Message boards :
Number crunching :
Panic Mode On (46) Server problems
Message board moderation
Previous · 1 . . . 7 · 8 · 9 · 10 · 11 · Next
Author | Message |
---|---|
Gundolf Jahn Send message Joined: 19 Sep 00 Posts: 3184 Credit: 446,358 RAC: 0 |
Am I seeing things or did the Server Status page lock up at exactly the same time as the day before (23:50 UTC)? Gruß, Gundolf |
Donald L. Johnson Send message Joined: 5 Aug 02 Posts: 8240 Credit: 14,654,533 RAC: 20 |
Am I seeing things or did the Server Status page lock up at exactly the same time as the day before (23:50 UTC)? Can't say about what time the status page locked, but downloads died about 20:00 PDT/03:00 UTC versus 19:00 PDT/02:00 UTC the day before. Cleaning crew issues again? Has Matt said if they got all the flakey drives on jocelyn fixed? Looks like we are still running everything off the primary database. Donald Infernal Optimist / Submariner, retired |
Highlander Send message Joined: 5 Oct 99 Posts: 167 Credit: 37,987,668 RAC: 16 |
Am I seeing things or did the Server Status page lock up at exactly the same time as the day before (23:50 UTC)? no, you must not change your mind and believe in ghosts :-). was wondering about it too. and additional since 9:40 UTC, i cannot upload anymore; can also only hope, this was the cleaning staff :-/, but i dont believe in ghosts too ... - Performance is not a simple linear function of the number of CPUs you throw at the problem. - |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
Am I seeing things or did the Server Status page lock up at exactly the same time as the day before (23:50 UTC)? Welcome to Groundhog Day! :) |
Miep Send message Joined: 23 Jul 99 Posts: 2412 Credit: 351,996 RAC: 0 |
Can we perhaps get a more eyecatching title on the next panic thread? something on the lines of 'up-/downloads or reports not working? look here!' ? Or I just write something I can copy/paste into each new 'I can't upload/download/report/why isn't this working' thread... Welcome to Groundhog Day! :) Am I allowed to bury under my blankets for the next 24 days? Carola ------- I'm multilingual - I can misunderstand people in several languages! |
Ellis Hardin Send message Joined: 15 Mar 01 Posts: 33 Credit: 26,603,764 RAC: 0 |
Bah. I have my preferences set to connect every 10 days and maintain an additional 10 days of work for a total of a 20 day cache. Since I've been using a 20 day cache, I've only ran out of work once, during the big outage late September to mid October last year. |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
I thought we were asked not to have caches that big as it adversely affected our wingmen? Some of us keep asking "be kind to the servers, please?", but not everyone chooses to listen. |
SciManStev Send message Joined: 20 Jun 99 Posts: 6652 Credit: 121,090,076 RAC: 0 |
I thought we were asked not to have caches that big as it adversely affected our wingmen? I carry a big cache, but I don't think my wingmen have to wait very long..... Steve Warning, addicted to SETI crunching! Crunching as a member of GPU Users Group. GPUUG Website |
Mike Send message Joined: 17 Feb 01 Posts: 34258 Credit: 79,922,639 RAC: 80 |
You can´t generalize that. I can download my units when the load on the servers is low. Evenso it makes no difference i download 100 units a day no matter whats actually in my queue. Just one example. With each crime and every kindness we birth our future. |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
I thought we were asked not to have caches that big as it adversely affected our wingmen? It isn't the size that matters, but the duration (latency). If people hold on to tasks for 20 days, as Ellis suggested, than the workunit datafile has to be held on the server for 20 days, in case it's needed again for a resend. Often longer, because at that setting shorties will jump the queue and delay mid-AR even longer. Try multiplying up 50,000 results per hour, x 0.5 for WUs, x 24, x 20, x 360 KiB. Then do the same for Astropulse, and add the two results. Then ask yourself what happens when a hard disk fails. |
Jason Safoutin Send message Joined: 8 Sep 05 Posts: 1386 Credit: 200,389 RAC: 0 |
I maintain work for 7 days, but usually takes me twice that to get through all the WU's as I don't always have BOINC running (mostly while I am at PC). I also run BETA, with just about the same cache. This is of course the first time I ran into any problem. "By faith we understand that the universe was formed at God's command, so that what is seen was not made out of what was visible". Hebrews 11.3 |
