Message boards :
Sixtrack Application :
"SixTrack Tasks NOT being distributed
Message board moderation
Previous · 1 · 2 · 3 · 4 · 5 · Next
| Author | Message |
|---|---|
|
Send message Joined: 12 Jul 11 Posts: 857 Credit: 1,619,050 RAC: 0 |
Welcome and thanks for the logfile. Eric. |
 Send message Joined: 28 Sep 04 Posts: 744 Credit: 51,947,201 RAC: 32,071   
|
I just enabled sixtract tasks again because Atlas task downloads are stalled in Boinc and this is what I get: 30-Jul-2017 10:38:58 [LHC@home] update requested by user 30-Jul-2017 10:39:00 [LHC@home] sched RPC pending: Requested by user 30-Jul-2017 10:39:00 [LHC@home] [sched_op] Starting scheduler request 30-Jul-2017 10:39:00 [LHC@home] Sending scheduler request: Requested by user. 30-Jul-2017 10:39:00 [LHC@home] Not requesting tasks: some download is stalled 30-Jul-2017 10:39:00 [LHC@home] [sched_op] CPU work request: 0.00 seconds; 0.00 devices 30-Jul-2017 10:39:00 [LHC@home] [sched_op] NVIDIA GPU work request: 0.00 seconds; 0.00 devices 30-Jul-2017 10:39:00 [LHC@home] [sched_op] Intel GPU work request: 0.00 seconds; 0.00 devices 30-Jul-2017 10:39:02 [LHC@home] Scheduler request completed 30-Jul-2017 10:39:02 [LHC@home] [sched_op] Server version 707 30-Jul-2017 10:39:02 [LHC@home] Project requested delay of 6 seconds 30-Jul-2017 10:39:02 [LHC@home] [sched_op] Deferring communication for 00:00:06 30-Jul-2017 10:39:02 [LHC@home] [sched_op] Reason: requested by project So because there are two Atlas task stalled no LHC tasks can be downloaded either. 
|
|
Send message Joined: 15 Nov 14 Posts: 602 Credit: 24,371,321 RAC: 0
|
So because there are two Atlas task stalled no LHC tasks can be downloaded either. Just abort the ATLAS download and it will resume the others. It happened to me too. |
|
Send message Joined: 28 Sep 04 Posts: 744 Credit: 51,947,201 RAC: 32,071
|
So because there are two Atlas task stalled no LHC tasks can be downloaded either. Yes I know. I just wanted to post it here as one of the reasons when sixtrack tasks can not be downloaded. EDIT: I had to restart Boinc to get the sixtrack downloads flowing again. Boinc kept insisting that there were still some stalled downloads although they were all aborted and reported to the server. After restart I got 15 sixtrack tasks on one request.
|
|
Send message Joined: 14 Jan 10 Posts: 1440 Credit: 9,657,607 RAC: 1,253  |
Data directory: E:\BOINC It's auto-managed. Normally the data storage is on C:\ProgramData\BOINC, but during BOINC installation you have chosen 'Advanced settings', and changed the storage location. |
|
Send message Joined: 12 Jul 11 Posts: 857 Credit: 1,619,050 RAC: 0 |
There are now at least 2 of us who see this as a REAL problem, at least for SixTrack. Hopefully we can pin it down and fix. Eric. |
 Send message Joined: 15 Jun 08 Posts: 2605 Credit: 262,026,635 RAC: 131,259 |
My list of sixtracks with status "validation pending" show a couple of older WUs (since 2017-07-04) that have not yet been sent to any wingman. Be so kind as to check if they are still in the backlog. https://lhcathome.cern.ch/lhcathome/workunit.php?wuid=72521104 https://lhcathome.cern.ch/lhcathome/workunit.php?wuid=72663470 https://lhcathome.cern.ch/lhcathome/workunit.php?wuid=72772214 https://lhcathome.cern.ch/lhcathome/workunit.php?wuid=72772256 https://lhcathome.cern.ch/lhcathome/workunit.php?wuid=72780157 |
|
Send message Joined: 12 Jul 11 Posts: 857 Credit: 1,619,050 RAC: 0 |
The 5 wingman tasks exist but are unsent. Eric. My list of sixtracks with status "validation pending" show a couple of older WUs (since 2017-07-04) that have not yet been sent to any wingman. |
|
Send message Joined: 2 May 07 Posts: 2260 Credit: 175,581,097 RAC: 15,522
|
Duration-time 13k seconds (Windows) against 160k (Linux): https://lhcathome.cern.ch/lhcathome/workunit.php?wuid=73084562 |
|
Send message Joined: 14 Jan 10 Posts: 1440 Credit: 9,657,607 RAC: 1,253 |
Duration-time 13k seconds (Windows) against 160k (Linux): For the Linux task the version without support for SSE2, SSE3 or PNI was used. I don't know why, but that 'old' version is 4 times slower and still used for modern machines. Me was told that the server would know which version is better for the client to be used, but also on my machine (meanwhile having a SixTrack history), I get several versions. 152748590 73077334 31 Jul 2017, 17:21:48 UTC 4 Aug 2017, 3:07:30 UTC Completed and validated 171,710.83 171,657.90 278.46 SixTrack v451.07 x86_64-pc-linux-gnu 152747027 73076542 31 Jul 2017, 16:29:06 UTC 1 Aug 2017, 4:47:30 UTC Completed and validated 42,471.49 42,460.27 266.14 SixTrack v451.07 (sse2) i686-pc-linux-gnu |
