Race condition prevents sending messages

[woetenwondla]

Using a supported version?

• I have searched searched open and closed issues for duplicates.
• I am using Signal-Desktop as provided by the Signal team, not a 3rd-party package.

Overall summary

Sometimes I encounter, that signal desktop hangs, when sending messages. So the message seems to never leave the computer and there is just this small dotted circle icon on the message itself. I think it happens when my laptop returns to work from suspend mode and then trying to send a message, but unfortunately I could not manage to find any clear source for it.
The only way, to get the message sent, is when I kill both signal processes via taskmanager.
There I observed, that one process always uses around 30% cpu, where as the other one constantly uses 10-15% cpu.
Therefor I think it is some race condition, that prevents the app from continuing to send messages.

When killing it via the taskmanager an error pops up from the process using less cpu. Haven't got a picture of it, sorry for that. Although I think the error message is not really descriptive, if remember correctly.
When I then reopen the signal app, the message appears to be sent.

Steps to reproduce

Bug:

1. Probably wake up computer from sleep
2. Write a message and send it
3. It does not get sent and the message displays the dotted circle forever
   Workaround:
4. Kill the application via task manager
5. Message appears to be sent when reopening signal

Expected result

Message should be sent immediately as always

Actual result

Message is apparently not sent (not 100% sure, maybe it is displayed incorrectly) and the dotted circle appears forever.

Screenshots

I'll add them, when the bug reappears.

Signal version

7.27.0

Operating system

Kali linux 2024.3

Version of Signal on your phone

7.18.2

Link to debug log

Note: It happened around 2 minutes before the end of the logs:
https://debuglogs.org/desktop/7.27.0/06dee372ee05dc44c267146e9002d18c0210c61aec4c25197914eb64fd070e81.gz

[scottnonnenberg-signal]

This appears to be the same issue: #6103
And this: #6577

[woetenwondla]

Thank you for this hint. Seems like there is a workaround to use another screenlocker.
Anyway, I am wondering if there are any plans to fix this or if it is possible?

[trevor-signal]

Thank you for this hint. Seems like there is a workaround to use another screenlocker. Anyway, I am wondering if there are any plans to fix this or if it is possible?

We believe the issue is likely going to be needed to be fixed upstream, but we'd be very open to solutions if you're aware of any.

[woetenwondla]

Ah okey, good to know. Unfortunately at the moment I am not aware of any other solutions. If this should change,
I'll be glad to inform you.