

ShardSplitTest flakiness investigation

2023-03-29 Thread Alex Deparvu
Hi, Following David's recent email about the failing tests (and obviously not knowing any better) I started to look at the ShardSplitTest failures. The easy part was turning the current NPEs into an actual assertion. The hard part is that the documents seem to sometimes be missing the _version_