|
Send message Joined: 15 Jun 08 Posts: 2605 Credit: 262,026,635 RAC: 131,259 |
The 5 wingman tasks exist but are unsent. Eric. Sorry Eric, I should have been more precise. A look at the server status page makes me guess that the sixtrack queue could be completely distributed whithin roughly 1 week. The oldest WU in my list was generated at 2017-06-29 (5 weeks ago) and is still not distributed to a second computer. Either I'm too unpatient (I apologise in this case) or there may still be a distribution problem that needs further investigation. |
|
Send message Joined: 9 Dec 14 Posts: 202 Credit: 2,533,875 RAC: 0
|
Duration-time 13k seconds (Windows) against 160k (Linux): maybe another reason for the big difference is because the linux machine uses a rather old amd a8 apu and the windows machine is using a pretty new i7 6800k. the benchmark difference is at least a factor of 4 to 5. |
|
Send message Joined: 12 Jul 11 Posts: 857 Credit: 1,619,050 RAC: 0 |
That's terrible! This was supposed to be solved. I'll report it AGAIN! Eric. Duration-time 13k seconds (Windows) against 160k (Linux): |
|
Send message Joined: 29 Feb 16 Posts: 157 Credit: 2,659,975 RAC: 0  |
I don't understand why the completed task does not figure in the summary per app version: https://lhcathome.cern.ch/lhcathome/host_app_versions.php?hostid=10491448 (see SixTrack 451.07 x86_64-pc-linux-gnu) I mean, it says 1 "consecutive valid tasks" but the "number of tasks completed" is 0... Other entries in that table do not seem to by fully consistent... |
|
Send message Joined: 14 Jan 10 Posts: 1440 Credit: 9,657,607 RAC: 1,253 |
I mean, it says 1 "consecutive valid tasks" but the "number of tasks completed" is 0... Like the Average turnaround time 0.00 days, although the example task's turnaround time was 2.81 days. On my Linux machine there are those details: SixTrack 451.07 x86_64-pc-linux-gnu Number of tasks completed 116 Max tasks per day 501 Number of tasks today 0 Consecutive valid tasks 1 Average processing rate 5.08 GFLOPS Average turnaround time 0.18 days |
|
Send message Joined: 19 Feb 08 Posts: 708 Credit: 4,336,250 RAC: 0
|
I can confirm that SixTrack tasks on my Window 10 PC take about 16k seconds, those on the 2 Linux boxes about ten times longer. They are the only LHC tasks which do not fail on my 3 PCs, for unknown reasons. VirtualBox is the latest on all 3. Tullio |
|
Send message Joined: 22 Mar 17 Posts: 30 Credit: 360,676 RAC: 0
|
A valid task in one that the host computed successfully and returned and the task validated ok. A completed task is valid task that is not outlier. So with SixTrack it's quite normal to have completed less than valid. I wouldn't mind a better term but I can't come up with anything that explains the difference but is still short. |
|
Send message Joined: 29 Feb 16 Posts: 157 Credit: 2,659,975 RAC: 0 |
A completed task is valid task that is not outlier. Good catch - it is the case. |
|
Send message Joined: 29 Feb 16 Posts: 157 Credit: 2,659,975 RAC: 0 |
Hi Tullio, I could find 4 machine of yours: - https://lhcathome.cern.ch/lhcathome/results.php?hostid=10464329 - windows - 1 6T task, of reasonable CPU time (given the title, I can guess what it is about); - https://lhcathome.cern.ch/lhcathome/results.php?hostid=10492633 - Linux - 1 6T task, error while downloading - have you re-tried to ask for new work? - https://lhcathome.cern.ch/lhcathome/results.php?hostid=10492932 - Linux - 1 6T task, on-going; - https://lhcathome.cern.ch/lhcathome/results.php?hostid=10454176 - Linux - a few 6T tasks, with the way large CPU times. This is really weird... Any mis-configuration? |
|
Send message Joined: 1 Feb 17 Posts: 1 Credit: 13,936 RAC: 0 |
I have been a member of LHC@Home since Feb 2017, but only have been able to process 4 work units. Not sure if this is the same problem as the "SixTrack Tasks," but I'm posting my info here to see if it helps me acquire more work units. I've looked for instructions on how to find the "client logging report," but a search of your site and BOINC help revealed no info, so I don't know how to show it to you. I guess if you want me to process data for you, I'll need some help and direction. Thank you. --Renata |
©2025 CERN