systemd-journald does not resume automatically after failed rotation (no free disk space) #3968
Closed
1 of 2 tasks
Labels
Comments
This was referenced Aug 16, 2016
poettering
pushed a commit
that referenced
this issue
Aug 17, 2016
If journals get into a closed state like when rotate fails due to ENOSPC, when space is made available it currently goes unnoticed leaving the journals in a closed state indefinitely. By calling system_journal_open() on entry to find_journal() we ensure the journal has been opened/created if possible. Also moved system_journal_open() up to after open_journal(), before find_journal(). Fixes #3968
|
@poettering , can this issue be re-opened, as the original correction did not correct totally the situation? |
|
@poettering I am seeing this issue on Arch Linux with latest updates. |
jsynacek
pushed a commit
to jsynacek/systemd-rhel
that referenced
this issue
Oct 10, 2017
If journals get into a closed state like when rotate fails due to ENOSPC, when space is made available it currently goes unnoticed leaving the journals in a closed state indefinitely. By calling system_journal_open() on entry to find_journal() we ensure the journal has been opened/created if possible. Also moved system_journal_open() up to after open_journal(), before find_journal(). Fixes systemd/systemd#3968 (cherry picked from commit 105bdb4) Resolves: #1493846
jsynacek
pushed a commit
to jsynacek/systemd-rhel
that referenced
this issue
Oct 19, 2017
If journals get into a closed state like when rotate fails due to ENOSPC, when space is made available it currently goes unnoticed leaving the journals in a closed state indefinitely. By calling system_journal_open() on entry to find_journal() we ensure the journal has been opened/created if possible. Also moved system_journal_open() up to after open_journal(), before find_journal(). Fixes systemd/systemd#3968 (cherry picked from commit 105bdb4) Resolves: #1493846
lnykryn
pushed a commit
to lnykryn/systemd-rhel
that referenced
this issue
Oct 30, 2017
If journals get into a closed state like when rotate fails due to ENOSPC, when space is made available it currently goes unnoticed leaving the journals in a closed state indefinitely. By calling system_journal_open() on entry to find_journal() we ensure the journal has been opened/created if possible. Also moved system_journal_open() up to after open_journal(), before find_journal(). Fixes systemd/systemd#3968 (cherry picked from commit 105bdb4) Resolves: #1493846
caobinxin
pushed a commit
to caobinxin/systemd-lx
that referenced
this issue
Oct 19, 2019
If journals get into a closed state like when rotate fails due to ENOSPC, when space is made available it currently goes unnoticed leaving the journals in a closed state indefinitely. By calling system_journal_open() on entry to find_journal() we ensure the journal has been opened/created if possible. Also moved system_journal_open() up to after open_journal(), before find_journal(). Fixes systemd/systemd#3968 (cherry picked from commit 105bdb46b4ac7eb658a2f27727216591d0bfe267) Resolves: #1493846
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Submission type
NOTE: Do not submit anything other than bug reports or RFEs via the issue tracker!
systemd version the issue has been seen with
231
NOTE: Do not submit bug reports about anything but the two most recently released systemd versions upstream!
Used distribution
Fedora 24 + systemd 231
In case of bug report: Expected behaviour you didn't see
When root partion is full, journal log rotation fails, and no new logs are seen in "journal -f" output. When disk space is freed, journal logging shuold resume without user restartingservices.
In case of bug report: Unexpected behaviour you saw
journal logging did not resume when disk space was freed. Only by giving service restart it resumed:
systemctl restart systemd-journald
In case of bug report: Steps to reproduce the problem
rm /var/log/journal/*/*systemctl restart systemd-journaldfill the root file system with dd command so that we get error "no space left on device"
# dd if=/dev/zero of=/var/log/journal/fillfile.bin bs=1M count=6244start logging continuously to journal:
while [ 1 == 1 ]; do logger -p 3 "testlogging to fill journal current file" ; sleep 0.01; done;wait that journal rotation happens, journal -f stops taking new logs, and dmesg output shows error:
4990.275879] systemd-journald[21722]: Failed to create new system journal: No space left on devicedelete the fillfile:
rm /var/log/journal/fillfile.bincheck the "journal -f" command again, no new logs are seen in journal output
The text was updated successfully, but these errors were encountered: