Sitsofe Wheeler

Hi @jidanjianiunai, If we focus on your second job for now can you: - Remove every option that doesn't stop the problem from happening (e.g. I can tell that `--group_reporting`...

This one is strange. I can reproduce the issue 100% of the time with the following: ``` $ rm -f /tmp/fio.tmp; ./fio --randseed=100 --zero_buffers --verify=md5 --name=test --size=128k --bs=3k --filename=/tmp/fio.tmp --rw=randrw...

Hmm, I would guess this is because the file is not preallocated before the second job runs so it has to be grown "on demand". One way to test if...

> it is known that my filesystem XFS [...] effectively serializes length extending writes Yup, looks like you know your way around this at least as well as I do...

> Isn't the bug that the files are unlinked even though the run isn't done using them? The files are all created up front when the run starts, even if...

Hello @AshCrismon, I think the crux of the problem is that fio has always assumed that if you want to a verify of a previous job it must be done...

Hi @AshCrismon, I'm still interested to hear your feedback to my assessment - if you have the time please feel free to follow up!

Forlornly closing due to lack of reply from reporter. If this issue is still happening with the latest fio (see https://github.com/axboe/fio/releases to find out which version that is) please (please!)...

Hi @struschev, Phew that's a whole lot of files to have on a network filesystem and I can see how the `stat` overhead is painful, so yes we would love...

> I propose just ignoring the new option when the filename is specified. Counter-proposal - your new option acts on all files of the job regardless of how they were...