| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: GO Fight Against Malaria
Thread: Being flooded with GFAM work units!!! |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 23
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
The baghwan, Sri knreedus, is on vacation. Really don't know if this is a bug or by design to overcome queue clogging. As written previously, only got officially that the User Aborted don't go to the [prioritized] repair [Reliable hosts] feeder anymore. All crunchers get them and considering that it's not really a repair, but a substitute, it makes perfect sense to me. Quorum 2 is quorum 2 and does not necessarily require the limited available 'Repair' hosts to be waited upon.
----------------------------------------edit: The implication is, that if members abort rare work 4, or work in excess of their crunching goals [HCC is getting Ni, and DDDT2 was famous for this members 'do double good 'service], it's everyone that can put the buckets out for these, not the high reliability devices only. [Edit 1 times, last edit by Former Member at Mar 30, 2013 6:32:16 PM] |
||
|
|
Johnny Cool
Ace Cruncher USA Joined: Jul 28, 2005 Post Count: 8621 Status: Offline Project Badges: 
|
The baghwan, Sri knreedus, is on vacation. Really don't know if this is a bug or by design to overcome queue clogging. As written previously, only got officially that the User Aborted don't go to the [prioritized] repair [Reliable hosts] feeder anymore. All crunchers get them and considering that it's not really a repair, but a substitute, it makes perfect sense to me. Quorum 2 is quorum 2 and does not necessarily require the limited available 'Repair' hosts to be waited upon. edit: The implication is, that if members abort rare work 4, or work in excess of their crunching goals [HCC is getting Ni, and DDDT2 was famous for this members 'do double good 'service], it's everyone that can put the buckets out for these, not the high reliability devices only. So, anyone with problems with the GFAM Flood will have to wait? No one else working at WCG can fix this? Just saying; and I do have much respect for knreed. I just let a trickle of GFAM work units last evening by going to mygrid. Added GFAM T= to my HCC gpu work units and lowered cache to 1.5 days. This morning, I went back to mygrid and *un-checked* the GFAM project until this is resolved.  Using Boinc 6.10.58, and despite rather confusing statements here at the WCG concerning where to put the app_info either H:\ProgramData\BOINC OR H:\ProgramData\BOINC\projects\www.worldcommunitygrid.org Yes, I use another partition on the second internal hard drive. The version 6.10.58 is 64-bits and that means I did DL it from http://boinc.berkeley.edu/ some time ago, That version did NOT have any cc_config or app_info. It may be possible that they added a cc_config months later after I DL'd it. Also, when I first used an app_info in the H:\ProgramData\BOINC\projects\www.worldcommunitygrid.org folder (because of ALL the statements here, it did not do anything promising AT ALL! I forge how long it was stuck in two HCC GPU threads for I don't know how long. I took a shot and put it in the aforementioned other folder. I placed mine in the H:\ProgramData\BOINC folder and ... voila! Statistics Date Total Run Time (y:d:h:m:s) Points Generated Results Returned 03/24/2013 - 0:008:23:07:40 - 179,838 - 400 03/23/2013 - 0:008:15:06:15 - 190,613 - 421 03/22/2013 - 0:006:07:57:54 - 186,834 - 412 03/21/2013 - 0:005:07:04:13 - 164,850 - 364 Have a great day all. You too, knreed!  |
||
|
|
Johnny Cool
Ace Cruncher USA Joined: Jul 28, 2005 Post Count: 8621 Status: Offline Project Badges:
|
