Hello,
Our Rocky 8.7 environment very relies on systemd-journal-upload and systemd-journal-remote for journald log central management. And we always met small part of systemd-journal-upload agent reports “systemd-journal-upload.service: Watchdog timeout (limit 3min)!” error, and even adjust Watchdog timeout to a longer time doesn’t help too much about the timeout issue. There is no network acl between systemd-journal-upload and system-journal-remote OS. And systemd-journal-upload.service and its
OS has no performance issue. So how can we debug more detail and see what processes block the service up after running for a short period?
Is there a reason why you’re still running Rocky Linux 8.7? It’s end of life and you are recommended to update systems to 8.10.
The screenshot you’ve provided show 6/ABRT and that core was dumped. Raising the timeout value isn’t going to solve this. I would update your system to 8.10 first to see if the issue reoccurs. If it does, then you will need to get the coredump and use gdb to get the backtrace.
Hello, here I updated one os to Rocky 8.10 and Watchdog timout of systemd-journal-upload service still exists. And I let make coredump for the Watchdog timeout and has below info. But I still have no idea about the root cause:
Current systemd-journal-upload and related rpm version for your reference.
