Message boards : ATLAS application : ATLAS issues
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 · 6 · Next

AuthorMessage
Guiri-One[Andalucia]

Send message
Joined: 1 Feb 06
Posts: 66
Credit: 9,723
RAC: 0
Message 38052 - Posted: 22 Feb 2019, 11:44:20 UTC - in response to Message 38049.  

This is not a reasonable reason :)
If it is "allowed" to do so, it must work, otherwise it must not be allowed. Now is kind of random, I am not even sure whether valid results I retunred, are really valid for science...

I can't have my laptop inactive during days for LHC (because Virtual Box shows "job unmanageable"...).
ID: 38052 · Report as offensive     Reply Quote
Jonathan

Send message
Joined: 25 Sep 17
Posts: 93
Credit: 3,074,875
RAC: 2,365
Message 38059 - Posted: 23 Feb 2019, 11:17:20 UTC - in response to Message 38052.  

You might be better off not running projects or tasks that require Virtual Box. Regular Boinc tasks tend to run smoother.
ID: 38059 · Report as offensive     Reply Quote
bronco

Send message
Joined: 13 Apr 18
Posts: 443
Credit: 8,438,885
RAC: 0
Message 38063 - Posted: 24 Feb 2019, 3:40:57 UTC - in response to Message 38052.  

Theory tasks are far more tolerant of interruptions. Maybe stick with Theory.
Have you tried assigning more CPUs to the tasks so that they can start and finish uninterrupted during your sleep hours?
The only other option I can think of is to convince them to offer smaller ATLAS tasks that process 100 events rather than the standard 200 events.

Remember ATLAS was not initially developed for BOINC. It was developed for a vast network of many thousands of non-BOINC hosts on CERN's own non-BOINC distributed computing network. BOINC is just an after thought, a mere add on attached to their DC network with bits of glue and twine. BOINC delivers up only a few hundreds of hosts that together return a miniscule amount of work compared to CERN's own DC network. While they certainly appreciate our contributions there is a limit on how much effort they should invest in keeping less capable BOINC hosts happy.
ID: 38063 · Report as offensive     Reply Quote
Profile rbpeake

Send message
Joined: 17 Sep 04
Posts: 99
Credit: 30,618,118
RAC: 3,938
Message 38068 - Posted: 26 Feb 2019, 20:58:09 UTC - in response to Message 38063.  

...Remember ATLAS was not initially developed for BOINC. It was developed for a vast network of many thousands of non-BOINC hosts on CERN's own non-BOINC distributed computing network. BOINC is just an after thought, a mere add on attached to their DC network with bits of glue and twine. BOINC delivers up only a few hundreds of hosts that together return a miniscule amount of work compared to CERN's own DC network. While they certainly appreciate our contributions there is a limit on how much effort they should invest in keeping less capable BOINC hosts happy.

Although from this table, https://lhcathome.cern.ch/lhcathome/atlas_job.php Boinc is not an insignificant contributor to the overall ATLAS effort. :)
Regards,
Bob P.
ID: 38068 · Report as offensive     Reply Quote
bronco

Send message
Joined: 13 Apr 18
Posts: 443
Credit: 8,438,885
RAC: 0
Message 38073 - Posted: 27 Feb 2019, 19:46:35 UTC - in response to Message 38068.  

For nearly a year I've been thinking BOINC is assigned the color black or perhaps dark grey and so I said "miniscule". Today I looked at that chart for the first time on a 50" screen where BOINC's color looks more like blue than grey/black. On that basis I agree... BOINC's contribution is not insignificant... and, if I am associating BOINC with the correct blue in the bars, it looks like BOINC is a significant contributor. Though our contribution is never huge on any day, it looks like it's there every day, assuming the chart is a typical month. Some of the other contributors (eg. red and violet) are huge some days but less than BOINC's on other days. Tortoise vs. hare.
ID: 38073 · Report as offensive     Reply Quote
marsinph

Send message
Joined: 4 Dec 07
Posts: 6
Credit: 1,338,198
RAC: 0
Message 38075 - Posted: 27 Feb 2019, 20:38:23 UTC - in response to Message 38068.  