Sutaru Tsureku Send message Joined: 6 Apr 07 Posts: 7105 Credit: 147,663,825 RAC: 5 |
Correct me if I'm wrong, AFAIK the max WU cache is 10 days (also if you set 10 + 10). - Best regards! - Sutaru Tsureku, team seti.international founder. - Optimize your PC for higher RAC. - SETI@home needs your help. - |
Miep Send message Joined: 23 Jul 99 Posts: 2412 Credit: 351,996 RAC: 0 |
26/05/2011 18:00:29 | | [work_fetch] target work buffer: 864000.00 + 864000.00 sec Looks suspiciously like 10 + 10 = 20 to me. I found running 2+2.5 is more than enough these days. Carola ------- I'm multilingual - I can misunderstand people in several languages! |
Tim Send message Joined: 19 May 99 Posts: 211 Credit: 278,575,259 RAC: 0 |
Yes Sutaru, I also think 10 days is the maximum. But even so, for me, with 5 gpus running 24/7, 5 days of cache is more than enough. |
Sutaru Tsureku Send message Joined: 6 Apr 07 Posts: 7105 Credit: 147,663,825 RAC: 5 |
26/05/2011 18:00:29 | | [work_fetch] target work buffer: 864000.00 + 864000.00 sec Thanks. In past I was running 'additional' 10 days & 'always connected' (no entry), all was fine (nearly always first in first out (FIFO) mode). If I set every 10 days connected and 0 days additonal, always if shorties came down, BOINC gone in 'high priority mode' (no work request until calculated). So I guess it will never happen (if 10 + 10) that BOINC will have full 20 days WUs. Yes/No? ----------------------------- BTW. News subforum : work generation temporarily halted We are having difficulties with one of our storage servers. Until we fix this we are unable to generate new work. - Best regards! - Sutaru Tsureku, team seti.international founder. - Optimize your PC for higher RAC. - SETI@home needs your help. - |
Sutaru Tsureku Send message Joined: 6 Apr 07 Posts: 7105 Credit: 147,663,825 RAC: 5 |
Yes Sutaru, I also think 10 days is the maximum. But even so, for me, with 5 gpus running 24/7, 5 days of cache is more than enough. IIRC, so it was in past, max 10 days. Maybe also if you set 10 + 10 in the prefs (and BOINC ask for), the S@h scheduler server send you max 10 days.. - Best regards! - Sutaru Tsureku, team seti.international founder. - Optimize your PC for higher RAC. - SETI@home needs your help. - |
Miep Send message Joined: 23 Jul 99 Posts: 2412 Credit: 351,996 RAC: 0 |
Maybe if you set 10 + 10 in the prefs and let it run for a while with the appropriate logging set in cc_config, I can convince you that 864000.00 + 864000.00 sec equals a 10 + 10 = 20 day cache? If you actually manage to grab 20 days worth of tasks is a different question but BOINC WILL try to get as much as you ask for. And that's without going into the deeper parts of client state to tamper with how much you'll be asking for. Carola ------- I'm multilingual - I can misunderstand people in several languages! |
Mike Send message Joined: 17 Feb 01 Posts: 34258 Credit: 79,922,639 RAC: 80 |
I´m surprised at least one got it. In reality its a little more complicated but you can get 20 days worth. With each crime and every kindness we birth our future. |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
If I set every 10 days connected and 0 days additonal, always if shorties came down, BOINC gone in 'high priority mode' (no work request until calculated). Yes. If you get too greedy, sometimes you don't get anything at all. That happens in life, too - often for children, but hopefully most of them grow out of it. Now, can we get back on topic, please? We're supposed to be panicking about the SERVER'S problems, not our own. I see from the front page that We are having difficulties with one of our storage servers. Until we fix this we are unable to generate new work. Please help the servers by not requesting, or holding on to, more work than you really need. Don't fight against them. |
SciManStev Send message Joined: 20 Jun 99 Posts: 6652 Credit: 121,090,076 RAC: 0 |
My cache is set for 8.5 days, and if I set it larger, BOINC replies that I have reached the maximum amount of tasks. There are about 6000 wu's at any given time, and they are never, ever returned late. Based on the last two large outages, I will run dry fairly close to the 8.5 days, sometimes a little less. Evey time I consider lowering my cache, we get another multi-day outage, so I've settled where I am. Steve Warning, addicted to SETI crunching! Crunching as a member of GPU Users Group. GPUUG 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.