Recent content by czechsys

Based on the DMARC analyzators. Because PMG doesn't support changing domains for generated mails by weubi i am looking for the most easiest way to do it without extra configurations on other systems. Target: signing PMG generated emails ONLY Problem: PMG is currenty sending as...

Another bug - when you remove REALM with users mapped, and after it try to remove user: api error (status = 400: unknown realm sub.example.org) Removing user fails.

PDM 1.1.1 tests (based on "Allow changing an LDAP realm to use anonymous search."): domains.cfg: ad: sub.example.org base-dn dc=sub,dc=example,dc=org mode ldaps server1 fqdn1 server2 fqdn2 comment AD authentication user.cfg: user: USER@sub.example.org enable true...

Try this parameter in the nvidia apt repository configuration for SHA1 problem: Allow-Insecure: yes

Option: You can mount nfs via /etc/fstab and use storage -> directory variant on the PVE side.

No, we never used linux bridges for PVE.

There was 2 changes: 1] update PVE8 from 9/2025 (5 hosts) and 1/2026 (2 hosts) to current - to get cluster on the same version 2] added "ovs_options mcast_snooping_enable=false" to vmbr0 - due failing ipv6 in VMs after manual migration So if there isn't some new bug ... I will try...

Hi, the was problem with booting (some) VMs after pve server reboot, shared multipathed iscsi SAN. VM (vg-dorad-prod, bulk start after pve reboot): 2026-04-14T04:10:06.683761+02:00 pve-14 pvesh[3655]: Starting VM 576 2026-04-14T04:10:06.690330+02:00 pve-14 pve-guests[3721]: start VM 576...

Hi, currently are our VMs network dualstack mainly. Only very small number of them is ipv6 only. The problem is currently detected on the dualstack VMs on the PVE 8.4. When doing manual offline migration (gui or ha-manager), some migrated VM's ipv6 doesn't work after boot. When doing automatic...

Busy database with huge number of the cpu cores? Solution: 1] disable HT or 2] don't overallocate nonHT cores to VM+PVE The HT cores aren't for busy/fast/etc workloads, but for idle/low workloads mainly (so you can overallocate low power VMs).

Test it with live iso (debian, ubuntu),

Apples and oranges. You are comparing PRODUCT (based on the debian + ubuntu) with pure OS (debian).

PVE (9.1.1): domains.cfg: ad: SUB.EXAMPLE.ORG comment AD authentication domain sub.example.org server1 fqdn1 base_dn DC=sub,DC=example,DC=org case-sensitive 1 default 1 mode ldaps server2 fqdn2 user.cfg...

AD realm is created manually in domains.cfg. This works Editing via gui "Sync options" failing. "Sync" via gui Realms failing - unedited error message: 2025-12-10T10:53:48+01:00: this is a DRY RUN - changes will not be persisted 2025-12-10T10:53:48+01:00: TASK ERROR: LDAP operation result: rc=1...

Any option to have "cluster name" or alias for cluster in remotes? Having cluster hidden under hostname is little confusing...