...Remember ATLAS was not initially developed for BOINC. It was developed for a vast network of many thousands of non-BOINC hosts on CERN's own non-BOINC distributed computing network. BOINC is just an after thought, a mere add on attached to their DC network with bits of glue and twine. BOINC delivers up only a few hundreds of hosts that together return a miniscule amount of work compared to CERN's own DC network. While they certainly appreciate our contributions there is a limit on how much effort they should invest in keeping less capable BOINC hosts happy.

Although from this table, https://lhcathome.cern.ch/lhcathome/atlas_job.php Boinc is not an insignificant contributor to the overall ATLAS effort. :)



Hello,
Thank you for your explanation about ATLAS not developped for BOINC.
Why they send WU if BOINC community is less usefull than CERN computing possibilities ???
I understand that all help is usefull and welcome. But there are more problem than others
So, I tyhink there are two possibilities
1. CERN develop ATLAS stable version for BOINC
2. ATLAS need to be removed from BOINC !!!

Not forget if BOINC community is to little to be able to help CERN, then BOINC community is to little to crunch.
I am in top 150 AVG cruncher (with only three little hosts).
If we consider the first user (CERN it self) with avg of 5.8M, they have a capacity of more than the whole planet (not considering the second in AVG)

In total credits, the same !!!

Then CERN ask, recruit MAC developper. OK so far.
But what for ??? They not develop BOINC (dixit Profile rbpeake).

Best regards


.
ID: 38075 · Report as offensive     Reply Quote
bronco

Send message
Joined: 13 Apr 18
Posts: 443
Credit: 8,438,885
RAC: 0
Message 38077 - Posted: 27 Feb 2019, 21:22:54 UTC - in response to Message 38075.  

But ATLAS is my favorite toy. It runs stable here (~99% success rate). You want to smash my ATLAS toy because you can't make your ATLAS toy work? Find a different toy and leave mine alone.
ID: 38077 · Report as offensive     Reply Quote
tullio

Send message
Joined: 19 Feb 08
Posts: 708
Credit: 4,336,250
RAC: 0
Message 38103 - Posted: 5 Mar 2019, 8:44:35 UTC

I am running ATLAS@home on my Windows 10 PC with good results. VBox is 6.04. But if I open the VirtualBox Manager I see a list of Inaccessible machines and I remove them.
Tullio
ID: 38103 · Report as offensive     Reply Quote
Guiri-One[Andalucia]

Send message
Joined: 1 Feb 06
Posts: 66
Credit: 9,723
RAC: 0
Message 38106 - Posted: 5 Mar 2019, 12:21:05 UTC - in response to Message 38075.  

So true.

Tired of having my laptop on for hours for random results...
ID: 38106 · Report as offensive     Reply Quote
bronco

Send message
Joined: 13 Apr 18
Posts: 443
Credit: 8,438,885
RAC: 0
Message 38108 - Posted: 5 Mar 2019, 18:32:55 UTC - in response to Message 38106.  

So true.

Tired of having my laptop on for hours for random results...

You may read Don Quixote by Cervantes (a well known author in your locale) followed by Yeti's Checklist. Pay special attention to item #21.
ID: 38108 · Report as offensive     Reply Quote
mmonnin

Send message
Joined: 22 Mar 17
Posts: 55
Credit: 10,223,976
RAC: 2,477
Message 38109 - Posted: 5 Mar 2019, 21:51:51 UTC - in response to Message 38108.  

So true.

Tired of having my laptop on for hours for random results...

You may read Don Quixote by Cervantes (a well known author in your locale) followed by Yeti's Checklist. Pay special attention to item #21.


