Message boards :
Number crunching :
Panic Mode On (10) Server problems
Message board moderation
Previous · 1 . . . 6 · 7 · 8 · 9 · 10 · 11 · 12 . . . 13 · Next
Author | Message |
---|---|
Lemat Send message Joined: 3 Apr 99 Posts: 16 Credit: 14,968,143 RAC: 0 |
Just wondering... Why is it that the servers always go down during the weekends? Servers and other equipment ALWAYS work good while skilled technician (or man with sledge hammer) is nearby. |
Zap de Ridder Send message Joined: 9 Jan 00 Posts: 227 Credit: 1,468,844 RAC: 1 |
Looks like some one is working on it at the moment. |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14679 Credit: 200,643,578 RAC: 874 |
The Cricket graph jumped to maximum download rate about half an hour ago: that's 7am, on a Saturday morning, lab time. Yet more unpaid overtime for the boys. |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14679 Credit: 200,643,578 RAC: 874 |
From Server Status page, [As of 15 Nov 2008 20:40:14 UTC]: Results ready to send 1,042,055 Somehow, I don't think so - there were none two hours ago, and I got "no work from project" at 20:23:18 |
Ingleside Send message Joined: 4 Feb 03 Posts: 1546 Credit: 15,832,022 RAC: 13 |
From Server Status page, [As of 15 Nov 2008 20:40:14 UTC]: Actually, it is possible the status-page is correct... For one thing, for performance-reasons the Scheduling-server doesn't check the database directly if any work is available, instead it looks on a shared memory-array the Feeder sets-up. Not sure how large this array of work SETI is using, and neither if Scheduling-server looks-through the whole array for work in case can't find any, but it's possible the array is "empty" one second, until Feeder re-fills the next second. Another way to generate tons of work in that seems very little time is, the Transitioners is responsible for triggering Validator then enough results is reported for a wu, but also to generate new "Tasks" in case of errors, timed-out tasks, or newly-split wu's has been added to the database. For some reason all the Transitioner-processes wasn't running for many hours, but everything else was. Meaning, Scheduling-server happily sent-out all work, and as available work dropped-below whatever limits is set, the Splitters fired-up splitting new wu, and continued splitting new wu's until reached the "disk full"-limit. Since it's much faster for the Transitioner to generate some new database-fields with the Task-info for each wu than it is to split a wu, many hours of already-split wu's can give a huge spike in available work in fairly short time then Transitioners was re-enabled. "I make so many mistakes. But then just think of all the mistakes I don't make, although I might." |
Zap de Ridder Send message Joined: 9 Jan 00 Posts: 227 Credit: 1,468,844 RAC: 1 |
Whatever, just before going to bed I checked the server state page and everything looks back to normal. Not that it's important to me cause my puter is only working 6/24 at 50% and next astro pulse wu wil be monday or tuesday reporting :-) Anyway hail to those or the one that took care of it all today. |
arkayn Send message Joined: 14 May 99 Posts: 4438 Credit: 55,006,323 RAC: 0 |
|
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13854 Credit: 208,696,464 RAC: 304 |
Something's still not quite right. System shows plenty of work available, but i've been getting "No work from project" messages for the last 4 hours. Grant Darwin NT |
Swibby Bear Send message Joined: 1 Aug 01 Posts: 246 Credit: 7,945,093 RAC: 0 |
Been getting "NO WORK FROM PROJECT" for most of the day. Almost out of WUs on all machines now. |
BroncoBob9 Send message Joined: 29 May 03 Posts: 62 Credit: 2,443,241 RAC: 0 |
I was getting "no work available" earlier today, but now have plenty. Everything seems fine. |
arkayn Send message Joined: 14 May 99 Posts: 4438 Credit: 55,006,323 RAC: 0 |
|
Terror Australis Send message Joined: 14 Feb 04 Posts: 1817 Credit: 262,693,308 RAC: 44 |
Same here, my Linux machines are failing downloads with an "http error" or just "0 new tasks" but the Windows boxes are downloading ok. Is this that DNS problem that happened a week or so ago re-occurring ? Brodo |
[B^S] madmac Send message Joined: 9 Feb 04 Posts: 1175 Credit: 4,754,897 RAC: 0 |
|
[B^S] madmac Send message Joined: 9 Feb 04 Posts: 1175 Credit: 4,754,897 RAC: 0 |
|
bounty.hunter Send message Joined: 22 Mar 04 Posts: 442 Credit: 459,063 RAC: 0 |
Was also getting No New Tasks. Exited and shut down the BOINC client and restarted. Straightaway got new tasks. I suspect one of the schedulers is acting up. |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13854 Credit: 208,696,464 RAC: 304 |
Just got a whole lot of work, although still getting the "No work from project" messages on some attempts. Grant Darwin NT |
Sirius B Send message Joined: 26 Dec 00 Posts: 24912 Credit: 3,081,182 RAC: 7 |
I was getting 0 new tasks all night long on the farm, now I'm snowed under. Also got quite a few AP wu's............lovely....... |
arkayn Send message Joined: 14 May 99 Posts: 4438 Credit: 55,006,323 RAC: 0 |
|
gomeyer Send message Joined: 21 May 99 Posts: 488 Credit: 50,370,425 RAC: 0 |
Both my Mac and the Vista machine downloaded new work overnight, but the linux machine is still stuck on downloads. I think it has locked into Vader as its download server and will not release it. This seems to be a similar Linux issue to the one we had a week or so ago. Try adding: 208.68.240.18 boinc2.ssl.berkeley.edu to your hosts file, log off/on and restart the client. This worked for me. |
Iztok s52d (and friends) Send message Joined: 12 Jan 01 Posts: 136 Credit: 393,469,375 RAC: 116 |
Simpler: 1. retry communications, so client wish to download 2. killall -TERM boinc 3. ping boinc2.ssl.berkeley.edu repeat until it returnx x.x.x.13 (we need x.x.x.18) 4. start boinc script again for ping, my linux box is alternating between x.x.x.13/18. If not, /etc/hosts is to be checjed. 73 Iztok |
©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.