LHC@home

I'm getting a mix of valid and invalid ATLAS native tasks on different BOINC clients since 12:36 UTC today.
All invalids fail after about 10 min runtime.

Examples:
https://lhcathome.cern.ch/lhcathome/result.php?resultid=270323978
https://lhcathome.cern.ch/lhcathome/result.php?resultid=270328004
https://lhcathome.cern.ch/lhcathome/result.php?resultid=270324483

With CentOS in native and Windows and NO proxy, up to now for today no problems.
Edit: RDP is not shown in Windows.

At least three of my computers running Windows and vbox tasks have suddenly run into validate errors today. Tasks failing in about 6 minutes.

Examples:
https://lhcathome.cern.ch/lhcathome/result.php?resultid=270325128
https://lhcathome.cern.ch/lhcathome/result.php?resultid=270340255
https://lhcathome.cern.ch/lhcathome/result.php?resultid=270358724

have now also three in native at 18:45 UTC with a stop after 10 min.
[2020-03-30 20:24:54] Starting ATLAS job with PandaID=4687202094
[2020-03-30 20:24:54] Running command: /usr/bin/singularity exec --pwd /var/lib/boinc/slots/0 -B /cvmfs,/var /cvmfs/atlas.cern.ch/repo/containers/images/singularity/x86_64-centos7.img sh start_atlas.sh
[2020-03-30 20:24:55] Job failed
[2020-03-30 20:24:55] ++ pwd

Hi,

There was a bunch of bad tasks submitted yesterday by accident. They were submitted over a period of 2 hours yesterday afternoon and so hopefully they will be flushed out the system soon.

Thanks.
So far all ATLAS tasks I got today are running fine.

I've still been getting some. One was sent in 1 Apr 2020, 2:04:13 UTC. Doesn't bother me. Only a small amount of time is wasted.

Hi,

There was a bunch of bad tasks submitted yesterday by accident. They were submitted over a period of 2 hours yesterday afternoon and so hopefully they will be flushed out the system soon.

Most tasks from the faulty bunch were sent out for the first time 2 days ago.
They will disappear when the #errors is high enough:
https://lhcathome.cern.ch/lhcathome/workunit.php?wuid=137171081


This afternoon I got a faulty task that was generated today.
Is it part of the same old bunch or part of another faulty one?
https://lhcathome.cern.ch/lhcathome/workunit.php?wuid=137380248

I too am experiencing ATLAS tasks failing after approximately 10 minutes. I just saw over a half dozen that did that.

Looking at the task details I found this

"Stderr output

<core_client_version>7.9.3</core_client_version>
<![CDATA[
<message>
process exited with code 195 (0xc3, -61)</message>
<stderr_txt>
20:53:25 (20061): wrapper (7.7.26015): starting
20:53:25 (20061): wrapper: running run_atlas (--nthreads 12)
awk: line 2: function strftime never defined
21:03:26 (20061): run_atlas exited; CPU time 0.007836
21:03:26 (20061): app exit status: 0x1
21:03:26 (20061): called boinc_finish(195)

</stderr_txt>
]]>
"



function strftime never defined error

Searching for "strftime" at the top of this page will deliver the answer.

OK, did that...

https://lhcathome.cern.ch/lhcathome/result.php?resultid=271425574

awk: line 2: function strftime never defined


The scripts used to control LHC native tasks expect gawk to be used.
It looks like this computer is using mawk instead.
Solution: install gawk and make sure the awk link points to gawk

What is 'awk" and what is gawk and mawk?

If I disable native in my preferences I too get near immediate computation errors...

https://lhcathome.cern.ch/lhcathome/result.php?resultid=274314793
[2020-05-22 20:26:23] *** Error codes and diagnostics ***
[2020-05-22 20:26:23] "exeErrorCode": 0,
[2020-05-22 20:26:23] "exeErrorDiag": "",
[2020-05-22 20:26:23] "pilotErrorCode": 1346,
[2020-05-22 20:26:23] "pilotErrorDiag": "Transform not found:/bin/bash: Sim_tf.py: command not found\n",
[2020-05-22 20:26:23] *** Listing of results directory ***
[2020-05-22 20:26:23] insgesamt 252860

since yesterday evening, all tasks are failing after about 14 minutes runtime and about 3-4 minutes CPU time.
Examples:

https://lhcathome.cern.ch/lhcathome/result.php?resultid=277285377
https://lhcathome.cern.ch/lhcathome/result.php?resultid=277286741
https://lhcathome.cern.ch/lhcathome/result.php?resultid=277286884

what's going wrong?

tasks are failing after about 14 minutes runtime and about 3-4 minutes CPU time.

Examples:

https://lhcathome.cern.ch/lhcathome/result.php?resultid=277284464
https://lhcathome.cern.ch/lhcathome/result.php?resultid=277285287
https://lhcathome.cern.ch/lhcathome/result.php?resultid=277286589
https://lhcathome.cern.ch/lhcathome/result.php?resultid=277286071

"pilotErrorDiag": "Transform not found:/bin/bash: Sim_tf.py: command not found\n",

I have the same problem with ATLAS native since yesterday.

---

Why mine when you can research? - GRIDCOIN - Real cryptocurrency without wasting hashes! https://gridcoin.us

"pilotErrorDiag": "Transform not found:/bin/bash: Sim_tf.py: command not found\n",

I have the same problem with ATLAS native since yesterday.

ah, okay, thanks for the information.
So there might be misconfigured tasks around, for VM as well as for native.
I was wondering whether the Windows10 Update to 2004 which was made on this computer 2 days ago has to do with the problem (because before, ATLAS tasks did NOT fail). So this then probably is not the case.

I now switched to CMS, and they run well.

I now switched to CMS, and they run well.

They follow the Law of Conservation of Working Projects at CERN.
When one thing starts working, another fails.
Ivan saved us.

Searching for "strftime" at the top of this page will deliver the answer.

Hmm, was that answer retracted???

Sorry, couldn't find anything matching your search query.

Extend the default search period (30 days in the past) and you will get a couple of hits.
Among them (posted 2020-04-12):
https://lhcathome.cern.ch/lhcathome/forum_thread.php?id=5391&postid=42138

Running Atlas-Tasks showing in the Server-Stats atm 5k, mostly 10k in the past.
There must be something wrong with Boinc-Tasks from Cern.
Isn't growing up:
https://lhcathome.cern.ch/lhcathome/img/progresschart.png