Did you mean another # as the list only goes to 19.
ID: 38109 · Report as offensive     Reply Quote
bronco

Send message
Joined: 13 Apr 18
Posts: 443
Credit: 8,438,885
RAC: 0
Message 38110 - Posted: 6 Mar 2019, 3:09:42 UTC - in response to Message 38109.  

It was a test to see if he actually read it.
ID: 38110 · Report as offensive     Reply Quote
Guiri-One[Andalucia]

Send message
Joined: 1 Feb 06
Posts: 66
Credit: 9,723
RAC: 0
Message 38112 - Posted: 6 Mar 2019, 8:42:22 UTC - in response to Message 38110.  

So funny.

Very tipical to provide a fancy answer instead of an answer at all (if known).

Atlas works some times and some other times, doesn't (same laptop same everything). Same with Theory.
I know is because my laptop needs to be down for 2 hours a day and project doesn't like to suspend - restore the computing.

Why? I have read Quijote's, there is no #21 and I have found no answer. Do you have one?

All this in addiiton og "job unmanageable" randomly appearing....
ID: 38112 · Report as offensive     Reply Quote
Jonathan

Send message
Joined: 25 Sep 17
Posts: 93
Credit: 3,074,875
RAC: 2,365
Message 38114 - Posted: 6 Mar 2019, 9:10:31 UTC - in response to Message 38112.  

Guiri-One, you would probably get better responses if you provided more information on what computer you are having trouble with, what tasks it is running and what projects. Your computers, tasks and important information to help are hidden because you are with Gridcoin. No computers show under your name recently. Pointing you to the check list is the best way of helping you sort it out.

You should probably just give up trying to run Virtual Box related projects and just stick to conventional Boinc projects. They are easier to manage and behave better.
ID: 38114 · Report as offensive     Reply Quote
Guiri-One[Andalucia]

Send message
Joined: 1 Feb 06
Posts: 66
Credit: 9,723
RAC: 0
Message 38115 - Posted: 6 Mar 2019, 13:03:20 UTC - in response to Message 38114.  

it was in the forum, but task is already old.

https://lhcathome.cern.ch/lhcathome/show_host_detail.php?hostid=10581367

Thats my host.

Enjoy :)
ID: 38115 · Report as offensive     Reply Quote
computezrmle
Volunteer moderator
Volunteer developer
Volunteer tester
Help desk expert
Avatar

Send message
Joined: 15 Jun 08
Posts: 2386
Credit: 222,942,387
RAC: 137,359
Message 38116 - Posted: 6 Mar 2019, 13:51:11 UTC - in response to Message 38115.  

From your logfile:
https://lhcathome.cern.ch/lhcathome/result.php?resultid=218612790

2019-03-04 21:16:44 (9476): Detected: VirtualBox VboxManage Interface (Version: 6.0.0)

You may upgrade to the most recent VirtualBox version.
Before you start the BOINC client again, the VBox environment should be cleaned, i.e. remove remains from crashed VMs.

2019-03-04 21:16:46 (9476): Setting Memory Size for VM. (5700MB)
2019-03-04 21:16:46 (9476): Setting CPU Count for VM. (3)

Depending on the overall system load this settings may be too heavy for your host.
First try to get a stable 1-core setup running for 1 or 2 weeks.
Then slightly (!) raise the #cores per task.

2019-03-04 21:16:51 (9476): Required extension pack not installed, remote desktop not enabled.

Install the extension pack to get the information from the VM consoles.
As suggested by other volunteers you may follow Yeti's checklist.
It's mentioned there.

2019-03-04 21:23:52 (9476): Guest Log: set up http_proxy http://http://10.49.1.1:8080/accelerated_pac_base.pac:80

The proxy setting seems to be malformed.
1. "http://http://" is wrong.
2. IIRC the BOINC client can't deal with pac files.
Just enter port "8080" and name "your_proxy.example.com" or IP "10.49.1.1" into the client's proxy form.
Theory tasks can forward the proxy configuration to the VM but this requires the proxy to listen on port 3128.

