Boinc v. 5.2.x

Message boards : BOINC client : Boinc v. 5.2.x
Message board moderation

To post messages, you must log in.

AuthorMessage
Metod, S56RKO

Send message
Joined: 9 Sep 05
Posts: 128
Slovenia
Message 689 - Posted: 11 Oct 2005, 13:45:00 UTC

Hello,

I upgraded my crunchers to boinc CC version 5.2.1.

I noticed that it doesn't do the file transfers for seti@home immediately. This is true both for DL and UL.

For example: one of my hosts got assigned 5 WUs by seti@home scheduller. Normally (vith older CC versions at least) it would start to download the associated files immediately, but it doesn't. It's been more than 2 hours and according to CC messages it didn't try to start transfers. If I force DL my hand (retry the transfers) it will DL files just fine.

Another example: my crucnher finished a seti@home WU. Normally it would upload the result immediately and postpone the reporting to later stage. Now it doesn't start to upload the result.

Is this intended behaviour or is it a bug?

I'll just let it be and observe, how and when they will do the transfer.
Metod ...
ID: 689 · Report as offensive
Keck_Komputers
Avatar

Send message
Joined: 29 Aug 05
Posts: 304
United States
Message 691 - Posted: 11 Oct 2005, 22:14:31 UTC

Check under the commands menu to see if you have a bad setting for networking. Also check the proxy settings in BOINC to make sure they are appropriate.
BOINC WIKI

BOINCing since 2002/12/8
ID: 691 · Report as offensive
Metod, S56RKO

Send message
Joined: 9 Sep 05
Posts: 128
Slovenia
Message 692 - Posted: 12 Oct 2005, 12:07:33 UTC - in response to Message 691.  
Last modified: 12 Oct 2005, 12:13:26 UTC

Check under the commands menu to see if you have a bad setting for networking. Also check the proxy settings in BOINC to make sure they are appropriate.


My settings for the hosts in question are the same as they used to be: connection at any time, no proxy.

The problem is that the transfer is not even initiated by CC. See a complete stdout from one of the hosts. The transfer at the end of it was hand-forced and went OK without a hic-cup.

As you can see, there's no problem with Einstein: 2005-10-12 11:31 the computation finished and the result was uploaded immediately. With seti: new WU was allocated on 2005-10-11 17:51, but only downloaded after forced transfer on 2005-10-12 13:57. A result computation finished on 2005-10-12 00:32 but did not upload until forced transfer.

