| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: BOINC Agent Support
Thread: Server Software Upgrade issues, Mar-6-2012 ~14:30 UTC - 21:00 UTC |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 177
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Could be related to the lost result reports... see my post, 2 up. The _2 copy went out just after the _1 copy was (re)sent [relatively sure about that]. It's not hard to reconstruct as the _1 was circulated on the 7th, but before _0 was returned i.e. this was an original 2 copy CEP2, not a 1 copy + an additional wingman to check on an inconclusive. Why the system thought the _1 was not enough and sent out another copy 47 minutes later is a ?
--//-- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I am now seeing the latest C4CW sent out today say Min quorum 2, Replication 2......but.....only one has been sent out?
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
What you will be missing out on too is the high credits currently granted... in three days time they'll bound to have gravitated to a lower average. Rob, It's not happening exactly the way you say it.  High credits have lasted only a few hours and the new claims are much lower than what they were before the upgrade.  See below (all results within the same benchmarking period): X0960059581455200510201243_ 0-- JmB-6600-Ub64 Valid 05/03/12 19:38 06/03/12 00:17 0,87 30,0 / 30,0It could differ per science. Last night we ran 914TFL on 395 CPU years or 2.13 TFL per year, where the running current average was 1.59TFL/Year. Still seeing very high credit claim [not what the device asked for] and grant. Here's a few samples of a device that benchmarks inflated 24/hr (HT), gets around 19-20 grant, and is now getting 38-50, some even 60/hr [depends on wingman too of course]. Name DeviceName Status LastUpdate Returned CPUTime Claimed Granted Elapsed PPH ProjectName OSName CPUModel Speed Sent SN2S_ AAC46900_ 0000023_ 0712_ 0-- 1854592 Valid 8-3-2012 7:32:00 7-3-2012 23:17:33 6,36 335,1 260,2 33,34 40,91 SN2S Windows 7 - 64 I7-2670QM 2700 6-3-2012 13:56:56 SN2S_ AAB88508_ 0000021_ 0911_ 1-- 1767290 Valid 8-3-2012 7:32:00 7-3-2012 1:04:07 4,03 250,0 153,2 33,73 38,01 SN2S Ubuntu 11.10 Q6600 2400 5-3-2012 15:20:02 SN2S_ AAC46900_ 0000021_ 0240_ 0-- 1854592 Valid 8-3-2012 0:28:00 7-3-2012 23:27:33 7,63 401,9 385,1 35,04 50,47 SN2S Windows 7 - 64 I7-2670QM 2700 6-3-2012 12:25:19 Enjoy it while it lasts... for sure WCG over at BOINCStats will see a positional boost today when the 91.4 Million credit is recorded.  --//-- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
And being an offline night cruncher, the last upload of 17 but for a few exceptions, show substantial lower claim levels again, particularly for Linux, which descended back to even below old claims. W7 still on ~31.5 where 24 was normal.
