LHC@home

https://lhcathome.cern.ch/lhcathome/workunit.php?wuid=226595315
PbPb heavyion-mb 2760 - - pythia8 8.302 default 100000 38

Probably it'll take nineteen days, or so. Ten days are gone, and only 52800 events processed out of 100000.
--
Bye, Lem

Yep.

They're ironing out a few wrinkles at the moment.
The CMS task problem has apparently been fixed. But I've got to wait, hoping, for a "10 Day" Theory task to finish before I get to see any and find out!

Here's my list of failure -
https://lhcathome.cern.ch/lhcathome/results.php?userid=95350
LOL

CMS task problem has apparently been fixed.

not entirely: the connection to Condor is up again, but the jobs queue has run dry; so one can download tasks, but they fail after about half an hour due to lack of jobs :-(

Probably it'll take nineteen days, or so. Ten days are gone, and only 52800 events processed out of 100000.

Probably much more: it has slowed down. In the last four days, less than 8000 events. At this pace, there are still more than 20 days left until the end of computation. Grand total, likely more than 35 days.
--
Bye, Lem

Power failure. :'-(
I'll buy an UPS, sooner or later...
--
Bye, Lem

Sorry about the power failure.
Lem - please remind me - where do I find the "events" processed?
Thanks

where do I find the "events" processed?

i use to monitor theory tasks

 tail -f /var/lib/boinc/slots/*/cernvm/shared/runRivet.log 

Herwig tasks did need a different command to filter out the spam.

where do I find the "events" processed?

i use to monitor theory tasks

 tail -f /var/lib/boinc/slots/*/cernvm/shared/runRivet.log 

Herwig tasks did need a different command to filter out the spam.

Right.
In my user directory I've created a file named: ~/.bash_aliases, which contains, among the other, the lines:

# My aliases
alias theorytasks='head -n 1 /var/lib/boinc/slots/?*/cernvm/shared/runRivet.log |grep -oPe "(?<= - )[^-][^[:blank:]]*" |sort |uniq -c'
alias theorystatus='for i in $( ls var/lib/boinc/slots/?*/cernvm/shared/runRivet.log ); do echo $i; head -n1 $i ; grep "Integrate " $i | tail -n1 ; grep " events processed" $i | tail -n1 ; echo; done'
alias atlasstatus='sudo bash -c "head -n1 /var/lib/boinc/slots/?*/PanDA_Pilot-*/eventLoopHeartBeat.txt"'
alias l@hstatus='theorystatus ; echo ; echo "---------------------------------" ; echo ; atlasstatus'

This file is recalled in ~/.bashrc, which contains also:

# Alias definitions.
# You may want to put all your additions into a separate file like
# ~/.bash_aliases, instead of adding them here directly.
# See /usr/share/doc/bash-doc/examples in the bash-doc package.

if [ -f ~/.bash_aliases ]; then
    . ~/.bash_aliases
fi

So I can use the commands theorytasks, theorystatus, atlasstatus, l@hstatus to quickly get some info about the tasks.

Herwig theory tasks have two steps: the first creates the events (and it may lasts days) to be processed, the second processes the events.
The first step can be monitored reading the file:... uhmm, let's see if I can recall correctly... You've got to dive into .../boinc/slots/*?/cernvm/shared/tmp/tmp.*/run-main/pwgevents.lhe
I'll be more precise as soon as I'll crunch some herwig task, I promise. Or someone else will post it sooner. :-)
--
Bye, Lem

Thanks for the responses.
I won't be able to implement this method -
I don't have

/var/lib/boinc/slots/*/cernvm

I do have

/var/lib/boinc/slots/*/shared

for the Theory task. But there's no runRivet.log or PanDA_Pilot anywhere. I searched for *event* and got nothing.
I do have a vague memory of being able to find the events processed in some xml file, without installing any extra software.

The suggested commands work only for native tasks.

You are running vbox tasks which write their logs to the filesystem inside the VM.
The VM then maps the logs (at least some of them) to it's internal apache workspace.
This in turn can be read using your local http browser (on the host) with an address that looks like:

http://localhost:12345/logs/

Each VM on the host has an individual port.
Get it from the stderr.txt, it looks like:

1066-10-14 16:39:37 (395588): Detected: Web Application Enabled (http://localhost:57455)

Thank you. There it is!

61,600 of 100,000 events processed
5 days so far.
Started:.....19 Nov 2024, 14:21:01 UTC
Deadline:..30 Nov 2024, 14:21:01 UTC

(This is my only PC, so BOINC only runs for ~75% of the time.)

What happens to a Theory task? Do they just count to 100,000 [default] and stop?

What happens to a Theory task? Do they just count to 100,000 [default] and stop?

When all (mostly 100,000) Theory-events are processed some data is gathered and the result is sent back to LHC@home.
This very last part of the task lasts less than 1 minute.

Perfect. Thanks. ;-)

A long Herwig one.




===> [runRivet] Mon Dec 15 08:37:16 UTC 2025 [boinc pp z1j 13000 280 - herwig7 7.2.1 nlo-pw-dipole 48000 421]

After 24 hours runtime 217 of 760 integrations done and thereafter 48000 events to process.

A long Herwig one.




===> [runRivet] Mon Dec 15 08:37:16 UTC 2025 [boinc pp z1j 13000 280 - herwig7 7.2.1 nlo-pw-dipole 48000 421]

After 24 hours runtime 217 of 760 integrations done and thereafter 48000 events to process.

This will be going very loooooong.

The integration's part was done after about 80 hours and the first 480 (1%) of the events took 4 hours, so another 17 days to go . . . . . . . . . . .

I see that there still are tasks that can run 10+ days.

Just noticed that I have one that already had a valid result. Am I wasting time finishing it, or will I get credited, assuming my result validates against the first one?

Here's another one whose first iteration got aborted after 10 days. It will probably run beyond the deadline on my PC. Should I keep running this one as well?

---

I see that there still are tasks that can run 10+ days.

Just noticed that I have one that already had a valid result. Am I wasting time finishing it, or will I get credited, assuming my result validates against the first one?

Here's another one whose first iteration got aborted after 10 days. It will probably run beyond the deadline on my PC. Should I keep running this one as well?

Those tasks have a 11-day deadline on the server. The client gets a 10 day deadline.
When they are returned before the server's deadline, you're OK. With BOINC Manager's "Show Graphics" you can follow the progress.
Your 2 linked tasks: The first one was too late but on time before your resend turned in. The second task may have restarted task's VM several times maybe starting from scratch.

EDIT: I see you reacted in the 6+ day thread.

Your 2 linked tasks: The first one was too late but on time before your resend turned in. The second task may have restarted task's VM several times maybe starting from scratch.

Both tasks are still running and it's not clear whether:
- the first task, which already had a valid result, will grant me credits if I finish it?
- if I won't finish the second task in 11 days, will it get cancelled and will I lose all 11 days of running time? If a third replicate gets sent out after 10 days I don't think anyone will finish it within 24 hours (before my hard deadline) so that shouldn't be an issue.

Edit: after re-reading your reply several times I think I figured out the misunderstanding. In both WUs that I linked, I'm running the resends, not the initial tasks.

---