Well, I just found that it does pay to do a detach-reattach. I ran BRP4s only up till tonight, but they're no longer being supplied to GPUs. So ran my cache empty, reported everything. Then detached and reattached. Before detach, the Einstein directory used 2.5GB of disk space. Now after reattach, it's only 56MB. Only BRP5 data files, no Gravitational Wave search files. It really makes a difference. :-)
I´m missing on the SERVER STATUS page at the lower right corner a data field called "BRP5 progress".
Me too. But for various technical reasons this will take a while. For one the first ~30k WUs that were sent out last weekend have to be completely processed. You will see that this happened when the number of "BRP5 Workunits waiting for assimilation" is dropping again.
BM
We've reached the 30000 BRP5 Wu waiting to be assimilated
We've reached the 30000 BRP5 Wu waiting to be assimilated
Thanks.
I have to wait for 636 WUs to be finished that were generated two weeks ago (first weekend of BRP5 before the validator was running) before I can start assimilation again.
Not sure if this has anything to do with no GPU work being available:
6/11/2013 10:42:08 AM | Einstein@Home | Sending scheduler request: To fetch work.
6/11/2013 10:42:08 AM | Einstein@Home | Requesting new tasks for ATI
6/11/2013 10:42:15 AM | Einstein@Home | Scheduler request completed: got 0 new tasks
6/11/2013 10:42:15 AM | Einstein@Home | No work sent
6/11/2013 10:42:15 AM | Einstein@Home | No work is available for Binary Radio Pulsar Search (Perseus Arm Survey)
6/11/2013 10:42:15 AM | Einstein@Home | see scheduler log messages on http://einstein.phys.uwm.edu//host_sched_logs/6889/6889983
6/11/2013 10:42:15 AM | Einstein@Home | No work available for the applications you have selected. Please check your preferences on the web site.
None of my machines can get work and most of them are completely out of BRP5 WUs.
I'm seeing the same thing even though the server status pages claims there are 5,350 BRP5 tasks ready to send:
[pre]11/06/2013 18:08:04 | Einstein@Home | update requested by user
11/06/2013 18:08:08 | Einstein@Home | [sched_op] Starting scheduler request
11/06/2013 18:08:08 | Einstein@Home | Sending scheduler request: Requested by user.
11/06/2013 18:08:08 | Einstein@Home | Requesting new tasks for NVIDIA
11/06/2013 18:08:08 | Einstein@Home | [sched_op] CPU work request: 0.00 seconds; 0.00 devices
11/06/2013 18:08:08 | Einstein@Home | [sched_op] NVIDIA work request: 27420.11 seconds; 1.00 devices
11/06/2013 18:08:11 | Einstein@Home | Scheduler request completed: got 0 new tasks
11/06/2013 18:08:11 | Einstein@Home | [sched_op] Server version 611
11/06/2013 18:08:11 | Einstein@Home | No work sent
11/06/2013 18:08:11 | Einstein@Home | No work is available for Binary Radio Pulsar Search (Perseus Arm Survey)
11/06/2013 18:08:11 | Einstein@Home | No work available for the applications you have selected. Please check your preferences on the web site.
11/06/2013 18:08:11 | Einstein@Home | Project requested delay of 60 seconds
11/06/2013 18:08:11 | Einstein@Home | [sched_op] Deferring communication for 1 min 0 sec
11/06/2013 18:08:11 | Einstein@Home | [sched_op] Reason: requested by project[/pre]
Something with the scheduler appears to be wrong...
5572 Einstein@Home 6/11/2013 11:47:14 AM Sending scheduler request: To fetch work.
5573 Einstein@Home 6/11/2013 11:47:14 AM Requesting new tasks for NVIDIA
5574 Einstein@Home 6/11/2013 11:47:18 AM Scheduler request completed: got 0 new tasks
5575 Einstein@Home 6/11/2013 11:47:18 AM No work sent
5576 Einstein@Home 6/11/2013 11:47:18 AM see scheduler log messages on http://einstein.phys.uwm.edu//host_sched_logs/6767/6767034
Scheduler log shows...
