Message boards :
Number crunching :
Panic Mode On (55) Server problems?
Message board moderation
Previous · 1 · 2 · 3 · 4 · 5 · 6 · 7 . . . 10 · Next
Author | Message |
---|---|
soft^spirit Send message Joined: 18 May 99 Posts: 6497 Credit: 34,134,168 RAC: 0 |
The limit I was referring to is total pending cpu/gpu. Janice |
zoom3+1=4 Send message Joined: 30 Nov 03 Posts: 65755 Credit: 55,293,173 RAC: 49 |
The limit I was referring to is total pending cpu/gpu. Well I was referring to whats on My application details page, which has details for both cpu and gpu, of course I quit doing cpu work months ago, but does not refer to pendings, wherever that is at. The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's |
Fred E. Send message Joined: 22 Jul 99 Posts: 768 Credit: 24,140,697 RAC: 0 |
but does not refer to pendings, wherever that is at. That's results returned pending validation. Don't know about the 450 limit. I was at about 460 and just got 2 batches totalling 54. They will be on E-Bay soon. :=) Another Fred Support SETI@home when you search the Web with GoodSearch or shop online with GoodShop. |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
The limit I was referring to is total pending cpu/gpu. The limit displayed there is nonsense. When I fired up my 24 core machine at work it showed a limit of like 200-300 after a few days with a 1 day cache, but then proceeded to download 1500 tasks in a few hours when I bumped the cache limit up. With it showing 'Max tasks per day: 250 Number of tasks today: '1500 It could have been that limit was just broken that day with all the mucking about they have been doing. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
Fred E. Send message Joined: 22 Jul 99 Posts: 768 Credit: 24,140,697 RAC: 0 |
Also don't think it's the daily limits on the account details page, which I guess are to guard against a host run amok. Should have mentioned below that the tasks I got were GPU on a gpu only request - CPU request was in 40 minute standoff. I still get the limit message when I request CPU tasks, which is where I'm running short. Another Fred Support SETI@home when you search the Web with GoodSearch or shop online with GoodShop. |
zoom3+1=4 Send message Joined: 30 Nov 03 Posts: 65755 Credit: 55,293,173 RAC: 49 |
The limit I was referring to is total pending cpu/gpu. Well something needs fixing, that's for sure, just not here on the PC, that works perfectly. The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's |
soft^spirit Send message Joined: 18 May 99 Posts: 6497 Credit: 34,134,168 RAC: 0 |
The limit I was referring to is total pending cpu/gpu. http://setiathome.berkeley.edu/results.php?hostid=3744107 Your in progress count. Mine seems maxed at 450 for several hours, never higher. I am guessing others are getting more because of 6 physical cpu cores(the HT and the 2nd CPU on mine are most likely ignored in their figuring) Turned in 9 tasks, got 9 tasks. Turned in 12, got 12. I think it is safe to say I am done loading up until they increase the server side limits. Janice |
zoom3+1=4 Send message Joined: 30 Nov 03 Posts: 65755 Credit: 55,293,173 RAC: 49 |
The limit I was referring to is total pending cpu/gpu. B4 I did the Flops setting in the app info file I was repeatedly getting only 4 wu's for the gpus, afterwards the amount went up to as high as 41, minimum I saw b4 putting boinc back to sleep was about 31 or so. At 11pm tonight Boinc will awake and crunch until 7am in the morning here so as to avoid overloading Me or the cooler. The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's |
W-K 666 Send message Joined: 18 May 99 Posts: 19070 Credit: 40,757,560 RAC: 67 |
Limits. At the moment, due to the obvious problems at the moment, and the problems that would occur when the present state of affairs is fixed. There is a temporary limit, and it is based on CPU tasks. Meet that limit and the computer cannot d/load any more tasks, GPU's included, until you return some more. One of the reasons Vic why you should consider AP. (I received 4 AP's yesterday which at the time filled my CPU requirement and allowed the remainder to go to the GPU.) The foreseen problem, is as most DCF's are now v.low, so that when the problem is solved the DCF will modify the completion estimates and computers will be requesting several 100 times more than required. = server overload^2 HAL9000, the normal limits I believe are per processor, so for a 24 cpu computer with 250 limit the max/day would be 6000. |
soft^spirit Send message Joined: 18 May 99 Posts: 6497 Credit: 34,134,168 RAC: 0 |
I currently have no "problem" with limits, I have gpu and cpu work. In the past problems have occured when the total limit is lower than the seperate cpu and gpu limits. This will cause people to run dry a LOT. I can wait until everyone is busy working again before backing up the truck. Janice |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
