Message boards :
Number crunching :
Panic Mode On (77) Server Problems?
Message board moderation
Previous · 1 . . . 11 · 12 · 13 · 14 · 15 · 16 · 17 . . . 22 · Next
Author | Message |
---|---|
juan BFP Send message Joined: 16 Mar 07 Posts: 9786 Credit: 572,710,851 RAC: 3,799 |
I finally did it. I manage to finish a daily quota... :-) Seems like somebody finaly find a way to fix the bandwith problem... 1 task per day for each host. :( |
Link Send message Joined: 18 Sep 03 Posts: 834 Credit: 1,807,369 RAC: 0 |
I finally did it. I manage to finish a daily quota... :-) No, it's just the resend lost tasks feature, which timed out over 100 VLARs at once (see error tasks for this computer) instead of sending them where they belong. The way this thing is working should really be reconsidered, it's not the first time such thing is reported here. |
fscheel Send message Joined: 13 Apr 12 Posts: 73 Credit: 11,135,641 RAC: 0 |
All I can say about the servers now is....WOW! |
.clair. Send message Joined: 4 Nov 04 Posts: 1300 Credit: 55,390,408 RAC: 69 |
new record for me 06/10/2012 19:49:36 | SETI@home | Scheduler request completed: got 115 new tasks Dont know why, Dont care, I`me geting work :¬) |
robertmiles Send message Joined: 16 Jan 12 Posts: 213 Credit: 4,117,756 RAC: 6 |
At times like these I'm glad to jump over to a backup project like Milkyway, LHC, or Cosmology to keep my machines crunching. I'll stop the backup projects when the crisis here is over Easy enough to arrange for an automatic switching - subscribe to both SETI@Home and at least one project, but give the backup project(s) a much lower resource share than SETI@Home, so it will only download from those projects if no workunits are available from SETI@Home. Some versions of BOINC even permit setting the resource share for backup projects to zero without causing problems. |
Wiggo Send message Joined: 24 Jan 00 Posts: 34744 Credit: 261,360,520 RAC: 489 |
My rigs are all happy and the "Results ready to send" are starting to look good but I see still that the 24my12ad file stuck at 14 on the splitters which must be getting close to 3 days in that state now. Cheers. |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13722 Credit: 208,696,464 RAC: 304 |
My rigs are all happy and the "Results ready to send" are starting to look good Starting to fall rapidly again; splitters have dropped to 8/s from 42/s for a couple of hours there. Never quite made it to the 300,000 mark. Grant Darwin NT |
.clair. Send message Joined: 4 Nov 04 Posts: 1300 Credit: 55,390,408 RAC: 69 |
Never quite made it to the 300,000 mark. Not posible. 300000 kilometers per second is nearly the speed of light, splitters can`t do it ;¬) |
Wiggo Send message Joined: 24 Jan 00 Posts: 34744 Credit: 261,360,520 RAC: 489 |
Never quite made it to the 300,000 mark. If that stuck file ever gets freed then we might get to see just how fast they could go. ;-) Yeah things dipped a bit when mine wanted more work but getting between 103-119 tasks at a time can be good. But it has its drawbacks when you get one of those dodgy setup wingmen out there that just pump out constant invalid or errored work, instead of a page or two at a time in the pendings of inconclusives it's now blown out to 3 or 4 pages (quite a few still being those I listed a few months ago). :( Cheers. |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13722 Credit: 208,696,464 RAC: 304 |
Yeah things dipped a bit when mine wanted more work but getting between 103-119 tasks at a time can be good. The most i've ever seen even when my caches were at their lowest was about 25. Over the last few months the most i've ever got in one hit is 12. And with the requests that result in nothing, it takes forever to replace WUs that have been done. Grant Darwin NT |
Wiggo Send message Joined: 24 Jan 00 Posts: 34744 Credit: 261,360,520 RAC: 489 |
Yeah things dipped a bit when mine wanted more work but getting between 103-119 tasks at a time can be good. These high receive rates have only been happening for about the last 24hrs here (basically I've doubled received work I get per request so maybe with the server swaps the feeder has been enlarged, I think that Morten Ross pointed this out in another thread) but they do keep mine from having to constantly hitting the servers up now like they were before (except for the other problem that they create that I've pointed out already which needs something different for a cure) so whatever was changed seems to have worked a treat here. Cheers. |
Nemesis Send message Joined: 14 Mar 07 Posts: 129 Credit: 31,295,655 RAC: 0 |
Lots of strange stuff going on around here... According to the server status page the upload server is disabled, but my uploads are still working. I'm getting plenty of work but also lots of resent lost tasks. My main cruncher got lots of work yesterday but it also has 150+ tasks that timed out yesterday due to the silly 5 minute due date problem... Oh well, so long as I keep getting work i'll be happy. |
rob smith Send message Joined: 7 Mar 03 Posts: 22160 Credit: 416,307,556 RAC: 380 |
The upload server comes and goes. Dropping off line to clear its caches/sort its life out then reappearing in no short order. Bob Smith Member of Seti PIPPS (Pluto is a Planet Protest Society) Somewhere in the (un)known Universe? |
Bernie Vine Send message Joined: 26 May 99 Posts: 9954 Credit: 103,452,613 RAC: 328 |
My main cruncher got lots of work yesterday but it also has 150+ tasks that timed out yesterday due to the silly 5 minute due date problem... This happens where VLARS that were sent to your CPU failed to be delivered for some reason, they were then resent when your GPU requested tasks. VLARS don't work well with NVIDIA GPUS and are not supposed to be sent so they timed out after a few minutes. I know a lot of people are unhappy about this and hopefully something will be done. Perhaps. |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13722 Credit: 208,696,464 RAC: 304 |
I notice not all the AP assimilators are running, and the number of WUs waiting for assimilation continues climb, just hitting 40,000. From what i can gather, it's usually no more than a few 100. Grant Darwin NT |
bill Send message Joined: 16 Jun 99 Posts: 861 Credit: 29,352,955 RAC: 0 |
"VLARS don't work well with NVIDIA GPUS" That is not 100% true. They work just fine on my GTX450, GTX460s, and GTX560Ti's. |
juan BFP Send message Joined: 16 Mar 07 Posts: 9786 Credit: 572,710,851 RAC: 3,799 |
"VLARS don't work well with NVIDIA GPUS" VLARS works on all NVIDIA GPUs but takes for 4x to 8x more time to process so is a waste of resource if you processes then on a GPU, so VLARS -> CPU Only in NVIDIA enviroments. You could see that in you compleated WU. |
bill Send message Joined: 16 Jun 99 Posts: 861 Credit: 29,352,955 RAC: 0 |
"VLARS don't work well with NVIDIA GPUS" According to my Nvidia gpu's, you are wrong. |
Wiggo Send message Joined: 24 Jan 00 Posts: 34744 Credit: 261,360,520 RAC: 489 |
"VLARS don't work well with NVIDIA GPUS" They may work fine, until you wind up with to many of them being crunched at the same time as I found out (when they get like that they seem to leave something behind that eventually clogs the card's memory up), but you'll also find in the end that they are not economical to run on nVIDIA at all. Cheers. |
juan BFP Send message Joined: 16 Mar 07 Posts: 9786 Credit: 572,710,851 RAC: 3,799 |
You must have some kind of out of this world GPUs because that is common knowledge the NVIDIA GPUs are incredible slow when process VLAR´s. You could verify that in several threads in this forum. |
©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.