2013-06-11 15:47:52.0638 [PID=14285] Request: [USER#xxxxx] [HOST#6767034] [IP xxx.xxx.xxx.142] client 7.0.64
2013-06-11 15:47:52.0758 [PID=14285] [send] effective_ncpus 8 max_jobs_on_host_cpu 999999 max_jobs_on_host 999999
2013-06-11 15:47:52.0758 [PID=14285] [send] effective_ngpus 1 max_jobs_on_host_gpu 999999
2013-06-11 15:47:52.0758 [PID=14285] [send] Not using matchmaker scheduling; Not using EDF sim
2013-06-11 15:47:52.0758 [PID=14285] [send] CPU: req 0.00 sec, 0.00 instances; est delay 0.00
2013-06-11 15:47:52.0758 [PID=14285] [send] CUDA: req 20311.03 sec, 0.00 instances; est delay 0.00
2013-06-11 15:47:52.0759 [PID=14285] [send] work_req_seconds: 0.00 secs
2013-06-11 15:47:52.0759 [PID=14285] [send] available disk 45.10 GB, work_buf_min 172800
2013-06-11 15:47:52.0759 [PID=14285] [send] active_frac 0.999972 on_frac 0.994627 DCF 3.199574
2013-06-11 15:47:52.0773 [PID=14285] [send] [HOST#6767034] is reliable
2013-06-11 15:47:52.0774 [PID=14285] [send] set_trust: random choice for error rate 0.000010: yes
2013-06-11 15:47:52.0926 [PID=14285] [version] Don't need CPU jobs, skipping version 104 for hsgamma_FGRP2 ()
2013-06-11 15:47:52.0926 [PID=14285] [version] no app version available: APP#21 (hsgamma_FGRP2) PLATFORM#9 (windows_x86_64) min_version 0
2013-06-11 15:47:52.0926 [PID=14285] [version] no app version available: APP#21 (hsgamma_FGRP2) PLATFORM#2 (windows_intelx86) min_version 0
2013-06-11 15:47:52.0926 [PID=14285] [version] Checking plan class 'BRP4X64'
2013-06-11 15:47:52.0933 [PID=14285] [version] reading plan classes from file '/BOINC/projects/EinsteinAtHome/plan_class_spec.xml'
2013-06-11 15:47:52.0933 [PID=14285] [version] project prefs setting 'also_run_cpu' (1.000000) prevents using plan class.
2013-06-11 15:47:52.0933 [PID=14285] [version] Checking plan class 'BRP4SSE'
2013-06-11 15:47:52.0934 [PID=14285] [version] project prefs setting 'also_run_cpu' (1.000000) prevents using plan class.
2013-06-11 15:47:52.0934 [PID=14285] [version] no app version available: APP#19 (einsteinbinary_BRP4) PLATFORM#9 (windows_x86_64) min_version 0
2013-06-11 15:47:52.0934 [PID=14285] [version] no app version available: APP#19 (einsteinbinary_BRP4) PLATFORM#2 (windows_intelx86) min_version 0
2013-06-11 15:47:52.9027 [PID=14285] [version] Checking plan class 'SSE2'
2013-06-11 15:47:52.9027 [PID=14285] [version] plan class ok
2013-06-11 15:47:52.9028 [PID=14285] [version] Don't need CPU jobs, skipping version 104 for einstein_S6BucketLVE (SSE2)
2013-06-11 15:47:52.9028 [PID=14285] [version] no app version available: APP#22 (einstein_S6BucketLVE) PLATFORM#9 (windows_x86_64) min_version 0
2013-06-11 15:47:52.9028 [PID=14285] [version] no app version available: APP#22 (einstein_S6BucketLVE) PLATFORM#2 (windows_intelx86) min_version 0
2013-06-11 15:47:52.9043 [PID=14285] [debug] [HOST#6767034] MSG(high) No work sent
2013-06-11 15:47:52.9043 [PID=14285] [debug] [HOST#6767034] MSG(high) see scheduler log messages on http://einstein.phys.uwm.edu//host_sched_logs/6767/6767034
2013-06-11 15:47:52.9043 [PID=14285] Sending reply to [HOST#6767034]: 0 results, delay req 60.00
2013-06-11 15:47:52.9046 [PID=14285] Scheduler ran 0.848 seconds
Is the scheduler even responding to the request for GPU WU's?
