I followed this tutorial for very long time and it's working fine.
Until now received this error from sa-update cronjob after updated to PMG 7.3 that shipped along with SpamAssassin 4.0
plugin: failed to parse plugin (from @INC): Can't locate...
PVE 7.1-6 latest updated.
Automatic guest-fstrim for Ubuntu KVM not working after migrate to another node or move disk. Discard option is checked. Qemu-guest-agent is installed and running.
However don’t found this problem for CentOS7-8 KVM.
I have to manually shutdown/start Ubuntu KVM once...
I can't paste the complete content of /etc/pmg/templates/postgresql.conf.in as it has more than 15000 chars but it's equivalent of size and content to /etc/pmg/templates/postgresql.conf.in.ucf-dist and also /var/lib/pmg/templates/postgresql.conf.in
First page of its content.
[% TAGS <* *> -%]
Hi @Stoiko Ivanov
I can now fixed this problem by using template and set it to version 13 and it works even after several rebooted. Version doesn't change back to 11 anymore! Thanks for your help :)
Below are answers for your questions.
I just rebooted the server, so process has changed as...
Yes I followed the guide.
pg_dropcluster --stop 13 main
pg_upgradecluster -v 13 11 main
systemctl unmask postfix pmg-smtp-filter pmgpolicy pmgdaemon pmgproxy
apt purge postgresql-11 postgresql-client-11
and optional "apt autoremove" as apt recommended.
it will work as long...
No, never touch postgres config file before.
Even “pmgconfig dump” shows “postgres.version = 13”
Weird thing after successfully upgraded is systemctl shows three service files
After upgraded by following this guide https://pmg.proxmox.com/wiki/index.php/Upgrade_from_6.x_to_7.0
PMG is working fine but after rebooting once, Postgresql failed to start as its config file /etc/postgresql/13/main/postgresql.conf version number changed from 13 to 11 on all lines.
- Will LXC (centos7) be able to boot successfully after upgraded to PVE7 without modifying grub CMDLINE?
If yes, what will happen on this LXC? Its functions still working as before or any function will be lost/strict?
Or modifying grub is the only way to keep it booting?
Does PMG deletes e-mail from /var/spool/pmg/cluster/* if admin chose to deliver false-positive message from quarantine page to receiver already?
I usually use sa-learn command to improve bayes by learning spam messages from /var/spool/pmg/cluster/ as it contains definitely spam and virus...
Related to ClamAV daily update error below.
Setting "ScriptedUpdates yes" is fixed this issue as well. After two weeks outdated ClamAV signature without any notice.
ClamAV update process started at Mon Feb 24 23:15:06 2020
WARNING: Your ClamAV installation is OUTDATED!
Actually I have no idea about this. I don't mean to use anything other than PMG using but it shows me like that in the log.
Is this normal process after restarting the server?
# ps axu|grep spam
spamass+ 142 0.0 0.0 104072 204 ? Ssl 18:52 0:00...
Why spam message like this doesn't show on Tracking Center? but receives "5.7.1 Blocked by SpamAssassin" instead.
Feb 12 16:57:33 mxgw spamd: spamd: identified spam (26.5/5.0) for user:111 in 5.6 seconds, 3891 bytes.
Feb 12 16:57:33 mxgw spamd: spamd: result: Y 26 -...
Yes me too.
Seems we have to edit /etc/cron.daily/logrotate
And insert an option ‘ -f ‘ to the line
/usr/sbin/logrotate -f /etc/logrotate.conf
Moreover, I have to comment out this condition checking at the top of this file also. Because on my system there’s that ‘system’ directory exists...
I have tried this too by running logrotate script from cron.daily directory but no logs getting rotated because there’s this condition at the top and that directory actually exists. So cronjob skipped.
# skip in favour of systemd timer
if [ -d /run/systemd/system ]; then
I'm sure I tried on the Firefox Private tab and also Chrome Incognito tab. They are the same.
I have using PMG about two years and this just happen few days ago. Nothing touched on logrotate config.
And yes, /var/log/syslog contains the mail log and its size is getting big.
Tracking center is showing "No data in database" no matter what date period is selected. Other statistic pages are showing as normal. Rebooting doesn't help.
This happened after upgrading some packages below.