Development for LHC@home

This new version will consider failed jobs as successful from a BOINC perspective. This means that the error output will be returned to the MCPlots server for further analysis and they will not be re-run on other hosts.

This seems to be working fine for me. I will put it in production tomorrow if there are no issues reported.

Hi Laurence,
have started a native.
https://lhcathomedev.cern.ch/lhcathome-dev/workunit.php?wuid=1970478
Is it possible to change this wish from Crystal from Production for Theory?
https://lhcathome.cern.ch/lhcathome/forum_thread.php?id=4888

Have now one Task with 0.5 CPU in use.
This Problem was in Production for Sixtrack also in the last weeks. Is now solved!
https://lhcathomedev.cern.ch/lhcathome-dev/result.php?resultid=2858356

Missing line with ===> and job info in Console F1 although job is running:

Should be something like:
11:13:48 CET +01:00 2020-01-14: cranky: [INFO] ===> [runRivet] Tue Jan 14 10:13:30 UTC 2020 [boinc pp w1j 7000 100 - pythia8 8.230 fischerPP1 100000 14]

This only seems to be an issue with the Windows version.

Version 5.20 should fix the issue.

My Linux VBox Theory 5.19s show the same output, so not just Windows.
Otherwise it works fine, just doesn't show the job in Alt-F1. Still available from 1st line of running.log.

5.20 for Windows does indeed fix this.

I have updated all platforms. Please let me know if there are any issues.

I have updated all platforms. Please let me know if there are any issues.

OK Laurence I guess your post woke me up at 4:44am so I am d/l this v5.20 vdi right now and going to give these a try on a Windows 10
I still have some high-speed on my Dish left so it should not take long (but I also have lots of CMS-dev running now and hundreds of Sixtracks but this is just a 4-core Ryzen so I still start these up asap)

I never tried before but now I am wondering if I can just copy this new version on a stick and put it on the other pc's so I only have to d/l this once.

---

Mad Scientist For Life

I never tried before but now I am wondering if I can just copy this new version on a stick and put it on the other pc's so I only have to d/l this once.

The best thing to do would be to create a local caching proxy with squid. This will automatically do this without you having to intervene. It may take a bit of effort to setup but overall a lot less effort than doing it manually each time.

I started a couple of these and so far so good


And before the wife sees that I am still up doing this at 5am I better go pretend

I have updated all platforms. Please let me know if there are any issues.

On Windows that job info in Console ALT-F1 is back again. It's looking good now.
The same issue is on the production server as well and that job info line is also not copied to the stderr.txt result file.
I tried a production task when using the dev-vdi and that went well. Task still busy but the result is https://lhcathome.cern.ch/lhcathome/result.php?resultid=259470817

My first 3 tasks worked like they should but now I have two running and one has been running 2.5 hours and now it changed to [ERROR] 'cvmfs_config probe sft.cern.ch' failed.


Even after they started

https://lhcathomedev.cern.ch/lhcathome-dev/show_host_detail.php?hostid=3792

So I will abort those two and try the rest that I have.