my two hosts stopped getting BRP GPU tasks as well, and i'm seeing the same messages as you guys in the event log. i went ahead and cut-n-pasted the URL in the event log into firefox, and this is what it reports:
2013-06-11 16:01:18.6662 [PID=18031] Request: [USER#xxxxx] [HOST#4505391] [IP xxx.xxx.xxx.6] client 7.0.64
2013-06-11 16:01:18.7011 [PID=18031] [send] effective_ncpus 2 max_jobs_on_host_cpu 999999 max_jobs_on_host 999999
2013-06-11 16:01:18.7011 [PID=18031] [send] effective_ngpus 2 max_jobs_on_host_gpu 999999
2013-06-11 16:01:18.7011 [PID=18031] [send] Not using matchmaker scheduling; Not using EDF sim
2013-06-11 16:01:18.7011 [PID=18031] [send] CPU: req 0.00 sec, 0.00 instances; est delay 0.00
2013-06-11 16:01:18.7012 [PID=18031] [send] CUDA: req 41079.29 sec, 0.00 instances; est delay 0.00
2013-06-11 16:01:18.7012 [PID=18031] [send] work_req_seconds: 0.00 secs
2013-06-11 16:01:18.7012 [PID=18031] [send] available disk 15.32 GB, work_buf_min 259200
2013-06-11 16:01:18.7012 [PID=18031] [send] active_frac 0.999941 on_frac 0.998303 DCF 2.713785
2013-06-11 16:01:18.7062 [PID=18031] [send] [HOST#4505391] is reliable
2013-06-11 16:01:18.7063 [PID=18031] [send] set_trust: random choice for error rate 0.000010: yes
2013-06-11 16:01:18.7188 [PID=18031] [version] Checking plan class 'BRP4X64'
2013-06-11 16:01:18.7192 [PID=18031] [version] reading plan classes from file '/BOINC/projects/EinsteinAtHome/plan_class_spec.xml'
2013-06-11 16:01:18.7193 [PID=18031] [version] project prefs setting 'also_run_cpu' (1.000000) prevents using plan class.
2013-06-11 16:01:18.7193 [PID=18031] [version] Checking plan class 'BRP4SSE'
2013-06-11 16:01:18.7193 [PID=18031] [version] project prefs setting 'also_run_cpu' (1.000000) prevents using plan class.
2013-06-11 16:01:18.7193 [PID=18031] [version] no app version available: APP#19 (einsteinbinary_BRP4) PLATFORM#9 (windows_x86_64) min_version 0
2013-06-11 16:01:18.7193 [PID=18031] [version] no app version available: APP#19 (einsteinbinary_BRP4) PLATFORM#2 (windows_intelx86) min_version 0
2013-06-11 16:01:18.7194 [PID=18031] [version] Don't need CPU jobs, skipping version 104 for hsgamma_FGRP2 ()
2013-06-11 16:01:18.7194 [PID=18031] [version] no app version available: APP#21 (hsgamma_FGRP2) PLATFORM#9 (windows_x86_64) min_version 0
2013-06-11 16:01:18.7194 [PID=18031] [version] no app version available: APP#21 (hsgamma_FGRP2) PLATFORM#2 (windows_intelx86) min_version 0
2013-06-11 16:01:20.3568 [PID=18031] [version] Checking plan class 'SSE2'
2013-06-11 16:01:20.3568 [PID=18031] [version] plan class ok
2013-06-11 16:01:20.3569 [PID=18031] [version] Don't need CPU jobs, skipping version 104 for einstein_S6BucketLVE (SSE2)
2013-06-11 16:01:20.3569 [PID=18031] [version] no app version available: APP#22 (einstein_S6BucketLVE) PLATFORM#9 (windows_x86_64) min_version 0
2013-06-11 16:01:20.3569 [PID=18031] [version] no app version available: APP#22 (einstein_S6BucketLVE) PLATFORM#2 (windows_intelx86) min_version 0
2013-06-11 16:01:20.3579 [PID=18031] [debug] [HOST#4505391] MSG(high) No work sent
2013-06-11 16:01:20.3580 [PID=18031] [debug] [HOST#4505391] MSG(high) see scheduler log messages on http://einstein.phys.uwm.edu//host_sched_logs/4505/4505391
2013-06-11 16:01:20.3580 [PID=18031] Sending reply to [HOST#4505391]: 0 results, delay req 60.00
2013-06-11 16:01:20.3583 [PID=18031] Scheduler ran 1.699 seconds
Well, I just found that it
)
Well, I just found that it does pay to do a detach-reattach. I ran BRP4s only up till tonight, but they're no longer being supplied to GPUs. So ran my cache empty, reported everything. Then detached and reattached. Before detach, the Einstein directory used 2.5GB of disk space. Now after reattach, it's only 56MB. Only BRP5 data files, no Gravitational Wave search files. It really makes a difference. :-)
RE: I´m missing on the
)
We've reached the 30000 BRP5 Wu waiting to be assimilated
RE: We've reached the 30000
)
Thanks.