2019-03-04 22:23:04 (9476): Guest Log: ormPsy.JtorbfTErxaen.sefxoercumtse. t2r0f1E9x-e0.3e-x0e4c u2t2e: 2200:1495,-30933- 0I4N F2O2 :E2V0N:T4t5o,H3I9T3S  IeNxFeOc uEtVoNrT rteotHuIrTnSs  e6x5e
2019-03-04 22:23:04 (9476): Guest Log: utoPry JroebtTurrannss f6o5r
2019-03-04 22:23:04 (9476): Guest Log: s.tPryfJEoxbeT.rvaanlsidfaotrem s2.0t1r9f-E0x3e-.0v4a l2i2d:a2t0e: 426,041292- E0R3R-0O4R  V2a2l:i2d0a:t4i6o,n4 2o2f  ErReRtOuRr nV acloiddea ftaiiolne do:f  Nroent-uzrenr oc ordeet ufrani lceodd:e  Nfroonm- zEeVrNoT troeHtIuTrSn  (c6o5d)e  (fErrormo r EcVoNdTet o6H5I)T
2019-03-04 22:23:04 (9476): Guest Log:  (6P5y)J o(bETrrraonrs fcoordmes .6t5rf)E
2019-03-04 22:23:04 (9476): Guest Log: e.PvaylJiodbaTtrea n2s0f1o9r-m0s3.-t0r4f E2x2e:2.0v:a4l6i,d4a3t9e  I2N0F1O9 -S0c3a-n0n4i n2g2 :lo2g0f:i4l6e, 4l3o9g .IENVFNOT tSocHaInTnSi nfgo rl eorgrfoirlse
2019-03-04 22:23:04 (9476): Guest Log: logP.EyVJNoTbtTorHaInTsfSo rfmosr. terrarnosrfso
2019-03-04 22:23:04 (9476): Guest Log: m.exPeycJuotbeT r2a0n1s9f-o0r3m-s0.4t r22a:n2s0f:o4r6m,.6e4x1e cCuRtIeT I2C0A1L9 -T0r3a-n0s4fo r2m2 :e2x0e:c4u6t,o6r4 1r aCisReIdT ITCrAaLn sTfroarnmsVfaolrimd aetxieoncEuxtcoerp triaoins:e dN oTnr-aznesrfoo rrmeVtaulrni dcaotdieo nfErxocme pEtViNoTnt:o HNIoTnS- z(e6r5o) ;r eLtougfrinl ec oedrer ofrr oimn  ElVoNgT.tEoVHNITTtSo HI(T6S5:) ;" ALtohgMfpiElvet LeororpoMrg ri n   l o FgA.TEAVLN TmtaokHeIPToSo:l  "fAatihlMepdE fvotrL oAotphMMgprE v t L o oFpAMTgArL. SmhaakreePdEovotlQ ufeauielPerdo vfiodre rA"t
2019-03-04 22:23:04 (9476): Guest Log: MpEvPtyLJooobpTMrgarn.sfSohramrsed.EtvratnQsufeouremP.reoxveicduetre" 
2019-03-04 22:23:04 (9476): Guest Log: 019-P0y3J-o0b4T r2a2n:s2f0o:r5m1s,.4t8r6a nWAsRfNoIrNmG. eTxreacnustfeo r2m0 1n9o-w0 3e-x0i4t i2ng2 :e2a0:r5l1y, 4w8i6t hW AeRxNiItN Gc oTdrea 6n5s f(oNromn -nzoewr oe xriettiunrgn  ecaordley f rwoimt hE VeNxTitto HcIoTdSe  (6655 )(;N oLong-zefriol er eetrurronr  ciond el ofgr.oEmV NETtVoNHTItToSH:I T"SA t(h6M5); Logfile erprEvotrL ooipnM glr o g . E FVANTTAtL omHaIkTePSo:o l f"aAitlheMpEvdt Lfooorp MAgtrh M p E v tFLAoToApLM gmra.kSehaProeodlE vftaQiuleeude Pfroorv iAdtehrM"p)E

Messed lines like this point out an overloaded host or a wrong priority setting.
It may help if you set the vboxwrapper to a higher priority.
Crystal Pellet and others mentioned that a couple of times in the message board.

