Questions and Answers : Unix/Linux : Boinc not suspending projects properly and mixing them up
Message board moderation

To post messages, you must log in.

AuthorMessage
Hermes

Send message
Joined: 1 Oct 04
Posts: 5
Credit: 119,485
RAC: 0
Message 7473 - Posted: 6 May 2005, 13:47:28 UTC

Boinc v4.32 running under Knoppix 3.4
Attached to Einstein (suspended), LHC and Predictor.
2005-05-06 14:22:02 [---] May run out of work in 0.01 days; requesting more
2005-05-06 14:22:02 [LHC@home] Requesting 16.86 seconds of work
2005-05-06 14:22:02 [LHC@home] Sending request to scheduler: http://lhcathome-sched1.cern.ch/scheduler/cgi
2005-05-06 14:22:33 [LHC@home] Computation for result v64diponlyinjnoskew-58s14_16519.0385_1_sixvf_41990_1 finished
2005-05-06 14:22:33 [ProteinPredictorAtHome] Starting result h0006A_1_13335_1 using mfoldB125 version 4.27
2005-05-06 14:22:34 [LHC@home] Started upload of v64diponlyinjnoskew-58s14_16519.0385_1_sixvf_41990_1_0
2005-05-06 14:22:38 [LHC@home] Scheduler RPC to http://lhcathome-sched1.cern.ch/scheduler/cgi succeeded
2005-05-06 14:22:39 [LHC@home] Started download of v64diponlyinjnoskew-58s16_18565.7693_1_sixvf_42068.zip
2005-05-06 14:22:40 [LHC@home] Finished upload of v64diponlyinjnoskew-58s14_16519.0385_1_sixvf_41990_1_0
2005-05-06 14:22:40 [LHC@home] Throughput 46843 bytes/sec
2005-05-06 14:22:43 [LHC@home] Finished download of v64diponlyinjnoskew-58s16_18565.7693_1_sixvf_42068.zip
2005-05-06 14:22:43 [LHC@home] Throughput 90236 bytes/sec
2005-05-06 14:22:43 [ProteinPredictorAtHome] Pausing result h0006A_1_13335_1 (left in memory)
2005-05-06 14:22:43 [LHC@home] Starting result v64diponlyinjnoskew-58s16_18565.7693_1_sixvf_42068_3 using sixtrack version 4.66
2005-05-06 15:22:43 [ProteinPredictorAtHome] Resuming result h0006A_1_13335_1 using mfoldB125 version 4.27

At this point both the mfold (Predictor) and the sixtrack (LHC) Program are running and increasing their CPU-time. But in the worktab only the CPU-time of the Predictor WU increases, the LHC WU stays at zero CPU-time.

After the next switch:
2005-05-06 15:22:43 [ProteinPredictorAtHome] Resuming result h0006A_1_13335_1 using mfoldB125 version 4.27
2005-05-06 15:22:43 [LHC@home] Pausing result v64diponlyinjnoskew-58s16_18565.7693_1_sixvf_42068_3 (left in memory)

The mfold program is now suspended, and sixtrack continues running. The CPU-time and percentage complete of the mfold-WU continue to rise, the sixtrack-WU stays at 0 for both values.
ID: 7473 · Report as offensive     Reply Quote
Thord

Send message
Joined: 30 Jul 05
Posts: 7
Credit: 18,326,959
RAC: 0
Message 9165 - Posted: 4 Aug 2005, 12:37:20 UTC

I am running boinc 4.43 on linux, and it seems the problem is still
there. If it is a dual cpu runnning multiple projects the manager
display is more often wrong than right.
/Thord.
ID: 9165 · Report as offensive     Reply Quote

Questions and Answers : Unix/Linux : Boinc not suspending projects properly and mixing them up


©2024 CERN