I have to wait for 636 WUs to be finished that were generated two weeks ago (first weekend of BRP5 before the validator was running) before I can start assimilation again.
Is this causing problems for anyone out there?
BM
BM
Not sure if this has anything
)
Not sure if this has anything to do with no GPU work being available:
None of my machines can get work and most of them are completely out of BRP5 WUs.
I'm seeing the same thing
)
I'm seeing the same thing even though the server status pages claims there are 5,350 BRP5 tasks ready to send:
[pre]11/06/2013 18:08:04 | Einstein@Home | update requested by user
11/06/2013 18:08:08 | Einstein@Home | [sched_op] Starting scheduler request
11/06/2013 18:08:08 | Einstein@Home | Sending scheduler request: Requested by user.
11/06/2013 18:08:08 | Einstein@Home | Requesting new tasks for NVIDIA
11/06/2013 18:08:08 | Einstein@Home | [sched_op] CPU work request: 0.00 seconds; 0.00 devices
11/06/2013 18:08:08 | Einstein@Home | [sched_op] NVIDIA work request: 27420.11 seconds; 1.00 devices
11/06/2013 18:08:11 | Einstein@Home | Scheduler request completed: got 0 new tasks
11/06/2013 18:08:11 | Einstein@Home | [sched_op] Server version 611
11/06/2013 18:08:11 | Einstein@Home | No work sent
11/06/2013 18:08:11 | Einstein@Home | No work is available for Binary Radio Pulsar Search (Perseus Arm Survey)
11/06/2013 18:08:11 | Einstein@Home | No work available for the applications you have selected. Please check your preferences on the web site.
11/06/2013 18:08:11 | Einstein@Home | Project requested delay of 60 seconds
11/06/2013 18:08:11 | Einstein@Home | [sched_op] Deferring communication for 1 min 0 sec
11/06/2013 18:08:11 | Einstein@Home | [sched_op] Reason: requested by project[/pre]
Something with the scheduler appears to be wrong...
I'm glad I'm not the only
)
I'm glad I'm not the only one... :D
5572 Einstein@Home 6/11/2013 11:47:14 AM Sending scheduler request: To fetch work.
5573 Einstein@Home 6/11/2013 11:47:14 AM Requesting new tasks for NVIDIA
5574 Einstein@Home 6/11/2013 11:47:18 AM Scheduler request completed: got 0 new tasks
5575 Einstein@Home 6/11/2013 11:47:18 AM No work sent
5576 Einstein@Home 6/11/2013 11:47:18 AM see scheduler log messages on http://einstein.phys.uwm.edu//host_sched_logs/6767/6767034
Scheduler log shows...