The limit I was referring to is total pending cpu/gpu. It looks like they might have set the limit for CPU only machines to 50. I saw that limit message on one of my home boxen. So I did a project reset. It was resent 20, 20, and then 10 tasks from the 57 it had previously. Then the next request was again the limit message. The estimated time is only about 3x the actual completion time 2.5 hour tasks taking 30 min, 7 hr tasks taking about 2 hours. Two of my other home machines have exactly 50 tasks in progress and are also getting the limit message. Some work > no work! SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
zoom3+1=4 Send message Joined: 30 Nov 03 Posts: 65755 Credit: 55,293,173 RAC: 49 |
I'll pass on AP, As I don't do cpu work anymore, Just gpu work as AP takes too long on the cpu and that is one of My reasons. The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's |
zoom3+1=4 Send message Joined: 30 Nov 03 Posts: 65755 Credit: 55,293,173 RAC: 49 |
Yeah in between the 4 wu's I was dry quite a bit of the time... The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's |
tbret Send message Joined: 28 May 99 Posts: 3380 Credit: 296,162,071 RAC: 40 |
It was pure coincidence. |
Cosmic_Ocean Send message Joined: 23 Dec 00 Posts: 3027 Credit: 13,516,867 RAC: 13 |
Ooo.. I just noticed a new message that I haven't seen before. 2011-09-23 01:53:18|SETI@home|Message from server: Not sending work - last request too recent: 77 sec I didn't bother looking at the messages tab to see how recent the last work request was and hit the update button. Guess it was only 77 seconds prior. Linux laptop: record uptime: 1511d 20h 19m (ended due to the power brick giving-up) |
S@NL Etienne Dokkum Send message Joined: 11 Jun 99 Posts: 212 Credit: 43,822,095 RAC: 0 |
this message has been in BOINC for a long time. Fact beeing that you cannot request work in under 5 minutes (300secs)... |
Cosmic_Ocean Send message Joined: 23 Dec 00 Posts: 3027 Credit: 13,516,867 RAC: 13 |
this message has been in BOINC for a long time. Fact beeing that you cannot request work in under 5 minutes (300secs)... I did it yesterday though. I wait about 60 seconds and hit update again and usually returns "no tasks available". I just hadn't seen that message before. I'm guessing that there were actually tasks available that particular time, but because I had requested work too recently, it decided not to send me anything. Linux laptop: record uptime: 1511d 20h 19m (ended due to the power brick giving-up) |
LadyL Send message Joined: 14 Sep 11 Posts: 1679 Credit: 5,230,097 RAC: 0 |
'The current limits are 50 CPU jobs + 400 GPU jobs.' As Joe stated here, there was a change to the way limits work, the project may not be aware of, as the limit used to be 'per C/GPU' when previously applied. We still have a flakey HE router, assimilators not really keeping up, a changeset that is crippeling Anonymous platform, one staff on tour and V7 being tested in beta. (I probably forgot something) You can't be everywhere at once and having to put out small fires, slows the progress on dealing with the wildfire. Eventually everything will be back on track, but that may take weeks. At which point, inevitably one of the servers will hiccup, bringing the project down for a few days. Look at the bright side - work is still flowing :) |
SciManStev Send message Joined: 20 Jun 99 Posts: 6652 Credit: 121,090,076 RAC: 0 |
Although keeping a fast rig fed does require 24/7 baby sitting with the current Quotas, science is getting done, even if it is in starts and stops. Last night before I went to bed, I checked that I had as many GPU tasks as allowed, and none were in the transfer tab. This morning I woke up to empty GPU's and a full transfer tab. A "Retry" got things going again, as BOINC was in a backoff again. I can use the "Retry" remotely while I'm at work, but at least I'm awake to do it. I agree with LadyL. It will clear in time, and in the meantime science is getting done. Steve Warning, addicted to SETI crunching! Crunching as a member of GPU Users Group. GPUUG Website |
soft^spirit Send message Joined: 18 May 99 Posts: 6497 Credit: 34,134,168 RAC: 0 |
this is the first time in weeks I can walk away, and baring backoffs, sleep the night and wake up to the machine still crunching. There are tons of issues, many of which have been the result of untested software thrown in. The router issues re-occuring after reboot would suggest a dead battery, or settings not being stored? On top of that, there are other chronic error issues. But we are kind of used to those. And it is a large to-do list with minimal staff. I do hope the software is repaired before the next version is implemented. I would hate to see old issues carried forward. Janice |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.