The baghwan, Sri knreedus, is on vacation. Really don't know if this is a bug or by design to overcome queue clogging. As written previously, only got officially that the User Aborted don't go to the [prioritized] repair [Reliable hosts] feeder anymore. All crunchers get them and considering that it's not really a repair, but a substitute, it makes perfect sense to me. Quorum 2 is quorum 2 and does not necessarily require the limited available 'Repair' hosts to be waited upon. edit: The implication is, that if members abort rare work 4, or work in excess of their crunching goals [HCC is getting Ni, and DDDT2 was famous for this members 'do double good 'service], it's everyone that can put the buckets out for these, not the high reliability devices only. knreed is back. I hope that this problem can be resolved. Also, still only running 8 threads; HCC GPU work-units.  PS, your edit makes no sense to me at all. I had to abort a ton of GFAM work-units as thE "No Reply" messages were consuming many pages in the Results section at mygrid. I hope that you are not pushing the blame on me for for what looks like a WCG (or a Boinc) glitch. |
||
|
|
PecosRiverM
Veteran Cruncher The Great State of Texas Joined: Apr 27, 2007 Post Count: 1054 Status: Offline Project Badges: 
|
Using Boinc 6.10.58, and despite rather confusing statements here at the WCG concerning where to put the app_info either H:\ProgramData\BOINC OR H:\ProgramData\BOINC\projects\www.worldcommunitygrid.org Also, when I first used an app_info in the H:\ProgramData\BOINC\projects\www.worldcommunitygrid.org folder (because of ALL the statements here, it did not do anything promising AT ALL! I forge how long it was stuck in two HCC GPU threads for I don't know how long. I took a shot and put it in the aforementioned other folder. I placed mine in the H:\ProgramData\BOINC folder and ... voila! Johnny I'm not real sure what your trying to run. You can go here and do some reading (these guys know their stuff). http://www.xtremesystems.org/forums/showthrea...09-Working-app_info-files    |
||
|
|
Johnny Cool
Ace Cruncher USA Joined: Jul 28, 2005 Post Count: 8621 Status: Offline Project Badges:
|
Using Boinc 6.10.58, and despite rather confusing statements here at the WCG concerning where to put the app_info either H:\ProgramData\BOINC OR H:\ProgramData\BOINC\projects\www.worldcommunitygrid.org Also, when I first used an app_info in the H:\ProgramData\BOINC\projects\www.worldcommunitygrid.org folder (because of ALL the statements here, it did not do anything promising AT ALL! I forge how long it was stuck in two HCC GPU threads for I don't know how long. I took a shot and put it in the aforementioned other folder. I placed mine in the H:\ProgramData\BOINC folder and ... voila! Johnny I'm not real sure what your trying to run. You can go here and do some reading (these guys know their stuff). http://www.xtremesystems.org/forums/showthrea...09-Working-app_info-files I remember seeing thread at Xtremesystems back in December. Like many peeps, they go there for this kind of info (from OldChap in this case). Been reading certain threads in the WCG sections there as well as at Boinc. <app_info> <app> <name> hcc1 </name> </app> <file_info> <name> wcg_hcc1_img_7.05_windows_intelx86__ati_hcc1 </name> <executable/> </file_info> <file_info> <name> hcckernel.cl.7.05 </name> <executable/> </file_info> <app_version> <app_name> hcc1 </app_name> <version_num> 705 </version_num> <platform> windows_intelx86 </platform> <plan_class> ati_hcc1 </plan_class> <max_ncpus> 1.0 </max_ncpus> <coproc> <type> ATI </type> <count> 0.5 </count> </coproc> <file_ref> <file_name> wcg_hcc1_img_7.05_windows_intelx86__ati_hcc1 </file_name> <main_program/> </file_ref> <file_ref> <file_name> hcckernel.cl.7.05 </file_name> <open_name> hcckernel.cl </open_name> </file_ref> </app_version> </app_info> Look familiar? |
||
|
|
PecosRiverM
Veteran Cruncher The Great State of Texas Joined: Apr 27, 2007 Post Count: 1054 Status: Offline Project Badges:
|
I don't see the part for GFAM..
