Message boards :
Number crunching :
Panic Mode On (57) Server problems?
Message board moderation
Previous · 1 . . . 3 · 4 · 5 · 6 · 7 · 8 · 9 . . . 10 · Next
Author | Message |
---|---|
Josef W. Segur Send message Joined: 30 Oct 99 Posts: 4504 Credit: 1,414,761 RAC: 0 |
The scheduler is really funny. That's 0.6 reserved for the ATI GPU (0.3 for each of 2 tasks it's hoping to get). AFAICT the servers don't do anything with that count to ensure you get at least that many tasks for CPU, maybe sometime in the future. Having a non-zero instances does condition some logic to at least consider sending some work, though. Joe |
Dimly Lit Lightbulb 😀 Send message Joined: 30 Aug 08 Posts: 15399 Credit: 7,423,413 RAC: 1 |
Well, thanks to something stuffing up my computer for the second time in as many weeks, a bunch of tasks errored out. Couple that with: 04/10/2011 23:16:35 | SETI@home | Project has no tasks available, I am now down to my final four tasks. It's OK though, they should keep me going until I can get my next fix, I mean, some more tasks :) |
zoom3+1=4 Send message Joined: 30 Nov 03 Posts: 65766 Credit: 55,293,173 RAC: 49 |
I can't even report wu's, http errors at the HE router I guess, I wish the nuts doing this would just get lost. The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's |
OTS Send message Joined: 6 Jan 08 Posts: 369 Credit: 20,533,537 RAC: 0 |
I'm sure all of you just read the latest news post. I wonder if Bank of America uses HE and SETI was caught in the crossfire. http://abcnews.go.com/blogs/business/2011/10/bank-of-america-under-hacking-attack/ |
Cosmic_Ocean Send message Joined: 23 Dec 00 Posts: 3027 Credit: 13,516,867 RAC: 13 |
I wonder if Bank of America uses HE and SETI was caught in the crossfire. Just did a traceroute to see. Goes through a handful of hops on Level3 in Dallas and then directly to BoA's own allocated network addresses. Linux laptop: record uptime: 1511d 20h 19m (ended due to the power brick giving-up) |
zoom3+1=4 Send message Joined: 30 Nov 03 Posts: 65766 Credit: 55,293,173 RAC: 49 |
I'm sure all of you just read the latest news post. Yep, they do as seen here: http://bgp.he.net/AS10794#_whois The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13746 Credit: 208,696,464 RAC: 304 |
Has anyone heard anything on the Scheduler issue? I didn't see anything in the News or Tech news threads. Even after the outage i still can't contact the Scheduler. Grant Darwin NT |
W-K 666 Send message Joined: 18 May 99 Posts: 19080 Credit: 40,757,560 RAC: 67 |
I've had a few contacts over the last few hours, but it has been only about one in five attempts that are sucessful. Saying that I've rec'd no new tasks. Also I noted about 3 hours ago there were 798,200 tasks available, that has now decreased to 783,867 and according to scarecrows graphs, creation rate is near 0.00 as you can get. Think we need a full database clean up and compaction asap, rather than contact with the servers. |
Speedy Send message Joined: 26 Jun 04 Posts: 1643 Credit: 12,921,799 RAC: 89 |
I was under the understanding that this was taken care of this morning while the servers were off line. |
Cosmic_Ocean Send message Joined: 23 Dec 00 Posts: 3027 Credit: 13,516,867 RAC: 13 |
Well normally the tuesday outage does two things. One, it defragments and compresses the database, and then a copy of it is made and deployed on the replica (as well as sent to off-site storage). So one would think that after defragging and compressing the database, it should be running smooth now, right? Well there's only so much of the database that will fit in RAM, and unfortunately, with 13M and climbing waiting to be purged, I'm guessing that is taking up RAM for task page queries or something else. The issues we are experiencing with contacting the scheduler, internal DB operations in the server closet are likely as intermittent/slow because of the bloat. I think turning the scheduler off and turning db_purge on until the backlog is cleared is about the only thing that can rectify this situation. Linux laptop: record uptime: 1511d 20h 19m (ended due to the power brick giving-up) |
W-K 666 Send message Joined: 18 May 99 Posts: 19080 Credit: 40,757,560 RAC: 67 |
Well 40 mins later and the MB's waiting to be sent still remains at 783,867, so definitely something is causing problems at the server end. |
BetelgeuseFive Send message Joined: 6 Jul 99 Posts: 158 Credit: 17,117,787 RAC: 19 |
I agree that things are getting out of control and something should be done about it. However, I think that turning off the scheduler is a bit drastic. Turning off the splitters seems more appropriate. This way people will still be able to report completed tasks (before the deadline). Tom Well normally the tuesday outage does two things. One, it defragments and compresses the database, and then a copy of it is made and deployed on the replica (as well as sent to off-site storage). |
kittyman Send message Joined: 9 Jul 00 Posts: 51468 Credit: 1,018,363,574 RAC: 1,004 |
This is not pretty. The kitties are getting rather depressed. My top rig has been doing MW since yesterday just to keep it warm because it runs 24/7. Two other rigs are out of Seti work, and can't even connect to report the last of what they've completed. The other 5 slower rigs are still crunching up their last. The main router is on the fritz, and the servers are tied up in knots when you can connect. Even the forums are doggy at times. Come on back, Seti. The kitties will leave the light on for ya. "Freedom is just Chaos, with better lighting." Alan Dean Foster |
Kevin Olley Send message Joined: 3 Aug 99 Posts: 906 Credit: 261,085,289 RAC: 572 |
This is not pretty. Set NNT and then update, they will then report. Don't forget to unset NNT after. Kevin |
soft^spirit Send message Joined: 18 May 99 Posts: 6497 Credit: 34,134,168 RAC: 0 |
This is not pretty. Is that a bug work-around to be able to contact the scheduler again? Janice |
W-K 666 Send message Joined: 18 May 99 Posts: 19080 Credit: 40,757,560 RAC: 67 |
Just to let you know I rec'd one task, GPU VHAR, at 07:49:46. |
LadyL Send message Joined: 14 Sep 11 Posts: 1679 Credit: 5,230,097 RAC: 0 |
On a 6.12.x client setting NNT forces 'report straight after upload' behaviour. Which would help if the problem was getting the client to report and not getting the report through... |
soft^spirit Send message Joined: 18 May 99 Posts: 6497 Credit: 34,134,168 RAC: 0 |
ahh gotcha. So no help in the just can not connect to scheduler. I have had 251 completed tasks here for hours on 6.10.58, and simply can not get there from here. Janice |
W-K 666 Send message Joined: 18 May 99 Posts: 19080 Credit: 40,757,560 RAC: 67 |
Time in last post was UTC But checking back I had rec'd another about 20 mins earlier. Strange thing is that Results ready to send is still at 783,867 thats over an hour after I last posted that number. So either the page is lying, or "Results ready to send" cannot be greater than that, for some reason. Unpurged WU/tasks maybe? |
S@NL - John van Gorsel Send message Joined: 5 Jul 99 Posts: 193 Credit: 139,673,078 RAC: 0 |
It could be a coincidence but it worked for me:
The scheduler is still unreachable when asking for new work though... Seti@Netherlands website |
©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.