Name DeviceName Status LastUpdate Returned CPUTime Claimed Granted Elapsed PPH ProjectName OSName CPUModel Speed Sent SN2S_ AAB88508_ 0000015_ 0767_ 1-- 1854592 Pending Validation 8-3-2012 9:51:00 8-3-2012 8:48:21 5,76 127,9 31,71 SN2S Windows 7 - 64 I7-2670QM 2700 5-3-2012 11:41:47 SN2S_ AAB88508_ 0000015_ 0123_ 1-- 1854592 Pending Validation 8-3-2012 9:51:00 8-3-2012 8:48:21 6,29 139,7 31,71 SN2S Windows 7 - 64 I7-2670QM 2700 5-3-2012 11:41:46 SN2S_ AAB88508_ 0000015_ 0664_ 0-- 1854592 Pending Validation 8-3-2012 9:51:00 8-3-2012 8:48:21 6,04 135,0 31,71 SN2S Windows 7 - 64 I7-2670QM 2700 5-3-2012 11:52:43 SN2S_ AAB88508_ 0000015_ 0601_ 1-- 1854592 Pending Validation 8-3-2012 9:51:00 8-3-2012 8:48:21 5,35 119,3 31,71 SN2S Windows 7 - 64 I7-2670QM 2700 5-3-2012 11:52:43 SN2S_ AAB88508_ 0000026_ 0360_ 0-- 1854592 Pending Validation 8-3-2012 9:51:00 8-3-2012 8:48:21 3,95 87,5 31,71 SN2S Windows 7 - 64 I7-2670QM 2700 5-3-2012 18:28:21 SN2S_ AAC46900_ 0000023_ 0651_ 1-- 1854592 Pending Validation 8-3-2012 9:51:00 8-3-2012 8:48:21 5,95 313,4 42,86 SN2S Windows 7 - 64 I7-2670QM 2700 6-3-2012 13:56:56 SN2S_ AAC46900_ 0000024_ 0270_ 0-- 1854592 Pending Validation 8-3-2012 9:51:00 8-3-2012 8:48:21 6,87 361,8 42,24 SN2S Windows 7 - 64 I7-2670QM 2700 6-3-2012 14:33:57 SN2S_ AAC46900_ 0000024_ 0548_ 1-- 1854592 Pending Validation 8-3-2012 9:51:00 8-3-2012 8:48:21 7,17 377,6 42,24 SN2S Windows 7 - 64 I7-2670QM 2700 6-3-2012 14:33:57 SN2S_ AAC46900_ 0000026_ 0440_ 1-- 1767290 Pending Validation 8-3-2012 9:51:00 8-3-2012 8:48:01 3,78 71,7 31,51 SN2S Ubuntu 11.10 Q6600 2400 7-3-2012 1:17:09 SN2S_ AAC46901_ 0000006_ 1201_ 1-- 1767290 Pending Validation 8-3-2012 9:51:00 8-3-2012 8:48:01 4,04 79,1 23,64 SN2S Ubuntu 11.10 Q6600 2400 7-3-2012 9:09:51 SN2S_ AAC46901_ 0000006_ 0684_ 1-- 1767290 Pending Validation 8-3-2012 9:51:00 8-3-2012 8:48:01 3,94 76,6 23,64 SN2S Ubuntu 11.10 Q6600 2400 7-3-2012 9:09:49 SN2S_ AAC46901_ 0000006_ 0724_ 1-- 1767290 Pending Validation 8-3-2012 9:51:00 8-3-2012 8:48:01 4,49 87,2 23,64 SN2S Ubuntu 11.10 Q6600 2400 7-3-2012 9:09:49 SN2S_ AAC46901_ 0000006_ 0644_ 1-- 1767290 Pending Validation 8-3-2012 9:51:00 8-3-2012 8:48:01 4,02 80,4 23,64 SN2S Ubuntu 11.10 Q6600 2400 7-3-2012 9:09:48 SN2S_ AAC46901_ 0000006_ 0609_ 1-- 1767290 Pending Validation 8-3-2012 9:51:00 8-3-2012 8:48:01 4,21 81,6 23,64 SN2S Ubuntu 11.10 Q6600 2400 7-3-2012 9:09:47 SN2S_ AAB88508_ 0000015_ 0084_ 0-- 1854592 Pending Validation 8-3-2012 9:51:00 8-3-2012 8:47:46 2,88 63,9 31,70 SN2S Windows 7 - 64 I7-2670QM 2700 5-3-2012 11:41:47 SN2S_ AAB88508_ 0000015_ 0770_ 0-- 1854592 Pending Validation 8-3-2012 9:51:00 8-3-2012 8:47:46 3,24 72,1 31,70 SN2S Windows 7 - 64 I7-2670QM 2700 5-3-2012 11:41:47 SN2S_ AAC46900_ 0000026_ 0004_ 1-- 1767290 Pending Validation 8-3-2012 9:51:00 8-3-2012 8:47:42 3,78 71,7 31,52 SN2S Ubuntu 11.10 Q6600 2400 7-3-2012 1:16:23 Short, not 5 days of Kevin, not 3 days of Rob, just 24 hours for the tractor beam to reign things in back to normal for Linux. Maybe we should start looking for Aliens... you'll really never know they have an instant cure for everything, once found ;P --//-- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
For the ''worried'' department, at least for SN2S the validators have not been running. Have many with complete quorum since 00:03 UTC, no "Try Validation". Hope this is with intend... to be there when the validators run, so the credit development can be monitored or some other unrevealed reason.