2013-06-11 15:47:52.0638 [PID=14285] Request: [USER#xxxxx] [HOST#6767034] [IP xxx.xxx.xxx.142] client 7.0.64
2013-06-11 15:47:52.0758 [PID=14285] [send] effective_ncpus 8 max_jobs_on_host_cpu 999999 max_jobs_on_host 999999
2013-06-11 15:47:52.0758 [PID=14285] [send] effective_ngpus 1 max_jobs_on_host_gpu 999999
2013-06-11 15:47:52.0758 [PID=14285] [send] Not using matchmaker scheduling; Not using EDF sim
2013-06-11 15:47:52.0758 [PID=14285] [send] CPU: req 0.00 sec, 0.00 instances; est delay 0.00
2013-06-11 15:47:52.0758 [PID=14285] [send] CUDA: req 20311.03 sec, 0.00 instances; est delay 0.00
2013-06-11 15:47:52.0759 [PID=14285] [send] work_req_seconds: 0.00 secs
2013-06-11 15:47:52.0759 [PID=14285] [send] available disk 45.10 GB, work_buf_min 172800
2013-06-11 15:47:52.0759 [PID=14285] [send] active_frac 0.999972 on_frac 0.994627 DCF 3.199574
2013-06-11 15:47:52.0773 [PID=14285] [send] [HOST#6767034] is reliable
2013-06-11 15:47:52.0774 [PID=14285] [send] set_trust: random choice for error rate 0.000010: yes
2013-06-11 15:47:52.0926 [PID=14285] [version] Don't need CPU jobs, skipping version 104 for hsgamma_FGRP2 ()
2013-06-11 15:47:52.0926 [PID=14285] [version] no app version available: APP#21 (hsgamma_FGRP2) PLATFORM#9 (windows_x86_64) min_version 0
2013-06-11 15:47:52.0926 [PID=14285] [version] no app version available: APP#21 (hsgamma_FGRP2) PLATFORM#2 (windows_intelx86) min_version 0
2013-06-11 15:47:52.0926 [PID=14285] [version] Checking plan class 'BRP4X64'
2013-06-11 15:47:52.0933 [PID=14285] [version] reading plan classes from file '/BOINC/projects/EinsteinAtHome/plan_class_spec.xml'
2013-06-11 15:47:52.0933 [PID=14285] [version] project prefs setting 'also_run_cpu' (1.000000) prevents using plan class.
2013-06-11 15:47:52.0933 [PID=14285] [version] Checking plan class 'BRP4SSE'
2013-06-11 15:47:52.0934 [PID=14285] [version] project prefs setting 'also_run_cpu' (1.000000) prevents using plan class.
2013-06-11 15:47:52.0934 [PID=14285] [version] no app version available: APP#19 (einsteinbinary_BRP4) PLATFORM#9 (windows_x86_64) min_version 0
2013-06-11 15:47:52.0934 [PID=14285] [version] no app version available: APP#19 (einsteinbinary_BRP4) PLATFORM#2 (windows_intelx86) min_version 0
2013-06-11 15:47:52.9027 [PID=14285] [version] Checking plan class 'SSE2'
2013-06-11 15:47:52.9027 [PID=14285] [version] plan class ok
2013-06-11 15:47:52.9028 [PID=14285] [version] Don't need CPU jobs, skipping version 104 for einstein_S6BucketLVE (SSE2)
2013-06-11 15:47:52.9028 [PID=14285] [version] no app version available: APP#22 (einstein_S6BucketLVE) PLATFORM#9 (windows_x86_64) min_version 0
2013-06-11 15:47:52.9028 [PID=14285] [version] no app version available: APP#22 (einstein_S6BucketLVE) PLATFORM#2 (windows_intelx86) min_version 0
2013-06-11 15:47:52.9043 [PID=14285] [debug] [HOST#6767034] MSG(high) No work sent
2013-06-11 15:47:52.9043 [PID=14285] [debug] [HOST#6767034] MSG(high) see scheduler log messages on http://einstein.phys.uwm.edu//host_sched_logs/6767/6767034
2013-06-11 15:47:52.9043 [PID=14285] Sending reply to [HOST#6767034]: 0 results, delay req 60.00
2013-06-11 15:47:52.9046 [PID=14285] Scheduler ran 0.848 seconds
Is the scheduler even responding to the request for GPU WU's?
