Hello,
The status page of rspam has reset and is showing zero learned messages.
The uptime reported by the rspam status page corresponds with a restart of the VM host, so I suspect my NethServer VM wasn’t shutdown properly and that somehow caused rspam to reinitialize.
Is there anyway to diagnose what might have happened?
everythings is stored under a redis database that you can find at /var/lib/redis/rspamd/ if something occured bad like a shutdown maybe rspamd started a new redis databse (if the former was corrupted)
@mrmarkuz
[root@nethserver ~]# rpm -qa rspamd
rspamd-2.5-156.x86_64
I’ve updated rspamd many times and never had the statistics page reset.
@stephdl
[root@nethserver rspamd]# cd /var/lib/redis/rspamd/
[root@nethserver rspamd]# ls -lh
total 17M
-rw-r–r-- 1 redis mail 8.2M Feb 20 10:58 dump.rdb
-rw-r–r-- 1 redis mail 8.2M Feb 7 20:45 temp-739.rdb
The date of Feb 7 20:45 matches up with the VM host reboot. Is the file temp-739.rdb an old database?
The only other time I can remember the rspamd statistics resetting themselves was after I restored my NethServer VM from a ZFS snapshot almost 1.5years ago.
ok we found so why you stats has been restarted from scratch. I bet it is better to accept to play with the new db of redis because you have a gap of time now between the two databases.
For anyone else that comes across this thread in the future,
please can I confirm that the cause of the stats being reset was a corrupted redis database?
no ideas yet, however you have always a risk when you alter a databse. Maybe stop rspamd redis-rspamd, remove the database, restart rspamd is less risky