LHC@home

There was a backlog of unsent Sixtrack WUs again, we have now removed some Sixtrack specific legacy scheduling options in the BOINC server configuration of the consolidated project to see if this should speed things up.

These settings could lead to rejection of tasks as discussed under the MAC app thread.

Could there be another problem with work distribution? I see 100Ks of sixtrack WUs ready to send, but I haven't been able to keep my caches full for over 24 hours.

As far as we can see, daemons and work distribution is ok. However, the current Sixtrack tasks are rather short, so it might be that the quota runs full pretty quickly. A reset may get new tasks. (My computer has been getting a steady stream of Sixtrack tasks over the weekend.)

Will do a daemon restart on the server, just in case.

A reset may get new tasks.

A reset also means I've to download 8 GB again for the 4 VM-vdi-files on just 1 computer.
Current running VM's will be killed too.

No good solution!

However, the current Sixtrack tasks are rather short, so it might be that the quota runs full pretty quickly...

What quota is that? Is there something other than 14 in-progress tasks/core? I should have added that I'm getting "No tasks are available for SixTrack" rather than something indicating I've reached a time-related or other quota.

However, the current Sixtrack tasks are rather short...

I'm wondering if the opposite is true, in some cases. I just noticed that I'm getting "Not requesting tasks: don't need...job cache full". Two hosts getting this message have tasks in their queues with 6-8 hour estimated run times, which, I think, is much longer than most recent tasks.

Currently there are less WU's to crack but I am not getting those anymore. I did during the afternoon.
Have I reached my maximum? If so can that be changed?

Thank you.

---

Greetings from,
TJ

I got a few Sixtracks yesterday and overnight by setting my Preferences to only accept Sixtrack and letting my cached Theory tasks run down. I have had a mixture of short, 4 min, and long, 6 hour, tasks.

Are there any special criteria for getting Sixtrack tasks. My client is ticked to receive them but I get very few, most of the time none even when there are tasks shown in the unsent display . I am on W7.

This is an ongoing problem, I have submitted a ticket to our IT support
and we shall see. Apologies. Eric.

---

It really is a shame: LHC takes part in this year's BOINC Pentathlon but does not deliver Sixtrack tasks even in the minimally required amounts.

Now you had a nice opportunity to really acquire some worth-to-mention compute power and what are you doing? Possibly having a coffee?
Well, at least send one over to us, too!

Michael.

---

There will be Sixtrack tasks coming, we have issued a reminder to our users to generate more work that will soon come in the pipeline.

Meanwhile, please feel free to churn our other applications.

Meanwhile, please feel free to churn our other applications.

Those other applications either require Virtualbox or a constant Internet connection. So, many of us can't support these generally and especially not before the LHC competition starts (For your information: to make it in the contest, teams acquire huge loads of tasks many days before the race actually starts. They bunker them, i.e. compute these tasks offline without uploading the results - which is done only once the race's start date arrives. Got the problem with the other apps / tasks now?).

Michael.

---

Virtualbox is no option. It is PITA to get this running on headless server (I had no success) and my experience with LHC (ATLAS) is you have to babysitt these instances.
Means for me: Bye LHC.

Well it is true that the Virtual Box combo is more demanding in terms of setup.

However there has been a steady flow Sixtrack tasks over the last week and there are still many more to crunch.

There are over 650,000 WU's but I can't receive them, have reloaded BOINC twice but nothing happening.

since couple of hours one of my computers isnt able to get any sixtracks tasks anymore. boinc shows no sixtrack tasks available.

16.09.2017 13:25:59 | LHC@home | Requesting new tasks for CPU and AMD/ATI GPU

16.09.2017 13:26:00 | LHC@home | Scheduler request completed: got 0 new tasks

16.09.2017 13:26:00 | LHC@home | No tasks sent

16.09.2017 13:26:00 | LHC@home | No tasks are available for SixTrack

16.09.2017 13:26:00 | LHC@home | No tasks are available for sixtracktest

EDIT: the moment this post was written it downloaded new tasks

I have not gotten any sixtrack for around 12 hours, with the same message. I just assumed it was the weekend bug that occurs when no one is around to fix it.

Nothing for my 32 cores so down to my last 5 tasks.
And if I switch back to VB tasks those take 18 hours.

There is a major problem with the distribution of tasks (dating from the
integration of the other projects with SixTrack). I myself am suffering
like you. My 4 processor Windows 10 couldn't get work either. I tried
reset, tried a re-install of the BOINC manager, etc and finally I got 100
Tasks! I was now using the client defaults, although I normally don't
cache tasks. So it seems you either get no tasks or too many. The problem
is being looked at but no solution seems to be available. I agree the
VirtualBox message is annoying but the IT emphasis is on trying to
get more clients running for the other sub-projects. In addition the
BOINC project management pages are so slow to access, several
minutes for each click, as to be unusable. This makes it very difficult for
me to follow up on other problems. I think we still have some HyperThreading/
BIOS problem, process creation failing on Windows, tasks not being
distributed. The slow start up on Windows 10 seems to be solved with the
new executables, which seem to be running well. The overall error rate
is down to about 2.5%, but we can do better.
Apologies and I await action from support. Eric.

---