LHC@home

OMG, it's Friday afternoon, isn't it?

SCNR ;-)

CGI testing, the whole day, see -dev forum.

CGI testing, the whole day, see -dev forum.

Thank you.

What a shame! I didn't check the dev MB today.
:-O

CGI testing, the whole day, see -dev forum.

do you have a link?

Sorry, either I am blind or dumm :-)

CGI testing, the whole day, see -dev forum.

do you have a link?

Sorry, either I am blind or dumm :-)

The news section of https://lhcathomedev.cern.ch/lhcathome-dev/

If that's on the dev site, what does it have to do with Atlas here?

---

I suspect that there are other reasons why no ATLAS tasks are available for download; if it was the CGI testing, other subprojects would be affected as well, right?

Anyone any idea when ATLAS tasks will become available again?

Heat holidays for cpu. That's fine.

---

I am surprised that still no-one from the LHC/ATLAS people can tell us what's going on, and when new tasks will be available again.

I'm optimistically hopeful that it will be fixed on Monday. Meanwhile, crunching other LHC projects, although today I was having problems with LHCb, so switched everything to Theory.

---

Regards,
Bob P.

It's always Friday afternoon these things happen... Something changed in the machinery to submit tasks to the BOINC server which stopped new WU getting submitted. It is fixed now and was unrelated to changes in the dev project. Sorry for the lack of work the last couple of days.

during last night, I got downloaded new ATLAS tasks. What catches my eye though is that the BOINC Manager shows a "remaining time" of about 2 1/2 days.
Is this a different type of task now ("longrunners" as we had them some time ago)?

Is this a different type of task now ("longrunners" as we had them some time ago)?

Same observation for me: the new tasks seem to be longrunners.

---

We are the product of random evolution.

I got an ATLAS WU this early morning.
Until now it has finished 67 events within 2.5 h on a 2-core VM.
This means:
- more events than the usual 50.
- may not be a typical longrunner

Unfortunately it is not displayed how many events are left.

@David
Is it possible to include the #total events in the console output?
Like: 32/100 would mean 32 calculated/100 total

@David
Is it possible to include the #total events in the console output?
Like: 32/100 would mean 32 calculated/100 total

I strongly endorse this wish.
Plus - as already discussed some time ago, there should be a seperate sub-category for this type of task that takes more than 2 days.

I got an ATLAS WU this early morning.
Until now it has finished 67 events within 2.5 h on a 2-core VM.
This means:
- more events than the usual 50.
- may not be a typical longrunner

Definitely not a longrunner (walltime 3 h 47 m 1 s) but a 100 eventer:
https://lhcathome.cern.ch/lhcathome/result.php?resultid=149821233

Correct, these are 100 event tasks. As can be seen here we are back to running task 11283914 since 11364822 is almost complete. The long time estimates are due to the same problem causing the lack of tasks and new tasks should now have the correct estimate.

for a few hours now, none of my hosts was able to download ATLAS and CMS tasks ("no tasks available")

What's wrong?

for a few hours now, none of my hosts was able to download ATLAS and CMS tasks ("no tasks available")

What's wrong?

The same old scheduler problem due to the massive number of SixTrack tasks available.

for a few hours now, none of my hosts was able to download ATLAS and CMS tasks ("no tasks available")
What's wrong?

The same old scheduler problem due to the massive number of SixTrack tasks available.

NO GOOD :-(

but thanks, anyway, for the quick reply :-)