my two hosts stopped getting
)
my two hosts stopped getting BRP GPU tasks as well, and i'm seeing the same messages as you guys in the event log. i went ahead and cut-n-pasted the URL in the event log into firefox, and this is what it reports:
2013-06-11 16:01:18.6662 [PID=18031] Request: [USER#xxxxx] [HOST#4505391] [IP xxx.xxx.xxx.6] client 7.0.64
2013-06-11 16:01:18.7011 [PID=18031] [send] effective_ncpus 2 max_jobs_on_host_cpu 999999 max_jobs_on_host 999999
2013-06-11 16:01:18.7011 [PID=18031] [send] effective_ngpus 2 max_jobs_on_host_gpu 999999
2013-06-11 16:01:18.7011 [PID=18031] [send] Not using matchmaker scheduling; Not using EDF sim
2013-06-11 16:01:18.7011 [PID=18031] [send] CPU: req 0.00 sec, 0.00 instances; est delay 0.00
2013-06-11 16:01:18.7012 [PID=18031] [send] CUDA: req 41079.29 sec, 0.00 instances; est delay 0.00
2013-06-11 16:01:18.7012 [PID=18031] [send] work_req_seconds: 0.00 secs
2013-06-11 16:01:18.7012 [PID=18031] [send] available disk 15.32 GB, work_buf_min 259200
2013-06-11 16:01:18.7012 [PID=18031] [send] active_frac 0.999941 on_frac 0.998303 DCF 2.713785
2013-06-11 16:01:18.7062 [PID=18031] [send] [HOST#4505391] is reliable
2013-06-11 16:01:18.7063 [PID=18031] [send] set_trust: random choice for error rate 0.000010: yes
2013-06-11 16:01:18.7188 [PID=18031] [version] Checking plan class 'BRP4X64'
2013-06-11 16:01:18.7192 [PID=18031] [version] reading plan classes from file '/BOINC/projects/EinsteinAtHome/plan_class_spec.xml'
2013-06-11 16:01:18.7193 [PID=18031] [version] project prefs setting 'also_run_cpu' (1.000000) prevents using plan class.
2013-06-11 16:01:18.7193 [PID=18031] [version] Checking plan class 'BRP4SSE'
2013-06-11 16:01:18.7193 [PID=18031] [version] project prefs setting 'also_run_cpu' (1.000000) prevents using plan class.
2013-06-11 16:01:18.7193 [PID=18031] [version] no app version available: APP#19 (einsteinbinary_BRP4) PLATFORM#9 (windows_x86_64) min_version 0
2013-06-11 16:01:18.7193 [PID=18031] [version] no app version available: APP#19 (einsteinbinary_BRP4) PLATFORM#2 (windows_intelx86) min_version 0
2013-06-11 16:01:18.7194 [PID=18031] [version] Don't need CPU jobs, skipping version 104 for hsgamma_FGRP2 ()
2013-06-11 16:01:18.7194 [PID=18031] [version] no app version available: APP#21 (hsgamma_FGRP2) PLATFORM#9 (windows_x86_64) min_version 0
2013-06-11 16:01:18.7194 [PID=18031] [version] no app version available: APP#21 (hsgamma_FGRP2) PLATFORM#2 (windows_intelx86) min_version 0
2013-06-11 16:01:20.3568 [PID=18031] [version] Checking plan class 'SSE2'
2013-06-11 16:01:20.3568 [PID=18031] [version] plan class ok
2013-06-11 16:01:20.3569 [PID=18031] [version] Don't need CPU jobs, skipping version 104 for einstein_S6BucketLVE (SSE2)
2013-06-11 16:01:20.3569 [PID=18031] [version] no app version available: APP#22 (einstein_S6BucketLVE) PLATFORM#9 (windows_x86_64) min_version 0
2013-06-11 16:01:20.3569 [PID=18031] [version] no app version available: APP#22 (einstein_S6BucketLVE) PLATFORM#2 (windows_intelx86) min_version 0
2013-06-11 16:01:20.3579 [PID=18031] [debug] [HOST#4505391] MSG(high) No work sent
2013-06-11 16:01:20.3580 [PID=18031] [debug] [HOST#4505391] MSG(high) see scheduler log messages on http://einstein.phys.uwm.edu//host_sched_logs/4505/4505391
2013-06-11 16:01:20.3580 [PID=18031] Sending reply to [HOST#4505391]: 0 results, delay req 60.00
2013-06-11 16:01:20.3583 [PID=18031] Scheduler ran 1.699 seconds
The problem whatever it was
)
The problem whatever it was seems to be fixed as I have 4 BRP5-tasks in cache now!
Looking better for me, too...
)
Looking better for me, too... :-)
5652 Einstein@Home 6/11/2013 1:23:12 PM Sending scheduler request: To fetch work.
5653 Einstein@Home 6/11/2013 1:23:12 PM Requesting new tasks for NVIDIA
5654 Einstein@Home 6/11/2013 1:23:15 PM Scheduler request completed: got 1 new tasks
fixed here too...
)
fixed here too...