Bug#967945: systemd is way too aggressive in deleting journals of yesterday

[Harald Dunkel]

Package: systemd
Version: 241-7~deb10u4

Everytime I look into the status of some failed service there is
a message saying

Warning: Journal has been rotated since unit was started. Log output is incomplete or unavailable.

How comes? Disk space is cheap. And I didn't mess with the
defaults:

# egrep -v ^\# /etc/systemd/system.conf

[Manager]
# egrep -v ^\# /etc/systemd/journald.conf

[Journal]
# df -h /var/log/journal/
Filesystem      Size  Used Avail Use% Mounted on
/dev/drbd1       11T  6.3T  4.1T  61% /
# ls -al /var/log/journal
total 56
drwxr-sr-x+  3 root systemd-journal  4096 Jan 13  2016 .
drwxr-xr-x  20 root root            12288 Aug  4 00:00 ..
drwxr-sr-x+  2 root systemd-journal 36864 Aug  4 08:38 42b7d4373eff9fb16ebd59084afbff3f
#
# ls -al /dev/pts
total 0
drwxr-xr-x 2 root root      0 Jul 18 10:18 .
drwxr-xr-x 6 root root    520 Jul 18 10:18 ..
crw--w---- 1 root tty  136, 0 Jul 18 10:18 0
crw--w---- 1 root tty  136, 1 Jul 18 10:18 1
crw--w---- 1 root tty  136, 2 Jul 18 10:18 2
crw--w---- 1 root tty  136, 3 Jul 18 10:18 3
crw------- 1 root tty  136, 4 Aug  4 09:27 4
crw--w---- 1 root tty  136, 5 Aug  4 09:59 5
crw-rw-rw- 1 root root   5, 2 Aug  4 09:59 ptmx


According to the specs this should give me 4 GByte journal.


There was no such problem for stretch, AFAIR. I would strongly
recommend to keep logs *by default* at least for a week, unless
they don't exceed 2 GByte.


Regards
Harri