Message boards :
CMS Application :
no new WUs available
Message board moderation
Previous · 1 . . . 15 · 16 · 17 · 18 · 19 · 20 · 21 . . . 24 · Next
| Author | Message |
|---|---|
|
Send message Joined: 18 Dec 15 Posts: 1821 Credit: 118,946,684 RAC: 18,876   
|
the queue has run dry again |
 Magic Quantum Mechanic Send message Joined: 24 Oct 04 Posts: 1176 Credit: 54,887,670 RAC: 4,726  
|
the queue has run dry again That figures.....I had to do the Windows 10 Updates on 4 of mine and when I finally get done I tried to get another Atlas......gone.......then I figured I guess I can just get back to CMS......gone. By the time I try to get Theory some Threadrippers will eat all of those too. |
|
Send message Joined: 2 May 07 Posts: 2244 Credit: 173,902,375 RAC: 374
|
By the time I try to get Theory some Threadrippers will eat all of those too. Total number of generated events: 6014.9 billions The Theory Team have an eye on it. |
 Send message Joined: 28 Sep 04 Posts: 732 Credit: 49,367,266 RAC: 17,281
|
The queue is empty again. Friday the 13th and weekend coming... 
|
|
Send message Joined: 18 Dec 15 Posts: 1821 Credit: 118,946,684 RAC: 18,876
|
The queue is empty again. Friday the 13th and weekend coming...this recently permanent on and off has become quite troublesome :-( |
|
Send message Joined: 18 Dec 15 Posts: 1821 Credit: 118,946,684 RAC: 18,876
|
since this afternoon, there are no jobs being provided for the tasks which can still be downloaded. Obviously, the automatic stop of task delivery in case of no jobs available does not work :-( |
 Send message Joined: 29 Aug 05 Posts: 1061 Credit: 7,737,455 RAC: 245 |
Sorry, there have been some disruptions that we can't control. At the moment I have several workflows stalled in the Agent for reasons that I have yet to ascertain.  |
|
Send message Joined: 18 Dec 15 Posts: 1821 Credit: 118,946,684 RAC: 18,876
|
Sorry, there have been some disruptions that we can't control. At the moment I have several workflows stalled in the Agent for reasons that I have yet to ascertain.hello Ivan, nice to see you back :-) Hope you are fully okay now, healthwise! Thanks for your efforts to make CMS run again (CMS is definitely my favorite subproject)! |
|
Send message Joined: 29 Aug 05 Posts: 1061 Credit: 7,737,455 RAC: 245 |
Sorry, there have been some disruptions that we can't control. At the moment I have several workflows stalled in the Agent for reasons that I have yet to ascertain.hello Ivan, nice to see you back :-) Hope you are fully okay now, healthwise! Hi, good to be back. I'm now undergoing a "phased" transition back to my duties and will be officially back to my "contracted hours" (i.e. 50%) from January. As for today's problems, I'm suspecting that a certificate[1] has expired, and I'm trying to track down someone to check it. [1] CN=Robot: WmCore Service Account |
 Send message Joined: 15 Jun 08 Posts: 2541 Credit: 254,608,838 RAC: 28,391 |
... good to be back. I'm now undergoing a "phased" transition back to my duties and will be officially back ... +1 +1 +1 |
|
Send message Joined: 29 Aug 05 Posts: 1061 Credit: 7,737,455 RAC: 245 |
As for today's problems, I'm suspecting that a certificate[1] has expired, and I'm trying to track down someone to check it. OK, we have some jobs running again, but not my usual workflows as yet. These will almost certainly have different performance profiles, as we are trying to get a different calculation running. Let us know how they perform. |
|
Send message Joined: 15 Jun 08 Posts: 2541 Credit: 254,608,838 RAC: 28,391 |
Got this task: https://lhcathome.cern.ch/lhcathome/result.php?resultid=402719666 After some downloads and the usual benchmark runs CPU usage dropped to effectively 0%. No "cmsRun" process at the top console. No try to contact the WMAgent service. Nonetheless, glidein reported "0" = success. |
|
Send message Joined: 14 Jan 10 Posts: 1422 Credit: 9,484,585 RAC: 1,038   |
Now I got a real sub-task from Ivan's flow: ireid_TC_SLC7_IDR_CMS_Home_231206_131958_9405 |
|
Send message Joined: 18 Dec 15 Posts: 1821 Credit: 118,946,684 RAC: 18,876
|
I have re-started CMS on some of my machines - everything seems to work fine. What seems to me is that the new series ("CMS_141....) consumes less memory. |
|
Send message Joined: 15 Jun 08 Posts: 2541 Credit: 254,608,838 RAC: 28,391 |
OK, we have some jobs running again, but not my usual workflows as yet. These will almost certainly have different performance profiles, as we are trying to get a different calculation running. Let us know how they perform. They run much longer than the standard CMS tasks. Looks like they process more than 5 times the #events. Thus, the runtimes are too close to the hard 18 h task limit which causes a couple of them to be shut down by the BOINC watchdog. In this case BOINC marks the task as valid but they don't return scientific results. |
|
Send message Joined: 15 Jun 08 Posts: 2541 Credit: 254,608,838 RAC: 28,391 |
Looks like the backend queue again doesn't send CMS subtasks. But the project server doesn't notice it and continues generating empty envelope tasks. |
|
Send message Joined: 18 Dec 15 Posts: 1821 Credit: 118,946,684 RAC: 18,876
|
Looks like the backend queue again doesn't send CMS subtasks.there was the same problem last week - Ivan, could you please look into this, so that once no subtasks are available, the generation of empty envelope tasks is stopped. This worked well some time ago, so this mechanism obviously got broken at some point of time, and was not repaired so far. |
|
Send message Joined: 18 Dec 15 Posts: 1821 Credit: 118,946,684 RAC: 18,876
|
no tasks available; this time, the automatic stop mechanism for submitting tasks if no subtasks are available seemed to work well. |
|
Send message Joined: 18 Dec 15 Posts: 1821 Credit: 118,946,684 RAC: 18,876
|
no tasks available; this time, the automatic stop mechanism for submitting tasks if no subtasks are available seemed to work well.some time later, new tasks could be downloaded. Today, again no new tasks ... :-( |
|
Send message Joined: 2 May 07 Posts: 2244 Credit: 173,902,375 RAC: 374
|
Win11pro, Boinc 7.24.1 The flow for a new CMS-Task (only one is running here) need one hour. Problem of the scheduler? https://lhcathome.cern.ch/lhcathome/results.php?hostid=10795955 |
©2024 CERN