--//-- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Someone was watching for a member to post on this. Ran WCGDAWS 5 minutes later and most all of the quorum complete passed muster.... but a few, all with old style credit.
Name DeviceName Status LastUpdate Returned CPUTime Claimed Granted Elapsed PPH ProjectName OSName CPUModel Speed Sent SN2S_ AAB88508_ 0000015_ 0123_ 1-- 1854592 Valid 8-3-2012 11:42:00 8-3-2012 8:48:21 6,29 139,7 177,8 31,71 28,27 SN2S Windows 7 - 64 I7-2670QM 2700 5-3-2012 11:41:46 SN2S_ AAB88508_ 0000026_ 0360_ 0-- 1854592 Valid 8-3-2012 11:42:00 8-3-2012 8:48:21 3,95 87,5 78,4 31,71 19,85 SN2S Windows 7 - 64 I7-2670QM 2700 5-3-2012 18:28:21 SN2S_ AAC46900_ 0000023_ 0651_ 1-- 1854592 Valid 8-3-2012 11:42:00 8-3-2012 8:48:21 5,95 313,4 241,1 42,86 40,52 SN2S Windows 7 - 64 I7-2670QM 2700 6-3-2012 13:56:56 SN2S_ AAC46900_ 0000024_ 0270_ 0-- 1854592 Valid 8-3-2012 11:42:00 8-3-2012 8:48:21 6,87 361,8 277,2 42,24 40,35 SN2S Windows 7 - 64 I7-2670QM 2700 6-3-2012 14:33:57 SN2S_ AAC46900_ 0000024_ 0548_ 1-- 1854592 Valid 8-3-2012 11:42:00 8-3-2012 8:48:21 7,17 377,6 350,5 42,24 48,88 SN2S Windows 7 - 64 I7-2670QM 2700 6-3-2012 14:33:57 SN2S_ AAC46901_ 0000006_ 1201_ 1-- 1767290 Valid 8-3-2012 11:42:00 8-3-2012 8:48:01 4,04 79,1 83,9 23,64 20,77 SN2S Ubuntu 11.10 Q6600 2400 7-3-2012 9:09:51 SN2S_ AAC46901_ 0000006_ 0684_ 1-- 1767290 Valid 8-3-2012 11:42:00 8-3-2012 8:48:01 3,94 76,6 71,6 23,64 18,17 SN2S Ubuntu 11.10 Q6600 2400 7-3-2012 9:09:49 SN2S_ AAC46901_ 0000006_ 0724_ 1-- 1767290 Valid 8-3-2012 11:42:00 8-3-2012 8:48:01 4,49 87,2 79,5 23,64 17,71 SN2S Ubuntu 11.10 Q6600 2400 7-3-2012 9:09:49 SN2S_ AAC46901_ 0000006_ 0644_ 1-- 1767290 Valid 8-3-2012 11:42:00 8-3-2012 8:48:01 4,02 80,4 82,2 23,64 20,45 SN2S Ubuntu 11.10 Q6600 2400 7-3-2012 9:09:48 SN2S_ AAC46901_ 0000006_ 0609_ 1-- 1767290 Valid 8-3-2012 11:42:00 8-3-2012 8:48:01 4,21 81,6 79,6 23,64 18,91 SN2S Ubuntu 11.10 Q6600 2400 7-3-2012 9:09:47 SN2S_ AAB88508_ 0000015_ 0084_ 0-- 1854592 Valid 8-3-2012 11:42:00 8-3-2012 8:47:46 2,88 63,9 57,7 31,70 20,03 SN2S Windows 7 - 64 I7-2670QM 2700 5-3-2012 11:41:47 SN2S_ AAB88508_ 0000015_ 0770_ 0-- 1854592 Valid 8-3-2012 11:42:00 8-3-2012 8:47:46 3,24 72,1 78,0 31,70 24,07 SN2S Windows 7 - 64 I7-2670QM 2700 5-3-2012 11:41:47 Me concerned about credit... No, just GFLs. :P --//-- |
||
|
|
kashie
Cruncher Joined: Mar 7, 2007 Post Count: 46 Status: Offline Project Badges: 
|
Still getting the higher, cross project normalised claimed amount on SN2S on my computer.
