-
Notifications
You must be signed in to change notification settings - Fork 850
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Fix deadlock when erroring writes are slow #16937
Fix deadlock when erroring writes are slow #16937
Conversation
Thank you for your contribution MasslessParticle! We will review the pull request and get back to you soon. |
46ad629
to
9793e6f
Compare
@mohsha-msft @seankane-msft Hello all! Is there any word on this PR? This fixes a goroutine leak that brings down our containers so we're excited to merge a fix and be off a fork. Let me know if I can help! |
Hey @MasslessParticle , Thanks a lot for your contribution to our repo. |
Thanks for your reply! No need to merge to a dev branch. I'm excited for the |
Hi @MasslessParticle. Thank you for your interest in helping to improve the Azure SDK experience and for your contribution. We've noticed that there hasn't been recent engagement on this pull request. If this is still an active work stream, please let us know by pushing some changes or leaving a comment. Otherwise, we'll close this out in 7 days. |
@mohsha-msft is there still interest in accepting this PR? |
Hi @MasslessParticle. Thank you for your interest in helping to improve the Azure SDK experience and for your contribution. We've noticed that there hasn't been recent engagement on this pull request. If this is still an active work stream, please let us know by pushing some changes or leaving a comment. Otherwise, we'll close this out in 7 days. |
@jhendrixMSFT @mohsha-msft Is there still interest in this PR? |
a204a13
to
2a3335d
Compare
2a3335d
to
26961ea
Compare
c.getErr
can return before slow writes complete. If there is more than one write and those writes result in errors, the second write blocks on a fullerrCh
. The result is a deadlockedcopyFromReader
call because the wait groups never clear.The test included tests demonstrates the deadlock.
This PR causes
close
to simultaneously drain theerrCh
while it waits for writes to complete. It combines any errors that are returned.I looked at the CHANGELOG, but not sure this change fits. Let me know if I need to do something else.