LHC@home

I'm getting alot of CMS Errors ~30% :

<message>
The global filename characters, * or ?, are entered incorrectly or too many global filename characters are specified.
 (0xd0) - exit code 208 (0xd0)</message>

In the last day or so, I was beginning to think I had some configuration issues to blame when all of my CMS and most all of my Atlas tasks were failing, but it seems
to be something more of a systemic problem. My CMS tasks are running under Virtualbox getting the 'glidein exited with return value 1" error and the Atlas tasks are
running native getting validation errors.

Hopefully things get better. I just returned to LHC after about a 5 year hiatus.

I am getting the same, CMS all started failing this morning

This CMS problem was starting around thursday June 26th 2025, 11:51:04 UTC

same problem here; I neglected to check regularly within the past 1-1/2 days - so tons of failed CMS tasks on all my hosts :-(

and of course it would be great if the distribution of tasks was stopped until the problem will be solved (probably not before next week)

Yep, all failing on my intel iMac, and it takes 20mn of "run" before failing so it's quite a waste of cycles.

At the moment I'm getting some atlas tasks, so it's fine.

[oh no it's not fine, I discover that all the atlas are in "validate error" status and all the hosts of the WU have the same...]

upto 80% fails now

After sorting out the lack of virtualisation on my PC (not turning it on in the BIOS....) it got a few tasks that got through the "no virtualisation" few seconds and ran for about 20 minutes then failed. All these "running" tasks appeared to have the same error - right a the top of Stderr output:

<core_client_version>8.0.2</core_client_version>
<![CDATA[
<message>
The global filename characters, * or ?, are entered incorrectly or too many global filename characters are specified.
(0xd0) - exit code 208 (0xd0)</message>
<stderr_txt>
2025-07-03 08:38:07 (3204): vboxwrapper version 26210
2025-07-03 08:38:07 (3204): BOINC client version: 8.0.2
2025-07-03 08:38:07 (3204): Detected: VirtualBox VboxManage Interface (Version: 7.0.14)
2025-07-03 08:38:08 (3204): Detected: Heartbeat check (file: 'heartbeat' every 1200.000000 seconds)
2025-07-03 08:38:08 (3204): Successfully copied 'init_data.xml' to the shared directory.

The rest of the stderr output looks "fairly normal"....

Hmm, that's a strange one?

Help?!?!?!?!?

[edit to add]
I've set No New Tasks.

this problem has existed for about 1 week now, and I am surprised that no one over there has found out yet, and that CMS tasks are still being disbributed. Too bad :-(

this problem has existed for about 1 week now, and I am surprised that no one over there has found out yet, and that CMS tasks are still being disbributed. Too bad :-(

Sorry for the lack of communication, combination of many factors. However, the good news is that we've finally tracked down the bug that was stopping jobs from running and the "Running Jobs" plot is starting to tick upwards again. Nice to have something fixed rather than broken on a Friday afternoon for a change!

---