You are not logged in.
Have you tried limiting your journal size, see below.
Thx, followed your advice and set it to:
SystemMaxUse=50M
SystemMaxFileSize=10M
But I still lose 9 seconds:
systemd-journald[259]: Time spent on flushing to /var is 9.616797s for 720 entries.
Is it normal for 720 entries to waste 9 seconds?
Is this due to the fact that I'm using a HDD instead of an SSD?
Failure is success in progress.
A.E.
Offline
You could strace "journalctl --flush", but the behavior is maybe skewed by the "cinnamon-session[553]: DEBUG" stuff (causing many entries in the session journal)?
Offline
strace did not provide me with extra info, but it seems logical as the command "journalctl --flush" took less than 1 second to finish...
So it's strange that it takes definitely longer to execute during the boot process.
Anyway I removed the files under "/var/log/journal" and now the problem is gone
I kept my earlier "journald.conf" content
SystemMaxUse=50M
SystemMaxFileSize=10M
Now if I run "journalclt --disk-usage" the output shows around 8MB
Last edited by chrisdb (2019-02-19 10:25:04)
Failure is success in progress.
A.E.
Offline
ftr, you'd have had to skip the first flush. Otherwise the following flushes would oc. take less time since there's not as much to flush left.
Offline
ftr, you'd have had to skip the first flush. Otherwise the following flushes would oc. take less time since there's not as much to flush left.
I did mask the service, isn't that enough to stop the flush during boot?
Failure is success in progress.
A.E.
Offline
Usually - did the flush still show up in the log/blames/critical-chain?
Otherwise: is /var maybe on an extra (late) partition?
Offline
Usually - did the flush still show up in the log/blames/critical-chain?
Otherwise: is /var maybe on an extra (late) partition?
No, it didn't show up
I only have 1 partition.
Failure is success in progress.
A.E.
Offline