Returned CPUTime Claimed Name 8/03/2012 10:41:49 AM 6.40 370.4 SN2S_ AAB86562_ 0000020_ 0871_ 0-- 8/03/2012 10:39:06 AM 5.43 314.4 SN2S_ AAB88508_ 0000009_ 0199_ 0-- 8/03/2012 10:30:33 AM 5.35 309.4 SN2S_ AAB88508_ 0000009_ 0115_ 1-- 8/03/2012 10:26:23 AM 4.97 287.5 SN2S_ AAB88508_ 0000009_ 0395_ 1-- 8/03/2012 9:49:05 AM 4.93 285.5 SN2S_ AAB86562_ 0000020_ 1218_ 1-- 8/03/2012 9:44:02 AM 4.53 261.9 SN2S_ AAB88508_ 0000009_ 0174_ 1-- 8/03/2012 3:57:31 AM 4.79 277.1 SN2S_ AAB86562_ 0000020_ 0610_ 0-- 8/03/2012 3:57:13 AM 3.37 195.1 SN2S_ AAB86562_ 0000020_ 0978_ 1-- 8/03/2012 3:56:55 AM 5.27 304.6 SN2S_ AAB86562_ 0000020_ 0796_ 0-- Perhaps your computers may still be in probation stage until the required number of successful tasks has been reported. I think host scaling isn't activated until probation has ended. Not sure how many successful tasks that takes, 10 or 20 perhaps? Your Windows computer shows 3 tasks at the higher rate so possibly this means probation for that computer has ended or just begun. Your next batch of returned results for that computer should tell you that. Why your computers appear to have had host scaling enabled at first and then disabled later is the real question. I didn't think the anti-cherrypicker algorithm was turned on but perhaps it is as I saw someone else who had their claimed credit increase then later decrease since the upgrade. This is based on a very limited understanding of the credit system so could be incorrect and it's something else entirely. It's not just my computer that needs to pass a sanity check with this credit scheme, I now need a sanity check too after reading about it.:) |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Probation [technical] was like 15... before you are allowed to do zero redundancy sciences alone, with the ever returning spotchecks. From the go-live, there have been 28 SN2S returned by this device. As the latest continue to look, the claim has no more 1:1 computational relation to the benchmark. There is device statistics involved how much a claim is, which means that anyone can do whatever with benchmarking [continue to run 6.10.17 on Linux for all we care], the servers do a performance recompute for each device and use the old rules of finding the median. At any rate, looking at the quorum detail, the claims are ever more drawing closer for devices in a quorum, so the mean will eventually for most be rather close. Anticipating that the outlier rules still applied for some sciences will eventually become superfluous... the whole thing self regulating. We're on watch :D
BTW, knreed switched off the reliability queue for the moment [devices where repairs go], else that queue would have bogged down the whole... there are no 'reliable' devices at this time. One needed like 76-77 consecutive error free results before getting to the 0.2% fail rating. Now that we have a science-application level monitoring, that may be different... taking allot longer before a device is all-round reliable. For sure if I switch the SN2S device to CEP2, things start from scratch again... 15 with wingman [or whatever the number is] before going it alone. For now, just rushing my devices to get into the repair approved class on SN2S and waiting for the word from above, before trying another science. --//-- |
||
|
|
kashie
Cruncher Joined: Mar 7, 2007 Post Count: 46 Status: Offline Project Badges:
|
I wasn't referring to probation before being allowed to do zero redundancy tasks alone but host_app_version.scale_probation which relates to when credit scaling is allowed for a host.
----------------------------------------Yes the change to everyone being untrusted for a while caused some concern. One of my team members thought he had lost a lot of his tasks that were in his Pending Validation list and only found them again when he clicked the Inconclusive drop down choice. They had changed how they had been classified after the upgrade. [Edit 1 times, last edit by kashie at Mar 8, 2012 1:08:32 PM] |
||
|
|
keithhenry
Ace Cruncher Senile old farts of the world ....uh.....uh..... nevermind Joined: Nov 18, 2004 Post Count: 18665 Status: Offline Project Badges: 
|
As an FYI, it appears that resends are going out with the standard 10 day deadline instead of the usual 4 day from prior to the upgrade.
---------------------------------------- |
||
|
|
|