2019-03-04 22:23:05 (9476): Guest Log: No HITS file was produced

This indicates: No scientific output was generated. :-(
ID: 38116 · Report as offensive     Reply Quote
Guiri-One[Andalucia]

Send message
Joined: 1 Feb 06
Posts: 66
Credit: 9,723
RAC: 0
Message 38117 - Posted: 6 Mar 2019, 15:04:34 UTC - in response to Message 38116.  

And this is an asnwer :)

I have already updated VB to latest version and will set boinc to use less cores as before (I had 3 and will start testing with 1).

Nevertheless, I think is a waste if I can't use all cores, in comparison to other proyects...

Will come back once this is tested :)

Thx very much¡
ID: 38117 · Report as offensive     Reply Quote
computezrmle
Volunteer moderator
Volunteer developer
Volunteer tester
Help desk expert
Avatar

Send message
Joined: 15 Jun 08
Posts: 2386
Credit: 222,942,387
RAC: 137,359
Message 38119 - Posted: 6 Mar 2019, 15:43:40 UTC - in response to Message 38117.  

I have already updated VB to latest version ...

As of today your host details show:
Virtualbox (6.0.0) installed
https://lhcathome.cern.ch/lhcathome/show_host_detail.php?hostid=10581367

Since January 28th, 2019 Virtualbox 6.0.4 is released.
https://www.virtualbox.org/

A small example but it shows why hosts and logs should be exposed to others if you ask for help.
And for the same reason they should be linked to the original username instead of a pool account.
ID: 38119 · Report as offensive     Reply Quote
Jonathan

Send message
Joined: 25 Sep 17
Posts: 93
Credit: 3,074,875
RAC: 2,365
Message 38120 - Posted: 6 Mar 2019, 16:23:02 UTC - in response to Message 38119.  

It looks like your computer got updated to Boinc version 7.14.2 from 7.7 but you still have the older VB version. Are you running any other Boinc projects on computer 10581367?
I don't know if VB has any problems or not but you should probably limit your applications to use two or less cores for a work unit, by using an app_config.xml for each Atlas and Theory

Below I edited mine to limit the Atlas work unit to run a single, two core task assigned 4.8Gb. I don't run Theory so I don't have one created for that.

You are memory limited on that computer with Atlas tasks, having only 8Gb and integrated graphics

<app_config>
 <app>
  <name>ATLAS</name>
  <max_concurrent>1</max_concurrent>
 </app>
 <app_version>
  <app_name>ATLAS</app_name>
  <plan_class>vbox64_mt_mcore_atlas</plan_class>
  <avg_ncpus>2.000000</avg_ncpus>
  <max_ncpus>2.000000</max_ncpus>
  <cmdline>--memory_size_mb 4800</cmdline>
 </app_version>
</app_config>
ID: 38120 · Report as offensive     Reply Quote
Guiri-One[Andalucia]

Send message
Joined: 1 Feb 06
Posts: 66
Credit: 9,723
RAC: 0
Message 38121 - Posted: 6 Mar 2019, 16:50:13 UTC - in response to Message 38120.  

I just updated, no communicated to boinc yet.

Where should I place such app_config.xml? in

D:\boinc2\projects\lhcathome.cern.ch_lhcathome ?


There is no subfolfer for each project there...


Thx¡

Javi
ID: 38121 · Report as offensive     Reply Quote
Previous · 1 · 2 · 3 · 4 · 5 · 6 · Next

Message boards : ATLAS application : ATLAS issues


©2024 CERN