2005-10-11 11:09:05 [---] Starting BOINC client version 5.2.1 for windows_intelx86
2005-10-11 11:09:05 [---] libcurl/7.14.0 OpenSSL/0.9.8 zlib/1.2.3
2005-10-11 11:09:05 [---] Executing as a daemon
2005-10-11 11:09:05 [---] Data directory: D:\\boinc
2005-10-11 11:09:05 [---] BOINC is running as a service and as a non-system user.
2005-10-11 11:09:05 [---] No application graphics will be available.
2005-10-11 11:09:06 [SETI@home] Found app_info.xml; using anonymous platform
2005-10-11 11:09:06 [---] Processor: 2 GenuineIntel x86 Family 6 Model 7 Stepping 3 598MHz
2005-10-11 11:09:06 [---] Memory: 1023.54 MB physical, 2.41 GB virtual
2005-10-11 11:09:06 [---] Disk: 50.87 GB total, 37.46 GB free
2005-10-11 11:09:06 [---] Version change detected (4.71.0 -> 5.2.1); running CPU benchmarks
2005-10-11 11:09:06 [climateprediction.net] Computer ID: 145118; location: work; project prefs: default
2005-10-11 11:09:06 [Einstein@Home] Computer ID: 17148; location: work; project prefs: default
2005-10-11 11:09:06 [SETI@home] Computer ID: 406496; location: work; project prefs: default
2005-10-11 11:09:06 [---] General prefs: from SETI@home (last modified 2005-10-01 16:22:59)
2005-10-11 11:09:06 [---] General prefs: no separate prefs for work; using your defaults
2005-10-11 11:09:06 [---] Remote control allowed
2005-10-11 11:09:08 [---] Running CPU benchmarks
2005-10-11 11:10:07 [---] Benchmark results:
2005-10-11 11:10:07 [---]    Number of CPUs: 2
2005-10-11 11:10:07 [---]    531 double precision MIPS (Whetstone) per CPU
2005-10-11 11:10:07 [---]    938 integer MIPS (Dhrystone) per CPU
2005-10-11 11:10:07 [---] Finished CPU benchmarks
2005-10-11 11:10:08 [climateprediction.net] Deferring computation for result 2yoo_100159659_1
2005-10-11 11:10:08 [Einstein@Home] Resuming computation for result l1_1258.0__1258.3_0.1_T00_S4lC_2 using einstein version 479
2005-10-11 11:10:08 [SETI@home] Resuming computation for result 15mr04aa.4144.31664.534648.30_3 using setiathome version 411
2005-10-11 11:10:08 [---] Resuming computation and network activity
2005-10-11 11:10:08 [---] request_reschedule_cpus: Resuming activities
2005-10-11 11:10:08 [climateprediction.net] Restarting result 2yoo_100159659_1 using hadsm3 version 413
2005-10-11 11:10:08 [SETI@home] Pausing result 15mr04aa.4144.31664.534648.30_3 (removed from memory)
2005-10-11 11:10:10 [---] request_reschedule_cpus: process exited
2005-10-11 13:10:10 [climateprediction.net] Pausing result 2yoo_100159659_1 (removed from memory)
2005-10-11 13:10:10 [SETI@home] Restarting result 15mr04aa.4144.31664.534648.30_3 using setiathome version 411
2005-10-11 13:10:32 [---] request_reschedule_cpus: process exited
2005-10-11 14:05:53 [Einstein@Home] Sending scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
2005-10-11 14:05:53 [Einstein@Home] Reason: To report results
2005-10-11 14:05:53 [Einstein@Home] Reporting 1 results
2005-10-11 14:05:58 [Einstein@Home] Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded
2005-10-11 14:06:03 [Einstein@Home] Deferring communication with project for 55 seconds
2005-10-11 15:10:33 [climateprediction.net] Restarting result 2yoo_100159659_1 using hadsm3 version 413
2005-10-11 15:10:33 [SETI@home] Pausing result 15mr04aa.4144.31664.534648.30_3 (removed from memory)
2005-10-11 15:10:34 [---] request_reschedule_cpus: process exited
2005-10-11 17:10:34 [Einstein@Home] Pausing result l1_1258.0__1258.3_0.1_T00_S4lC_2 (removed from memory)
2005-10-11 17:10:34 [SETI@home] Restarting result 15mr04aa.4144.31664.534648.30_3 using setiathome version 411
2005-10-11 17:10:36 [---] request_reschedule_cpus: process exited
2005-10-11 17:51:11 [SETI@home] Sending scheduler request to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi
2005-10-11 17:51:11 [SETI@home] Reason: To fetch work
2005-10-11 17:51:11 [SETI@home] Requesting 30 seconds of new work
2005-10-11 17:51:16 [SETI@home] Scheduler request to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi succeeded
2005-10-11 17:51:21 [SETI@home] Deferring communication with project for 10 minutes and 0 seconds
2005-10-11 19:10:37 [Einstein@Home] Restarting result l1_1258.0__1258.3_0.1_T00_S4lC_2 using einstein version 479
2005-10-11 19:10:37 [SETI@home] Pausing result 15mr04aa.4144.31664.534648.30_3 (removed from memory)
2005-10-11 19:10:38 [---] request_reschedule_cpus: process exited
2005-10-11 23:10:39 [climateprediction.net] Pausing result 2yoo_100159659_1 (removed from memory)
2005-10-11 23:10:39 [SETI@home] Restarting result 15mr04aa.4144.31664.534648.30_3 using setiathome version 411
2005-10-11 23:10:56 [---] request_reschedule_cpus: process exited
2005-10-12 00:32:53 [---] request_reschedule_cpus: process exited
2005-10-12 00:32:53 [SETI@home] Computation for result 15mr04aa.4144.31664.534648.30_3 finished
2005-10-12 00:32:53 [climateprediction.net] Restarting result 2yoo_100159659_1 using hadsm3 version 413
2005-10-12 02:28:01 [climateprediction.net] Sending scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
2005-10-12 02:28:01 [climateprediction.net] Reason: To send trickle-up message
2005-10-12 02:28:01 [climateprediction.net] Note: not requesting new work or reporting results
2005-10-12 02:28:06 [climateprediction.net] Scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi succeeded
2005-10-12 04:42:34 [Einstein@Home] Sending scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
2005-10-12 04:42:34 [Einstein@Home] Reason: To fetch work
2005-10-12 04:42:34 [Einstein@Home] Requesting 15 seconds of new work
2005-10-12 04:42:39 [Einstein@Home] Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded
2005-10-12 04:42:40 [---] request_reschedule_cpus: files downloaded
2005-10-12 04:42:44 [Einstein@Home] Deferring communication with project for 55 seconds
2005-10-12 10:11:31 [---] request_reschedule_cpus: process exited
2005-10-12 10:11:31 [Einstein@Home] Computation for result l1_1258.0__1258.3_0.1_T00_S4lC_2 finished
2005-10-12 10:11:31 [Einstein@Home] Starting result l1_1258.0__1258.1_0.1_T02_S4lC_3 using einstein version 479
2005-10-12 10:11:33 [Einstein@Home] Started upload of l1_1258.0__1258.3_0.1_T00_S4lC_2_0
2005-10-12 10:11:42 [Einstein@Home] Finished upload of l1_1258.0__1258.3_0.1_T00_S4lC_2_0
2005-10-12 10:11:42 [Einstein@Home] Throughput 19446 bytes/sec
2005-10-12 13:57:40 [SETI@home] Started download of 20oc03ab.15181.3361.261086.125
2005-10-12 13:57:40 [SETI@home] Started upload of 15mr04aa.4144.31664.534648.30_3_0
2005-10-12 13:57:47 [SETI@home] Finished upload of 15mr04aa.4144.31664.534648.30_3_0
2005-10-12 13:57:47 [SETI@home] Throughput 13113 bytes/sec
2005-10-12 13:57:55 [SETI@home] Finished download of 20oc03ab.15181.3361.261086.125
2005-10-12 13:57:55 [SETI@home] Throughput 26479 bytes/sec
2005-10-12 13:57:56 [---] request_reschedule_cpus: files downloaded
2005-10-12 13:57:56 [Einstein@Home] Pausing result l1_1258.0__1258.1_0.1_T02_S4lC_3 (removed from memory)
2005-10-12 13:57:56 [SETI@home] Starting result 20oc03ab.15181.3361.261086.125_1 using setiathome version 411
2005-10-12 13:57:57 [---] request_reschedule_cpus: process exited

