LHC@home

I've applied your fix, although your command didn't work for me. I created the file manually and moved it as root to the correct folder.
My Atlas native now works fine, so thank you for your help.
Now, if only I can find out why Theory native doesn't work on this machine ...

Since 0:00 UTC, Atlas-Tasks running without doing anything. 0.00% CPU-using.

Since 0:00 UTC, Atlas-Tasks running without doing anything. 0.00% CPU-using.

Meanwhile resolved or local problem.
To test this I loaded 5 tasks and after 20 minutes initializing time of the VM's, all five happily processing events.

Ok, thanks, sorting them out. Maybe Squid... or other reasons.

Since 0:00 UTC, Atlas-Tasks running without doing anything. 0.00% CPU-using.

Meanwhile resolved or local problem.
To test this I loaded 5 tasks and after 20 minutes initializing time of the VM's, all five happily processing events.

but the problem of failing downloads still exists :-(

...but the problem of failing downloads still exists :-(

Yeah, there must be something awful rotten in the database for several ATLAS-tasks.
Even new tasks created last night suffer from this checksum issue and more painful after the return keep the tasks in progress and resent them again as 'lost' tasks.

...but the problem of failing downloads still exists :-(

Yeah, there must be something awful rotten in the database for several ATLAS-tasks.
Even new tasks created last night suffer from this checksum issue and more painful after the return keep the tasks in progress and resent them again as 'lost' tasks.

I am wondering how long it will take to get all this straightened out

We do have a new server running the ATLAS assimilator and validator, but not have any changes to the upload/download servers for months. Could it be a scheduler issue?

My own clients do not get this error, and as far as I can see not our other servers either.

Would suggest to detach/remove LHC@home and re-attach to the project if you observe this as suggested by Crystal Pellet.

Edit: BOINC client 7.20.2 does not have any download issues.

Do Volunteers have old Boinc-Versions for this problem?
https://lhcathome.cern.ch/lhcathome/host_stats.php?boinc_version=1
Boinc-Website is the Standard for Boinc-Version in Windows, Linux and other.

I have 8.02 and get the error so I don't think its to do with the boinc version?

The downloads are 0 so looking in my logs and the code its looks to be a 404 error which is resource not found, I don't see what we can do about this.

Ok, have only Win11pro, no Linux for Atlas.
Are there other Download Server for Linux?

They are all the same servers for both OS as far as I understand

Nils Høimyr wrote:

We do have a new server running the ATLAS assimilator and validator, but not have any changes to the upload/download servers for months. Could it be a scheduler issue?

My own clients do not get this error, and as far as I can see not our other servers either.

Would suggest to detach/remove LHC@home and re-attach to the project if you observe this as suggested by Crystal Pellet.

Edit: BOINC client 7.20.2 does not have any download issues.

On one of my hosts, for testing purposes, I now detached and re-attached LHC@home - and this seems to work :-) (just resetting the project did NOT solve the problem)

I get permission denied when I try these commands?

which commands, whats error?

echo "kernel.apparmor_restrict_unprivileged_userns = 0" >/etc/sysctl.d/99-userns.conf
sysctl --system

I have tried switching to the VM tasks and now getting validate error

I get permission denied when I try these commands?

You need root privileges, either 'su' or use sudo

---

I got some Atlas tasks on my intel iMac but they end with "validate error" status like this one.
In the log I see this

2025-06-29 10:41:05 (2284): Guest Log:  *** Error codes and diagnostics ***
2025-06-29 10:41:05 (2284): Guest Log:     "exeErrorCode": 0,
2025-06-29 10:41:05 (2284): Guest Log:     "exeErrorDiag": "",
2025-06-29 10:41:05 (2284): Guest Log:     "pilotErrorCode": 1305,
2025-06-29 10:41:05 (2284): Guest Log:     "pilotErrorDiag": "Failed to execute payload:/bin/bash: Sim_tf.py: command not found",
2025-06-29 10:41:05 (2284): Guest Log:  *** Listing of results directory ***

I can see the other wingmen of the WU are having errors too.

Edith : I can see I have another task that is currently running for 2 hours without failing (37%), in the stderr.txt of the task I found this

2025-06-29 12:07:54 (6239): Guest Log: [INFO] Detected branch: prod
2025-06-29 12:07:54 (6239): Guest Log: This is the prod version of the ATLAS job wrapper
2025-06-29 12:07:54 (6239): Guest Log: Copying input files
2025-06-29 12:08:02 (6239): Guest Log: Copied input files into RunAtlas.
2025-06-29 12:08:02 (6239): Guest Log: This VM did not configure a local http proxy via BOINC.
2025-06-29 12:08:02 (6239): Guest Log: Small home clusters do not require a local http proxy but it is suggested if
2025-06-29 12:08:02 (6239): Guest Log: more than 10 cores throughout the same LAN segment are regularly running ATLAS like tasks.
2025-06-29 12:08:02 (6239): Guest Log: Further information can be found at the LHC@home message board.
2025-06-29 12:08:11 (6239): Guest Log: Running cvmfs_config stat atlas.cern.ch

what does this mean ?

any idea why the task
https://lhcathome.cern.ch/lhcathome/result.php?resultid=427017087
failed after 1 hr 48 mins?

Stderr does not tell much, at least not to me