You may also want to check the CERN Grafana pages:
LHC@home menu -> Jobs -> CMS Jobs -> Running Jobs

If you are running CMS tasks but the most recent number there is down to zero you are wasting cycles with empty envelope tasks.

Thanks for the tip. I managed to register an "outsider" account there and I see 500-600 jobs in the "production" category. I assume that means all is good. I do have a question though. The BOINC-based server status page shows over 10k tasks in progress. Why the huge discrepancy? Does it mean that most tasks running are envelope tasks?

---

These are tasks that are sent to the client but not yet started.
When a VM (task) starts it registers via WMAgent/HTCondor to get a job.

BOINC server shows # sent tasks
Grafana shows # registered jobs

Thanks for the insight, it makes sense now!

---

Apologies. I didn't catch the last batch before the job queue drained. A new batch is in the system but it will take an hour or two before a) jobs get into the condor pending queue, and b) BOINC server sees there are new jobs available and starts creating new tasks in the task queue.
Edit: Ha, spoke too soon -- new jobs/tasks are already available!

---

the queue has run dry

and many of the last ones we got were the no work version
https://lhcathome.cern.ch/lhcathome/result.php?resultid=430573034 I had around 40

the queue has run dry

Yes. Looks like the JobSubmitter component in the WMAgent died around midnight. We've restarted it, and jobs are starting to run again.

---

YES they seem to be running here again Ivan.......not over at -dev though

YES they seem to be running here again Ivan.......not over at -dev though

Hi; The vccs-dev machine had been removed from the firewall. Laurence reinstated it, and I now have a job successfully running, so try again.

---

I'm not sure if I should use this topic, as the subject doesn't quite describe my problem.

I definitely have CMS selected in my LHC preferences. The LHC status page has, recently, consistently shown roughly 150 to 200 CMS tasks available (right now: 197), with around 1700 in progress (so apparently some people are getting these tasks). However, I haven't gotten any in months. Even if I suspend all of my other active projects, and "Update" LHC, or let BOINC try to retrieve tasks on its own, I get nothing: "No tasks are available for CMS Simulation".

What am I doing something wrong?

Thanks.

I see this sometimes, I don't really know how to fix it properly, it just fixes itself after a while.

The only way to force is to set you setting to only get CMS and don't accept work from other project if there is non avaliable, but then you only run CMS.

Well, it hasn't fixed itself in some months. The only other LHC application I have selected is XTrack, which hasn't had any tasks in months and doesn't seem to be going anywhere any time soon, so I could unselect that. I already have "accept work from other applications" deselected.

It does seem like there should be a straightforward fix for this, if someone knew what it was.

Thanks.

OK few months seems like too long, maybe there is another setting, like you have #CPU set more than 4 so you could not run them? Your computer ran other tasks OK so it seems like a config that is incorrect.

@doug

Please check if you have "run native ..." enabled at your prefs page (https://lhcathome.cern.ch/lhcathome/prefs.php?subset=project).
This option must be disabled for Windows hosts.

In addition:
Your host reports only 4 cores and 16 GB RAM.
Since VirtualBox recommends not to run VMs with "#cores > maxcores/2" you shouldn't run VMs with more than 2 cores.

So:
- you shouldn't run CMS (as it requires 4-core VMs)
- you should limit ATLAS to a 2-core setup
- you are free to run Theory (best option for this host)

Thank you all for your help and input.

I do NOT have "Run native if available" enabled, as it specifically says "not recommended for Windows".

My current machine does have only 4 cores. I should be getting a new machine with many more cores in a month or so, so I can revisit this situation then.

computezrmle: thanks specifically for your clear list of what applications to run and not run. For now, I'll have to avoid CMS. If my memory serves, I've had problems with Atlas in the past, perhaps because of that 2-core limitation, so I've avoided it in recent times. Your list is a bit of a bummer, as I don't know of any way to limit the cores to 2 for just Atlas.

I guess I'll be limited to Theory for the time being.

Thanks again to all.

1/10/2026 3:16:01 PM | LHC@home | No tasks are available for CMS Simulation

(after trying for an hour I finally got the one I was trying to get)

1/10/2026 3:16:01 PM | LHC@home | No tasks are available for CMS Simulation

(after trying for an hour I finally got the one I was trying to get)

this seems to be a common problem lately. Happens also here once in while. In most cases it then takes between 20 and 30 minutes until tasks finally come in. No idea what's going on.

https://lhcathome.cern.ch/lhcathome/server_status.php

It looks like we ran out of CMS work again and the ones still running appear to not have actual work

In reply to Magic Quantum Mechanic's message of 16 Feb 2026:

https://lhcathome.cern.ch/lhcathome/server_status.php

It looks like we ran out of CMS work again and the ones still running appear to not have actual work

Yes, I missed submitting a new workflow by a couple of hours. More jobs were submitted before we were fully out.

---