LHC@home

Restarting feeder, new tasks available in ~ 15 minutes.

Thanks, Nils.
Got work for CMS and Theory after your restart + grace period.
ATLAS still mentions "No work available".

<edit>
As of now, CMS has also no more work.
</edit>

As of now, CMS has also no more work.

Yes, here too.

---

As of now, CMS has also no more work.

Yes, here too.

Hmm, that may have been because my work machines already had "enough" jobs. My home machine got a CMS job 10 minutes later.

---

Gave up on Sixtrack and tried coming back to ATLAS.. *sigh*.

Managed to get 4 to d/l on one machine but the other machine stubbornly continues to get the no tasks available error.

... but the other machine stubbornly continues to get the no tasks available error.

same thing here with one of my machines: neither ATLAS, nor CMS, nor LHCb, nor Theory can download. Although there should be plenty available, according to the Project Status Page.

Very strange :-(

Seemed to be in the backed up condition.

Did other things for 15 minutes came back and did a single update and got 4 ATLAS.

Set the [work_fetch_debug]1[/work_fetch_debug] state and will see what that says if this happens again.

The ATLAS tasks are very short (2,xxx sec, 29 credit) compared to any I've seen before.

Server status page lists 7648 unsent tasks, but repeatedly ...

So 16 Sep 2018 08:29:19 CEST | LHC@home | No tasks are available for ATLAS Simulation

All of my clients are affected.
I guess the server needs a few good words.

Same here.

EDIT:
After the last cached task completed and uploaded it gave me 2 new tasks. It normally allows my 4 CPU host to cache 4 tasks (in preferences I have "Max # tasks" = unlimited). Doing ATLAS native not ATLAS VBox.

Server status page lists 7648 unsent tasks, but repeatedly ...

So 16 Sep 2018 08:29:19 CEST | LHC@home | No tasks are available for ATLAS Simulation

All of my clients are affected.

from this morning on, I've been experiencing the same problem :-(

The interesting thing is that the server status, after about 12 hours now, still shows exactly 7648 unsent tasks. Which implies that not a single task has been sent out all day long?

I remember that we had this for a while already about 8-10 months ago.

One night later, right now, the "unsent tasks" figure on the Project Status Pate is still showing 7648.
So, obviously, no ATLAS tasks could be downloaded since around yesterday morning.

Thanks for reporting, I see the same on my BOINC clients. We are looking into this. The server components (feeder, scheduler etc) all seem to run fine, but for some reason the ATLAS tasks are not selected by the feeder and distributed. AFAIK it is not the same problem as we had seen in the past.

We have switched to our fail-over server and ATLAS tasks should again be distributed. Please note that there will be some pauses and server timeouts over the next days, as our server hosts are being rebooted.

Seems to be working. Thank you Nils.

Since 2:30 UTC there is no flow of new work.

Since 2:30 UTC there is no flow of new work.

We had a problem overnight with the machine at CERN which submits tasks to LHC@Home but it's up and running again and new tasks are flowing in now.

We are losing the ready to send queue on server status page. We are now below 2000 tasks vs the usual 9300. I hope that somebody checks the situation before we run out of tasks.

---

We are losing the ready to send queue on server status page. We are now below 2000 tasks vs the usual 9300. I hope that somebody checks the situation before we run out of tasks.

Seems to be the new normal now ~1950 tasks ready to be sent.

---

2432 available now. Maybe the new normal = no normal. I bet it drops to 0 before Sunday due to yet another "blockage" or "infrastructure failure". Maybe they'll come up with some new name to blame it on.

Meanwhile they continue doling out thousands of tasks to hosts that fail 100 tasks every day and haven't returned a successful result for as far back as their results list go. Absolutely amazing.

bronco wrote:

... I bet it drops to 0 before Sunday ...

Seems you lost your bet.

In our region lost bets of those kind must be paid.
Usually in barrels of beer (or something equivalent) at the next team BBQ.
David Cameron may send you a BBQ schedule.
:-)

In my region to win beer you must first have the conviction to put some of your own beer on the line. You put nothing on the line prior to the outcome therefore there was no bet. It appears you thought there was a high probability of it going to 0 too.

I'll send a barrel of our fine Canadian beer but I warn you now... after a few glasses of ours you won't be able to go back to your region's beer :)