Hello,
Le 13/09/2020 à 12:50, Mike Gerber a écrit :
Hi, I have a number of Debian hosts that have these ever-growing fail2ban databases. I tried to solve the problem by updating to 0.11.1-2 (the bullseye version). Observations: * Disk space usage DOUBLES by a backup copy of the database on upgrade: total 574M -rw------- 1 root root 300M Sep 13 12:43 fail2ban.sqlite3 -rw------- 1 root root 274M Sep 13 12:31 fail2ban.sqlite3.20200913-103150 * The database seems to be purged, observeable by using the sqlite3 binary to open the database and querying SELECT COUNT(*) FROM bans. This seems to happen every hour, so you have to be a little patient. * It is however not VACUUMed, so it stays at its size. So an update is definitely not enough. Deleting the backup copy and manually VACUUMing (an hour after the update) is required too.
You should probably report that upstream. And maybe not adding more info to this bug (as it is marked as closed)
Thanks S _______________________________________________ Python-modules-team mailing list Python-modules-team@alioth-lists.debian.net https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/python-modules-team
