Development for LHC@home

To fix this problem I've released 0.99 and submitted a bunch of test tasks.

Thanks for testing and reporting problems, if 0.99 looks ok then hopefully I can release it into production this week.

I tested several of the 0.99 tasks. Seems OK to me, but better look yourself to the results: https://lhcathomedev.cern.ch/lhcathome-dev/results.php?hostid=3717&offset=0&show_names=0&state=0&appid=5

The part where there was obviously a tar-problem looks like this now:

di  7 jan 2020 21:49:54 CET: CVMFS is ok
di  7 jan 2020 21:49:54 CET: System is not Red Hat/CentOS 7, singularity is required
di  7 jan 2020 21:49:54 CET: Using singularity image /cvmfs/atlas.cern.ch/repo/containers/images/singularity/x86_64-centos7.img
di  7 jan 2020 21:49:54 CET: Checking for singularity binary...
di  7 jan 2020 21:49:54 CET: Singularity is not installed, using version from CVMFS
di  7 jan 2020 21:49:54 CET: Checking singularity works with /cvmfs/atlas.cern.ch/repo/containers/sw/singularity/x86_64-el7/current/bin/singularity exec -B /cvmfs /cvmfs/atlas.cern.ch/repo/containers/images/singularity/x86_64-centos7.img hostname
di  7 jan 2020 21:50:34 CET: INFO:  Convert SIF file to sandbox... LinAH125 INFO:  Cleaning up image...
di  7 jan 2020 21:50:34 CET: Singularity works
di  7 jan 2020 21:50:34 CET: Set ATHENA_PROC_NUMBER=4
di  7 jan 2020 21:50:34 CET: Starting ATLAS job with PandaID=4002876565
di  7 jan 2020 21:50:34 CET: Running command: /cvmfs/atlas.cern.ch/repo/containers/sw/singularity/x86_64-el7/current/bin/singularity exec --pwd /var/lib/boinc-client/slots/0 -B /cvmfs,/var /cvmfs/atlas.cern.ch/repo/containers/images/singularity/x86_64-centos7.img sh start_atlas.sh

This task was running on HP i7 without docker.
https://lhcathomedev.cern.ch/lhcathome-dev/result.php?resultid=2856616
Mi 8. Jan 12:51:06 CET 2020: *** Error codes and diagnostics ***
Mi 8. Jan 12:51:06 CET 2020: "exeErrorCode": 65,
Mi 8. Jan 12:51:06 CET 2020: "exeErrorDiag": "EVNTtoHITS got a SIGBUS signal (exit code 135)",
Mi 8. Jan 12:51:06 CET 2020: "pilotErrorCode": 0,
Mi 8. Jan 12:51:06 CET 2020: "pilotErrorDiag": "",

Don't know what is needed.
Atlas-native on Production is running well:
https://lhcathome.cern.ch/lhcathome/results.php?hostid=10618519

HP i7 Centos7-VM now rebooted after last reboot from 1.12.19(!).
Had upgraded to 15 GByte RAM in use. Task is now running 0.99 without Docker.
https://lhcathomedev.cern.ch/lhcathome-dev/workunit.php?wuid=1969419
After writing this lines. Task is crashed.....
Mi 8. Jan 18:45:45 CET 2020: "exeErrorDiag": "EVNTtoHITS got a SIGBUS signal (exit code 135)",

Edit: Get tomorrow 32 GByte more RAM for Ryzen 2700.
Will than testing CentOS7-VM WITH Docker again.

Edit2: Is a other Version of Python needed?

This version has now been released on the production server. Once it looks ok I will update the docker configuration to pull WU from there instead of the LHC-dev server.

Well, that didn't go too well.. lots of the SIGBUS errors like maeax reported. Sorry for not looking more closely at these problems here.

I will submit some more test WU to try and get to the bottom of the problem.

no risc, no fun ;-)