Development for LHC@home

Well most of my 82 tasks are already finished.....sure would be nice if they weren't't just 1 or 2 hour versions
And in the middle of running these my Event Log tells me why I can't send all of them back and why it takes several minutes just to load this page.

5/13/2020 1:10:18 PM | | Project communication failed: attempting access to reference site
5/13/2020 1:10:54 PM | | Internet access OK - project servers may be temporarily down.
5/13/2020 1:15:26 PM | | Project communication failed: attempting access to reference site
5/13/2020 1:15:52 PM | | Internet access OK - project servers may be temporarily down.
5/13/2020 1:22:20 PM | | BOINC can't access Internet - check network connection or proxy configuration


why do I stay up until 3:30am and get back up5 hours later just to get this is a better question right now

For the last 24 hours I have not been able to send in the rest of my 82 finished tasks
(I did see they are sent back from Germany but not over here)

I can do all the rest of the usual on the WWW including running tasks and the usual websites and I have tried at full-speed and the slower speed on my end)



My team member from Canada is off running those boinc Pentathlon right now or I would check with him

Can you take a look when you get a chance over there Laurence?

Five of these just to get 8 to run.......and isp running fast and all the other pc's unplugged.


&^%$#@

Five of these just to get 8 to run.......and isp running fast and all the other pc's unplugged.


&^%$#@

Then 80 Valids in a row and once again trying to start up 10 more......one at a time....and 4 more of these "getting time from the pool Failed" running at 3Mbps

The server sure is slow at sending these out the last 24 hours and it says it has plenty to send but right now I had to run that 60 seconds 3 times just to get 3 tasks.

Well it 2am and I'm running at full speed internet ( fast as 50Mbs) yet I just got 4....make that 5 ....no make that 6 of these in a row after starting up several of them one at a time.

this is one of those times I wish that crow would fly over Cern and drop that bread on a power supply and shut the server down

Yesterday 16 Theory tasks finished including ONE bad.
Today up to now 8 finished including ONE bad.
http://mcplots-dev.cern.ch/cache/stats/host-2-378-4064.txt
All Theory tasks are correct shown:
https://lhcathomedev.cern.ch/lhcathome-dev/results.php?hostid=4064
Is there a problem with mcplots?

Is there a problem with mcplots?

I could find 1 task of those 2 errors with

Container 'runc' finished with status code 1.

That job had an error on that PC, but BOINC is not aware of it.

Thank you Crystal,
you have a deeper look with the watching eyes ;-)

https://lhcathomedev.cern.ch/lhcathome-dev/result.php?resultid=2916002

Well this is a nice long one.....not Valid......but it did run over 10 days and even claimed to run the CPU the entire time.

I had two Sherpa's doing this after several that did run Valids

Well as usual this is why you can't let them run without watching them.



After starting up lots of them on other hosts and all of the Theory tasks over at LHC I just got 3 of these in a row to Abort.

So I guess I will stop at 3 since I don't like seeing more Computer Errors on my stats when I know it isn't my computer having the errors.

It isn't my internet speed either since I start them one at a time at 3mbps with all the other hosts suspended and unplugged ethernet ( as I have to do every day with these)

These VB tasks are sure clock dependent and I don't only mean here https://www.pool.ntp.org/zone/ch

As I mentioned a few days ago about the 10 day Invalid Sherpa tasks I had and some after the first two that I just Aborted since they keep doing that.

So wondering if it was just a problem again with both versions (here and over at LHC are different versions)

So I checked another member I know who runs more tasks than most members and uses Windows and he had the same problem every time with that version and I saw it on each of his hosts.

The Sherpa's did work for a while after the last fix but it is back at it again so they will just get aborted or for those who run over 100 threads they will run 10 days are if they get caught doing that Aborted over there too.

You think you could send me something since the server says Theory Simulation 197 Unsent ?
Don't care if they are over 18 hour ones or not.
My new month of high speed starts in a few hours and since I am trapped inside thanks to forest fires on the entire west coast.

Even when I get them lately I only get 10 on any of the 8-cores and I have seen others get over 50

(btw it sure would be nice if these tasks can go from suspended to run and STILL start using the CPU again instead of running until valid but with only 2mins CPU time and the Container 'runc' finished with status code 1) and Run time 7 hours 18 min 21 sec
CPU time 6 min 33 sec

---

Mad Scientist For Life

First time I ever got one of these so I guessed it was an update so I did reboot after the d/l and started them up again and after that it didn't have that on the VM console so I did the same thing on the other host just now.

https://lhcathomedev.cern.ch/lhcathome-dev/result.php?resultid=2926681

It doesn't always have to be a Sherpa to do this.

Here is a Pythia that would have run for 10 days and use the internet and the CPU the entire time before ending up a Failed 0 Credit .....I decided to let it be replaced by a task I might be able to trust (of course I will still watch it)

There is no success so long in the MCProduction matrix:
pythia8 8.301 dire-default
0+12/23

pythia and sherpa still running

pythia and sherpa still running

Are you still getting those 9 day tasks? I have been aborting them and only running the standard 2+ hours tasks.

No it has been a while since getting one of those.
Longest I have got since last year are the 18 hour version and even the Sherpas have been running without that problem.

The Theory tasks have been the shorter ones and the longer ones have been CMS

No it has been a while since getting one of those.
Longest I have got since last year are the 18 hour version and even the Sherpas have been running without that problem.

The Theory tasks have been the shorter ones and the longer ones have been CMS

I've been getting a bunch of the 9 day tasks lately and am still aborting them as this laptop just can't come close to finishing them on time. I wish they would let us choose the tasks types so I could avoid this as today I spent over 5 hours on one task, over 3 hours on another one and over 2 hours on 2 more of them, I am wasting their time and mine BUT at least they count toward my wuprop hours!!