AEI (Hannover) issue tracker

Today

12:54 Atlas Hardware Repair #3783 (Assigned): a7401: high number of different ECC errors across the board

3c25 | 08/06/26 | 12:04:24 UTC | Memory | Correctable ECC | Asserted | CPU 1 DIMM 9
3c26 | 08/06/26 | 12:04:25 UTC |... Carsten Aulbert

2026-08-05

08:06 Atlas ToDo #3775 (Resolved): proxmox licenses

got these via email from Adam Carsten Aulbert

2026-08-04

14:12 Warranty / Broken parts Vendor/misc (out of warranty) #3782 (Assigned): a3411: GPU#4 broken

Unable to determine the device handle for GPU4: 0000:89:00.0: Unknown Error Carsten Aulbert

11:43 Warranty / Broken parts Vendor/megware #3781 (Resolved): g9922: GPU #11 broken

| 10 NVIDIA RTX 4000 Ada Gene... On | 00000000:8C:00.0 N/A | N/A |
|ERR! ERR! ERR! ... Carsten Aulbert

2026-08-03

07:43 Atlas Bug #3780 (Assigned): update labels on nodes

This morning, I accidentally disconnected power from @condorhub-mkgc@ without realizing this as the machine still has... Carsten Aulbert

07:38 Atlas Bug #3779 (Resolved): central laptop not working

I tried to get some reaction to the keyboard/mouse/monitors at the table in the aisle between rows 5 and 6 but to no ... Carsten Aulbert

2026-07-31

13:36 Atlas ToDo #3769 (Assigned): Remove Nvidia/Mellanox NIC from ark01/work59c

what do you mean node is not reponding, I'm currently logged in.
hostname -a
ark01
ip a
1: lo: <LOOPBACK,UP,L... Carsten Aulbert

2026-07-29

07:09 Atlas Hardware Repair #3778: gx232: Riser re-initialization

asking megware whether we ought to remove all NVMe and the three other GPU packs for this re-init. Carsten Aulbert

07:01 Atlas Hardware Repair #3778: gx232: Riser re-initialization

there are currently five DAGs depending on this node, thus likely not happening before Monday Carsten Aulbert

06:57 Atlas Hardware Repair #3778 (Assigned): gx232: Riser re-initialization

We got instructions from Gigabyte how we could try to re-init the riser, unfortunately this is quite involved:
> T... Carsten Aulbert