----------------------------------------Here's the one HCC only that worked for me: (some small changes from yours) <app_info> <app> <name>hcc1</name> <user_friendly_name>Help Conquer Cancer</user_friendly_name> </app> <file_info> <name>wcg_hcc1_img_7.05_windows_intelx86__ati_hcc1</name> <executable/> </file_info> <file_info> <name>hcckernel.cl.7.05</name> <executable/> </file_info> <app_version> <app_name>hcc1</app_name> <version_num>705</version_num> <platform>windows_intelx86</platform> <plan_class>ati_hcc1</plan_class> <avg_ncpus>1.0</avg_ncpus> <max_ncpus>1.0</max_ncpus> <coproc> <type>ATI</type> <count>.5</count> </coproc> <file_ref> <file_name>wcg_hcc1_img_7.05_windows_intelx86__ati_hcc1</file_name> <main_program/> </file_ref> <file_ref> <file_name>hcckernel.cl.7.05</file_name> <open_name>hcckernel.cl</open_name> </file_ref> </app_version> </app_info> |
||
|
|
Johnny Cool
Ace Cruncher USA Joined: Jul 28, 2005 Post Count: 8621 Status: Offline Project Badges:
|
I don't see the part for GFAM.. Here's the one HCC only that worked for me: (some small changes from yours) <app_info> <app> <name>hcc1</name> <user_friendly_name>Help Conquer Cancer</user_friendly_name> </app> <file_info> <name>wcg_hcc1_img_7.05_windows_intelx86__ati_hcc1</name> <executable/> </file_info> <file_info> <name>hcckernel.cl.7.05</name> <executable/> </file_info> <app_version> <app_name>hcc1</app_name> <version_num>705</version_num> <platform>windows_intelx86</platform> <plan_class>ati_hcc1</plan_class> <avg_ncpus>1.0</avg_ncpus> <max_ncpus>1.0</max_ncpus> <coproc> <type>ATI</type> <count>.5</count> </coproc> <file_ref> <file_name>wcg_hcc1_img_7.05_windows_intelx86__ati_hcc1</file_name> <main_program/> </file_ref> <file_ref> <file_name>hcckernel.cl.7.05</file_name> <open_name>hcckernel.cl</open_name> </file_ref> </app_version> </app_info> PecosRiverM, that looks good to me. I just copied it and will probably try it out as soon as I get a chance. Thanks for helping me! |
||
|
|
PecosRiverM
Veteran Cruncher The Great State of Texas Joined: Apr 27, 2007 Post Count: 1054 Status: Offline Project Badges:
|
PecosRiverM, that looks good to me. I just copied it and will probably try it out as soon as I get a chance. Thanks for helping me! Anytime (with my limited knowledge) I can. Don't forget to save as a .xml and not .txt file .I tried that by mistake (doesn't work).  |
||
|
|
Johnny Cool
Ace Cruncher USA Joined: Jul 28, 2005 Post Count: 8621 Status: Offline Project Badges:
|
PecosRiverM, that looks good to me. I just copied it and will probably try it out as soon as I get a chance. Thanks for helping me! Anytime (with my limited knowledge) I can. Don't forget to save as a .xml and not .txt file .I tried that by mistake (doesn't work). (Churchill Downs) The Run For The Roses "And they are off" And yes, I saved it as text like you posted.  Nah! j/K |
||
|
|
Johnny Cool
Ace Cruncher USA Joined: Jul 28, 2005 Post Count: 8621 Status: Offline Project Badges:
|
I was hoping that knreed would post here about the GFAM Flood issue (if in fact, it's a non-issue) using Boinc version 6.10.58 and the reason why I have been hampered to only running 8 threads. This last part is the most important. basically, a crunching volunteer can be "penalized' for aborting GFAM's that were never going to be on time! Yes, the cloudy term "No Reply" and other issues.
----------------------------------------BTW, PecosRiverM, the performance of that app_info is really the same as the one I had despite a few word changes. Again, thanks very much! We are fortunate to have crunchers here who are very helpful.  I also sent email to WCG the same day I noticed this "GFAM Flood" issue and posted it here. No Reply from WCG or even a WCG Tech here ... |
||
|
|
|