Roman Gershman
Roman Gershman
Seems the problem on my side. Again!!! 🤦🏼♂️ Sorry for wasting your time. On a bright side, seems that 6.2 is working perfectly well!!!
I couldn't identify the root cause in my previous comment, but I have now. I'm pasting the example that reproduces the problem. It seems that another socket is needed to...
but I do not want to close `listen_fd`. It's not closed by design and it uses direct fd=0. I want to close another 10000 sockets that come afterwards and hold...
I am just happy that could be helpful this time! (even a broken clock is right twice a day)
Yeah, that won't fly. We have high frequency connections usecases, so I can not rely on that for sockets. But, it's fine - we have not been using direct fds...
@axboe has it be resolved by any chance?
The team is busy with other tasks. If someone would debug the issue on their side (operator logs etc) and submit the fix to the operator we would review such...
Can I review it @kostasrim ?
In general this PR is fairly big and it is hard to reason about design choices at this point. And also do a proper code review. I suggest to start...
Are you saying these issues are not a good fit for external contributors? I can paste here the contents of the issue if that helps.