LHC@home

... but it worked for 22 and I've seen nothing to state that runc should be upgraded.

Your computer is under your responsibility, especially if you do major upgrades.
Nobody will ever inform you prior to an upgrade that program "xy" outside the maintainer's scope will not run any more.

Runc issues date back to 2019 (use the forum search yourself to find the posts) and it has been discussed many times whether to prefer the local version or that from CVMFS.
Even if there are always some people claiming "the CVMFS version runs fine here" this is/was not necessarily true in all cases.

Good to see my extended logging was helpful.
;-)

... but it worked for 22 and I've seen nothing to state that runc should be upgraded.

Your computer is under your responsibility, especially if you do major upgrades.
Nobody will ever inform you prior to an upgrade that program "xy" outside the maintainer's scope will not run any more.

Runc issues date back to 2019 (use the forum search yourself to find the posts) and it has been discussed many times whether to prefer the local version or that from CVMFS.
Even if there are always some people claiming "the CVMFS version runs fine here" this is/was not necessarily true in all cases.

Good to see my extended logging was helpful.
;-)

I'm not going to go back and read all those old posts, it was bad enough just trying to follow the loosely knotted threads of forum posts to get me to the point I'm at now.

I agree, it's my responsibility, but then runc should not have been packaged up as part of the original installation if it only needed to be installed as a dependency to run the sub-project. Simply quote the required version level, much like Sudo-Version does in the stderr logs.

Well kudos to you then, they make getting things working a whole lot better than it used to be, didn't know you were doing the coding now !

Unfortunately, even though cvmfs config looks OK, the Theory app still fails.

stderr below

Any idea what's still wrong with the setup?

23:42:47 (2718132): wrapper (7.15.26016): starting
23:42:47 (2718132): wrapper (7.15.26016): starting
23:42:47 (2718132): wrapper: running ../../projects/lhcathome.cern.ch_lhcathome/cranky-0.1.4 ()
23:42:47 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] Detected Theory App
23:42:47 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] This application must have permanent access to
23:42:47 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] online repositories via a local CVMFS service.
23:42:47 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] It supports suspend/resume if a couple of
23:42:47 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] requirements are fulfilled.
23:42:47 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] Most important:
23:42:47 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] - init process is systemd
23:42:47 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] - cgroups v2 is enabled and 'freezer' is available
23:42:47 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] - the user running this application is a member of the 'boinc' group
23:42:47 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] - sudo is at least version 1.9.10
23:42:47 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] - sudoer file provided by LHC@home is installed
23:42:47 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] Checking local requirements.
23:42:47 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] Found Sudo-Version 1.9.15p5.
23:42:50 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] Probing /cvmfs/alice.cern.ch... OK
23:42:50 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] Probing /cvmfs/cernvm-prod.cern.ch... OK
23:42:50 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] Probing /cvmfs/grid.cern.ch... OK
23:42:50 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] Probing /cvmfs/sft.cern.ch... OK
23:42:50 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] Excerpt from "cvmfs_config stat": VERSION HOST PROXY
23:42:50 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] 2.12.7.0 http://s1ral-cvmfs.openhtc.io/cvmfs/alice.cern.ch DIRECT
23:42:50 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] Found a local runc version 1.2.2.
23:42:50 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] Creating container filesystem.
23:42:50 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] Using /cvmfs/cernvm-prod.cern.ch/cvm4
23:42:50 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] Starting runc container.
23:42:50 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] To get some details on systemd level run
23:42:50 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] systemctl status Theory_2843-4272113-527_2.scope
23:42:50 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] mcplots runspec: boinc pp z1j 13000 35 - pythia6 6.428 p2010 100000 527
23:42:50 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] ----,^^^^,<<<~_____---,^^^,<<~____--,^^,<~__;_
time="2025-03-16T23:42:50+01:00" level=fatal msg="nsexec-1[2719159]: failed to unshare remaining namespaces: Operation not permitted"
time="2025-03-16T23:42:50+01:00" level=fatal msg="nsexec-0[2719158]: failed to sync with stage-1: next state: Success"
time="2025-03-16T23:42:50+01:00" level=error msg="runc run failed: unable to start container process: can't get final child's PID from pipe: EOF"
23:42:50 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] Container Theory_2843-4272113-527_2 finished with status code 1.
23:42:50 CET +01:00 2025-03-16: cranky-0.1.4: [INFO] Preparing output.
23:42:50 CET +01:00 2025-03-16: cranky-0.1.4: [ERROR] No output found.
23:42:51 (2718132): cranky exited; CPU time 0.203514
23:42:51 (2718132): app exit status: 0xce
23:42:51 (2718132): called boinc_finish(195)

---

Unfortunately, even though cvmfs config looks OK, the Theory app still fails.

stderr below

Any idea what's still wrong with the setup?

time="2025-03-16T23:42:50+01:00" level=fatal msg="nsexec-1[2719159]: failed to unshare remaining namespaces: Operation not permitted"
time="2025-03-16T23:42:50+01:00" level=fatal msg="nsexec-0[2719158]: failed to sync with stage-1: next state: Success"
time="2025-03-16T23:42:50+01:00" level=error msg="runc run failed: unable to start container process: can't get final child's PID from pipe: EOF"

The error messages are caused by runc.
You need to find a runc version working fine on Ubuntu.

Volunteers running Ubuntu report runc issues every now and then.
On 22.04 runc v1.1.12 returned valid resluts but there may even be a more recent version that succeeds.

All right! Now it works, first workunit returned and valid!

Turns out runc had been installed from the containerd.io package as part of docker. Removing docker and installing runc separate did the trick.

Thanks for the help and sorry for the earlier ot + double post - didn't spot your message until I had already reposted thinking the earlier post somehow failed.

---

Is there anything we can do about the proxy failing or will we have to wait for a new version for that?

Is there anything we can do about the proxy failing or will we have to wait for a new version for that?

Do you have proxy issues with Theory native?
If so, you should provide more information, at least a link to a failing task.

If you meant Theory vbox then you need to wait for a new app_version.

In the other topic just created by Laurence it says "new version is fixing the proxy issue" :)

does anyone know when theory native will be back.
Don't like this v-box things.

As soon a the next version of the BOINC client is released. We had to stop sending tasks for the current native version due to the need for having an EL9 container. Work on the EL9 version was focused on the new BUDA from BOINC. It replaces the runc component with docker or podman. We have been testing it on the Dev project and it is looking promising.

By reading the page you mention I don't understand if this BUDA thingy is "only for the server side of things" or also for the boinc client ? like we would not need VB anymore and boinc would run a packaged boinc application via docker on the participant machine ?