Development for LHC@home

https://lhcathomedev.cern.ch/lhcathome-dev/result.php?resultid=3396872

running now for 4 min. Chapeau

==========================================================>> LT1 slot 0 <<==========================================================
===> [runRivet] Thu Apr 17 19:22:31 UTC 2025 [boinc pp winclusive 7000 - - pythia8 8.313 CP2-CR1 100000 607]
====================================================================================================================================

Pythia::next(): 7000 events have been generated
7000 events processed
7100 events processed
7200 events processed
7300 events processed
7400 events processed
7500 events processed
7600 events processed
7700 events processed

==========================================================>> LT1 slot 1 <<==========================================================
===> [runRivet] Thu Apr 17 19:31:29 UTC 2025 [boinc pp zinclusive 13000 -,-,106,- - pythia8 8.313 tune-AU2ct10 100000 608]
====================================================================================================================================
6 13 mu- 23 5 0 0 0 0 0 0.740 47.237 -560.165 562.154 0.106
7 -13 mu+ 23 5 0 0 0 0 0 -0.740 -47.237 -104.069 114.290 0.106
Charge sum: 0.000 Momentum sum: 0.000 0.000 -664.234 676.444 127.944

-------- End PYTHIA Event Listing -----------------------------------------------------------------------------------------------
Rivet.AnalysisHandler: INFO Only using nominal weight. Variation weights will be ignored.
Rivet.Analysis.CMS_2022_I2079374: INFO Assuming a mixed electron/muon sample. Set LMODE=EL/MU if your sample contains a single flavor.
0 events processed
100 events processed
200 events processed

New version (v7.57) with this fix:

https://github.com/lfield/lhcathome/pull/14

set local proxy:
wsl
CVMFS_HTTP_PROXY="http://xx.yy.zz:3128"

In reply to Laurence CERN's message of 17 Apr 2025:

New version (v7.57) with this fix:

https://github.com/lfield/lhcathome/pull/14

I have let this version run with 3 tasks concurrently overnight and all 47 returned v7.57-tasks so far were successsful from BOINC's point of view.
Several tasks did not make it to the 'event processing' part of the job, but this also happens with the VBox version on the production site.

In reply to maeax's message of 18 Apr 2025:

set local proxy:
wsl
CVMFS_HTTP_PROXY="http://xx.yy.zz:3128"

This is incomplete.
To avoid CVMFS configures an unwanted backup proxy (e.g. from CERN) 'DIRECT' must be added.
CVMFS_HTTP_PROXY="http://xx.yy.zz:3128;DIRECT"

Thank you, have it done now.
22 GByte Download, 0.5 GByte upload up to now.

Strange behaviour.
The 7.57 wus in Windows start very fast (50% done after 3 minutes) then slow down and arrive to 99% after 40 minutes.

Is it normal?

maeax@RYZENMP:/mnt/c/Users/Megaport$ ifconfig
docker0: flags=4099<UP,BROADCAST,MULTICAST> mtu 1500
inet 172.17.0.1 netmask 255.255.0.0 broadcast 172.17.255.255
ether 02:42:cb:01:0c:bd txqueuelen 0 (Ethernet)
RX packets 0 bytes 0 (0.0 B)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 0 bytes 0 (0.0 B)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

eth0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1492
inet 172.21.117.136 netmask 255.255.240.0 broadcast 172.21.127.2

This is not my own network adress area.

In reply to Crystal Pellet's message of 18 Apr 2025:

In reply to Laurence CERN's message of 17 Apr 2025:

New version (v7.57) with this fix:

Several tasks did not make it to the 'event processing' part of the job, but this also happens with the VBox version on the production site.

Crystal, have Boinc 8.2.1, you are using 8.2.0.

Win11pro Workstation running with Boinc 8.2.0
https://lhcathomedev.cern.ch/lhcathome-dev/results.php?hostid=5116

In reply to maeax's message of 18 Apr 2025:

set local proxy:
wsl
CVMFS_HTTP_PROXY="http://xx.yy.zz:3128"

command output:
Got a proxy from the local environmentWill use it for CVMFS and Frontier
Using custom CVMFS.Probing CVMFS repositories ...Probing /cvmfs/alice.cern.ch... OKProbing /cvmfs/cvmfs-config.cern.ch... OKProbing /cvmfs/grid.cern.ch... OKProbing /cvmfs/sft.cern.ch... OK
Excerpt from "cvmfs_config stat":VERSION HOST PROXY2.12.7.0 http://s1cern-cvmfs.openhtc.io http://10.116.178.201:3128Environment HTTP proxy: http://10.116.178.201:3128job: htmld=/var/www/lighttpdjob: unpack exitcode=0job: run exitcode=1job: diskusage=6996job: logsize=16 kjob: times=0m0.006s 0m0.000s0m29.436s 0m16.141sjob: cpuusage=46
Job Finished===>
[runRivet] Tue Apr 22 07:36:45 UTC 2025 [boinc pp jets 13000 650 - pythia8 8.313 ppthermNoScat 100000 642]
Filesystem Used Use% Mounted oncvmfs2 159K 1% /cvmfs/cvmfs-config.cern.chcvmfs2 19M 1% /cvmfs/alice.cern.chcvmfs2 17M 1% /cvmfs/grid.cern.chcvmfs2 1.9G 47% /cvmfs/sft.cern.chtotal 1.9G 12% -boinc_shutdown called with exit code 0EOMstderr from container:

SUCCESS after 6 days of work!

I recently made a new host online (Windows 10 running BOINC client 8.2.1 without Docker/ Podman). I thought I had VBox installed and good to go but the host is reporting Virtualization: None, that I'll look into later this week. However, before I noticed this, I'd attached it to LHC Dev, assuming I'd get a VBox task. However, I was assigned a Docker task (https://lhcathomedev.cern.ch/lhcathome-dev/result.php?resultid=3406162) which failed as expected.

Is the assignment logic only based on the client version or is there a check for Docker/ Podman that's supposed to prevents assignments such as these?

<core_client_version>8.2.1</core_client_version>
<![CDATA[
<message>
Incorrect function.
 (0x1) - exit code 1 (0x1)</message>
<stderr_txt>
docker_wrapper config:
   workdir: /boinc_slot_dir
   use GPU: no
   create args: --cap-add=SYS_ADMIN --device /dev/fuse
   verbose: 1
wsl_init(): no usable WSL distro
wsl_init() failed: -1
2025-05-11 20:14:23 (1532): called boinc_finish(1)

Here's the contents of the event log at BOINC startup:

Not the way its setup at the moment, it would need to be a different project then you could deslect it.

WSL and Vbox don't work together so you need to switch off virtulisation e.g.

bcdedit /set hypervisorlaunchtype off

Shouldn't the server have seen that I have no WSL support and not assigned a task requiring WSL? E.g, https://lhcathomedev.cern.ch/lhcathome-dev/result.php?resultid=3413586. The only other project that had Docker apps (BOINC Central) has run out of tasks and I can't test this anywhere else.

For example, if I do not have Virualisation on and VirtaulBox installed, I wouldn't have received VBox tasks. I'd have expected the same for docker tasks.

Well you would think and hope that but it doesn't work with these Dockers so I stopped after I figured it was going to work on a 28 core and all I got was nothing but computer errors so I didn't try that again and went back to CMS

---

Mad Scientist For Life