Metod ...
ID: 692 · Report as offensive
MJKelleher

Send message
Joined: 29 Aug 05
Posts: 8
United States
Message 693 - Posted: 13 Oct 2005, 0:32:08 UTC

I'm seeing the dealy, wait a bit and it'll report.

10/12/2005 9:52:41 AM||request_reschedule_cpus: process exited
10/12/2005 9:52:41 AM|Predictor @ Home|Computation for result h0009B_1_38639_0 finished
10/12/2005 9:52:42 AM|SETI@home|Restarting result 09oc03ab.7118.25298.767346.210_2 using setiathome version 418
10/12/2005 9:52:44 AM|Predictor @ Home|Started upload of h0009B_1_38639_0_0
10/12/2005 9:52:44 AM|Predictor @ Home|Started upload of h0009B_1_38639_0_1
10/12/2005 9:52:52 AM|Predictor @ Home|Finished upload of h0009B_1_38639_0_0
10/12/2005 9:52:52 AM|Predictor @ Home|Throughput 1341 bytes/sec
10/12/2005 9:52:52 AM|Predictor @ Home|Finished upload of h0009B_1_38639_0_1
10/12/2005 9:52:52 AM|Predictor @ Home|Throughput 8123 bytes/sec
10/12/2005 9:52:52 AM|Predictor @ Home|Started upload of h0009B_1_38639_0_2
10/12/2005 9:52:57 AM|Predictor @ Home|Finished upload of h0009B_1_38639_0_2
10/12/2005 9:52:57 AM|Predictor @ Home|Throughput 10601 bytes/sec
10/12/2005 10:52:42 AM|SETI@home|Pausing result 09oc03ab.7118.25298.767346.210_2 (removed from memory)
10/12/2005 10:52:42 AM|Predictor @ Home|Starting result h0009B_1_78831_0 using mfoldB125 version 428
10/12/2005 10:52:44 AM||request_reschedule_cpus: process exited
10/12/2005 11:52:45 AM|SETI@home|Restarting result 09oc03ab.7118.25298.767346.210_2 using setiathome version 418
10/12/2005 11:52:45 AM|Predictor @ Home|Pausing result h0009B_1_78831_0 (removed from memory)
10/12/2005 11:52:46 AM||request_reschedule_cpus: process exited
10/12/2005 1:52:48 PM|SETI@home|Pausing result 09oc03ab.7118.25298.767346.210_2 (removed from memory)
10/12/2005 1:52:48 PM|Predictor @ Home|Restarting result h0009B_1_78831_0 using mfoldB125 version 428
10/12/2005 1:52:49 PM||request_reschedule_cpus: process exited
10/12/2005 2:41:05 PM|Predictor @ Home|Sending scheduler request to http://predictor.scripps.edu/predictor_cgi/cgi
10/12/2005 2:41:05 PM|Predictor @ Home|Reason: To report results
10/12/2005 2:41:05 PM|Predictor @ Home|Reporting 1 results
10/12/2005 2:41:10 PM|Predictor @ Home|Scheduler request to http://predictor.scripps.edu/predictor_cgi/cgi succeeded
10/12/2005 2:41:16 PM|Predictor @ Home|Deferring communication with project for 0 seconds


If I recall correctly, it's as designed. Upload when complete, but wait for something else connecting, or the "Connect to network about every time" after the work uploads. That delay is about right for this computer.

