Closed harryoooooooooo closed 1 year ago
Hi and thanks for reporting the bug with a good stack trace!
I've tried to fix it in the commit referenced above this commit, could you have a look and see that it resolves the issue (and does not cause other issues)? Thanks!
Hi diwic@, I've stress tested the patch in the past day and it worked fine. Great thanks for your quick fix!
Hi @diwic, could you tell when will be the next dbus-tokio release? That would be great if the fix above is included in any release tag. Thanks!
@harryoooooooooo 0.7.6 released now, enjoy! :-)
Hi diwic@, I'm Hsin-chen from Google.
We noticed a deadlock in our program. According to the stack trace (attached below), the scenario is like:
Crossroads::handle_message
on a new method is receivedIt's clear that in step 1 and step 4 the thread tries to lock the same mutex twice. I've inserted some logs and verified it.
For what it's worth, we also noticed that the deadlock usually happens when there is a really near (<5 us) outgoing method call.
Though we're not using the latest released version, it seems the call path is still valid on HEAD. Could you help take a look? Please let me know if you need more logs or help reproducing the issue. Thanks.