Hi @the_MiCrO
thanks for posting on the forum!
Could you please provide:
the output of the trace from the tracking center for this email
pmgversion -v
pmgdb dump
Best regards
Jonas
on a hunch - please compare the maximal message size (in the Mail Proxy Options) with the Maximal Spam Detector size - if the second is lower the issue is that the mail is not being shown to spamassassin completely which explains the modified...
Your blocklist object lists:
* '.*@warsawexpo.e.*` as regex
* warsawaexpo.eu as domain
the mail's envelope sender is: 'sm.21177883674.m9f0c8k9bh6zheoe05-zaproszenie=warsawexpo.eu@emsgrid.com'
and does not match your object (there is no @ before...
might indeed be the cause - check your configured LDAP profiles in the gui (and in /etc/pmg/ldap.conf) - is everything in place as it should be?
EDIT: I just also checked your term.log - this one does state that the upgrade ran into errors:
so...
Could you please share:
* the journal since the upgrade ran
* /var/log/apt/history.log (from the time of the upgrade)
* /var/log/apt/term.log (from the time of the upgrade)
* `ls -laR /etc/pmg`
* `ls -laR /var/lib/pmg`
Thanks!
please do not double-post as this just splits the discussion into 2 places - posting a link here to the thread in the PVE forum for reference - and closing this one...
Thanks for the report - and quite well to know that most things seem to run okay with this profile...
We do plan to evaluate and change the use of /tmp in general in PMG - and move most things to /run where sensibly possible - so that should fix...
Finally, after 7.0 arrived in the enterprise repos, I upgraded all hosts (after testing one) and removed the pinning of 6.14 kernel.
Ceph works, no OSDs out - at least for my hardware the issue seems to be solved.
Hi @tom.sanchez
thank you for posting on the forum!
Sourcing from the upgrade guide of Proxmox Mail Gateway 8 to 9 [1]:
So you can safely use the option "Keep the local version currently installed"
Yours sincerely
Jonas
[1]...
Hello Mr.Ivanov,
thx for yout time.
1.
root@lenovo:~# pveversion -v
Linux lenovo 6.17.2-1-pve #1 SMP PREEMPT_DYNAMIC PMX 6.17.2-1 (2025-10-21T11:55Z) x86_64
The programs included with the Debian GNU/Linux system are free software;
the exact...
Thanks for chiming in here! Very much appreciated!
The 2 locations are probably indeed a oversight from me - Thanks for noticing! - I sent a patch:
https://lore.proxmox.com/pve-devel/20260623102957.246597-1-s.ivanov@proxmox.com/T/#u
memtest...
I sent a patch to add the override for abstractions/consoles to the pve-devel list:
https://lore.proxmox.com/pve-devel/20260622164522.836988-1-s.ivanov@proxmox.com/T/#u
A colleague managed to reproduce the issue locally - I failed because I checked inside tmux or via ssh :)
It seems the issue is with `/etc/apparmor.d/dig` (or rather the included `/etc/apparmor.d/abstractions/console`)
Could you try the...
Standard SysAdmin best practises from my perspective:
- tiny replaceable OS/root partitions
- Data storage to be separate from root/os storage
-- then moving it between faster(nvme zfs) vs redundant (read ceph) vs cheaper (HDD zfs) depending on...
Hm - could not reproduce dig not providing any output in a container that I installed fresh from the ubuntu 26.04 template (guessed the version based on `1:9.20.18-1ubuntu2.1` version for bind9-dnsutils) - here it produces the root-nameservers...
Thanks for digging into this, and the pointers to the OVH documentation!
I was not aware that there is a way to explicitly tell which boot-entry to load - so that seems like a workaround if you want to keep the option to boot into rescue mode...
Hello,
I encountered the same problem after updating to Proxmox 9.2, whereas in Proxmox 9.1 the issue doesn't occur.
Since this happened on two different machines, with two mothercard series, I used IPMI to try to figure out what rEFINd was...
No need to apologize! - I do see that it's a bit harder to read - I think I never noticed/wasn't bothered by it, as it looks fine in the template itself.
I'll see if I can adapt the template to add linkebreaks there (without adding tons of...