MJ

ID: 693 · Report as offensive
Metod, S56RKO

Send message
Joined: 9 Sep 05
Posts: 128
Slovenia
Message 696 - Posted: 13 Oct 2005, 16:14:50 UTC - in response to Message 693.  
Last modified: 13 Oct 2005, 16:19:54 UTC

I'm seeing the dealy, wait a bit and it'll report.


I'm not complaining about delays before reporting. I'm complaining about delays for uploading and downloading the result files.

A short summary of delayed DL:


2005-10-11 17:51:11 [SETI@home] Sending scheduler request to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi
2005-10-11 17:51:11 [SETI@home] Reason: To fetch work
2005-10-11 17:51:11 [SETI@home] Requesting 30 seconds of new work
2005-10-11 17:51:16 [SETI@home] Scheduler request to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi succeeded
...
2005-10-12 13:57:40 [SETI@home] Started download of 20oc03ab.15181.3361.261086.125
2005-10-12 13:57:55 [SETI@home] Finished download of 20oc03ab.15181.3361.261086.125
2005-10-12 13:57:55 [SETI@home] Throughput 26479 bytes/sec


And this DL only happened because I forced it. Do you see it now?

And here's delayed UL summary:


2005-10-12 00:32:53 [---] request_reschedule_cpus: process exited
2005-10-12 00:32:53 [SETI@home] Computation for result 15mr04aa.4144.31664.534648.30_3 finished
...
2005-10-12 13:57:40 [SETI@home] Started upload of 15mr04aa.4144.31664.534648.30_3_0
2005-10-12 13:57:47 [SETI@home] Finished upload of 15mr04aa.4144.31664.534648.30_3_0
2005-10-12 13:57:47 [SETI@home] Throughput 13113 bytes/sec


Again, this UL only happened because I forced it.

At this stage, uploaded WU was in state 'ready to report' and those will (even with CC 5.2.1) get reported when scheduller gets contacted next time.

And, again, I don't see this (erroneous IMHO) behaviour with Einstein@home and I didn't see it with earlier versions of CC (tested: 5.1.8, 5.1.6, 4.72, 4.71, 4.45, etc).
Metod ...
ID: 696 · Report as offensive
Metod, S56RKO

Send message
Joined: 9 Sep 05
Posts: 128
Slovenia
Message 698 - Posted: 14 Oct 2005, 10:13:10 UTC - in response to Message 696.  

I'm seeing the dealy, wait a bit and it'll report.


I'm not complaining about delays before reporting. I'm complaining about delays for uploading and downloading the result files.


Another thought occured to me this morning (European time): I had all the hosts connection settings to based on preferences. I changed them to allways and after a couple of hours it seems that the problem dissapeared. If this is it, then there might be some bug interpreting the settings. From far it seems to be something similar to the bug in 4.7x, where CC would try to contact climate@home scheduller for trickle uploads regardless of network connectivity settings (even if it was set to off explicitly).

I'll let it be like this and will report in another couple of hours.
Metod ...
ID: 698 · Report as offensive
Metod, S56RKO

Send message
Joined: 9 Sep 05
Posts: 128
Slovenia
Message 699 - Posted: 14 Oct 2005, 16:57:44 UTC - in response to Message 698.  


Another thought occured to me this morning (European time): I had all the hosts connection settings to based on preferences. I changed them to allways and after a couple of hours it seems that the problem dissapeared. If this is it, then there might be some bug interpreting the settings. From far it seems to be something similar to the bug in 4.7x, where CC would try to contact climate@home scheduller for trickle uploads regardless of network connectivity settings (even if it was set to off explicitly).

I'll let it be like this and will report in another couple of hours.


After another half a dozen hours I'm further convinced, that my theory above is true. I didn't notice any delayed uploads and/or downloads since change of network connectivity to allways.
I could only see normal delayed reporting of already uploaded results.
Metod ...
ID: 699 · Report as offensive
Profile Jord
Volunteer tester
Help desk expert
Avatar

Send message
Joined: 29 Aug 05
Posts: 15492
Netherlands
Message 700 - Posted: 14 Oct 2005, 19:36:33 UTC

My 5.2.1 is always on always (forgive the pun ;)), yet I had upload problems with Primegrid. I still have them on another computer, although we're testing if it's to do with the standard time-out period of PG.
ID: 700 · Report as offensive

Message boards : BOINC client : Boinc v. 5.2.x

Copyright © 2024 University of California.
Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.