Panic Mode On (100) Server Problems?

Author	Message
Jord Volunteer tester Send message Joined: 9 Jun 99 Posts: 15184 Credit: 4,362,181 RAC: 3	Message 1723779 - Posted: 9 Sep 2015, 8:46:00 UTC - in response to Message 1723776. It's always a bit lumpy on a Wednesday, today more so than most. Perhaps this guy is really putting his 32 servers online, in a new account. ID: 1723779 ·

Jimbocous Volunteer tester Send message Joined: 1 Apr 13 Posts: 1853 Credit: 268,616,081 RAC: 1,349	Message 1723781 - Posted: 9 Sep 2015, 8:47:59 UTC - in response to Message 1723772. Am I the only one not getting any tasks? It would appear so ... My caches aren't up to the server side imposed limits, but they're very close to it. Ditto. Even saw a tape with APs split a little bit ago, but wasn't able to get any. But almost full on MBs. ID: 1723781 ·

Jimbocous Volunteer tester Send message Joined: 1 Apr 13 Posts: 1853 Credit: 268,616,081 RAC: 1,349	Message 1723784 - Posted: 9 Sep 2015, 8:52:45 UTC - in response to Message 1723779. It's always a bit lumpy on a Wednesday, today more so than most. Perhaps this guy is really putting his 32 servers online, in a new account. Well, dunno about 32, but he did bring 4 new machines up (none with GPUs), and grabbed a couple hundred MBs between them. Given that each CPU is limited to 100 tasks, anything like that isn't really significant in the context of the ~100k per hour demand and throughput we see. ID: 1723784 ·

William Volunteer tester Send message Joined: 14 Feb 13 Posts: 2037 Credit: 17,689,662 RAC: 0	Message 1723811 - Posted: 9 Sep 2015, 11:24:23 UTC lol. or we are down in the history books as the group that found ET :D A person who won't read has no advantage over one who can't read. (Mark Twain) ID: 1723811 ·

qbit Volunteer tester Send message Joined: 19 Sep 04 Posts: 630 Credit: 6,868,528 RAC: 0	Message 1723814 - Posted: 9 Sep 2015, 12:12:38 UTC Finaly got some tasks. Yippie. ID: 1723814 ·

betreger Send message Joined: 29 Jun 99 Posts: 11361 Credit: 29,581,041 RAC: 66	Message 1723922 - Posted: 9 Sep 2015, 17:34:19 UTC The splitters continue to not keep up. ID: 1723922 ·

kittyman Volunteer tester Send message Joined: 9 Jul 00 Posts: 51468 Credit: 1,018,363,574 RAC: 1,004	Message 1723925 - Posted: 9 Sep 2015, 17:47:33 UTC - in response to Message 1723922. Last modified: 9 Sep 2015, 17:48:19 UTC The splitters continue to not keep up. Hopefully they shall over the next few hours. I have finally just in the last 15 minutes or so, filled all 9 rigs to quota for the first time since yesterday's outage. If that is any indication, maybe many other caches are finally being filled also. And results received is only at 91k, so maybe the shorty storm has abated somewhat as well. Meow! "Freedom is just Chaos, with better lighting." Alan Dean Foster ID: 1723925 ·

Richard Haselgrove Volunteer tester Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874	Message 1723946 - Posted: 9 Sep 2015, 19:41:21 UTC OK, I concede. Current result creation rate 0.7919/sec (with zero ready to send, and tapes loaded) is not a happy state of affairs. ID: 1723946 ·

Jord Volunteer tester Send message Joined: 9 Jun 99 Posts: 15184 Credit: 4,362,181 RAC: 3	Message 1723962 - Posted: 9 Sep 2015, 20:32:14 UTC Wasn't there something wrong with these 2011 tapes, perhaps that they couldn't be used by the previous v6 application and that that's why we're re-analyzing them now with v7? If so, perhaps the same problem is still choking the splitters. ET must be on these, making it extremely difficult to get that data extracted. ID: 1723962 ·

Cosmic_Ocean Send message Joined: 23 Dec 00 Posts: 3027 Credit: 13,516,867 RAC: 13	Message 1723994 - Posted: 9 Sep 2015, 22:21:59 UTC - in response to Message 1723962. Wasn't there something wrong with these 2011 tapes, perhaps that they couldn't be used by the previous v6 application and that that's why we're re-analyzing them now with v7? If so, perhaps the same problem is still choking the splitters. ET must be on these, making it extremely difficult to get that data extracted. I think we're just going through them again because we have auto-corr now. And maybe the ability to blank-out the radar-affected areas of the tapes improved since the last time we crunched them on v6. Linux laptop: record uptime: 1511d 20h 19m (ended due to the power brick giving-up) ID: 1723994 ·

jdzukley Send message Joined: 6 Apr 11 Posts: 19 Credit: 26,357,809 RAC: 74	Message 1724018 - Posted: 9 Sep 2015, 23:58:25 UTC Last modified: 10 Sep 2015, 0:31:46 UTC Ya know, it feels great to know that we - the crunchers are beating the servers! YES, TES. It's time to look at the glass as half full. Think about what is going on today with the server farm, the number of tasks being created, crunched, and returned. And noting that the equipment is running all out, not being restrained by the network connections... It has been rare, or perhaps never that the crunchers could beat the servers! WOW - Great - fantastic, better than anyone could have imagined, at least I could not. I am taking the time to smile and be happy! (And, but lets not let the glass get less than half full either!) ID: 1724018 ·

Grant (SSSF) Volunteer tester Send message Joined: 19 Aug 99 Posts: 13731 Credit: 208,696,464 RAC: 304	Message 1724182 - Posted: 10 Sep 2015, 8:17:07 UTC - in response to Message 1724018. Received-in-the-last-hour has dropped down to a much more reasonable number for several hours now, but still the splitters can't build up any sort of buffer. Grant Darwin NT ID: 1724182 ·

Richard Haselgrove Volunteer tester Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874	Message 1724183 - Posted: 10 Sep 2015, 8:27:00 UTC - in response to Message 1724182. Received-in-the-last-hour has dropped down to a much more reasonable number for several hours now, but still the splitters can't build up any sort of buffer. "Results out in the field" is rising steadily, as it was before last night's glitch - so at least the splitters are filling people's caches. ID: 1724183 ·

Wiggo Send message Joined: 24 Jan 00 Posts: 34744 Credit: 261,360,520 RAC: 489	Message 1724186 - Posted: 10 Sep 2015, 8:47:04 UTC - in response to Message 1724183. Last modified: 10 Sep 2015, 8:47:31 UTC Received-in-the-last-hour has dropped down to a much more reasonable number for several hours now, but still the splitters can't build up any sort of buffer. "Results out in the field" is rising steadily, as it was before last night's glitch - so at least the splitters are filling people's caches. Between the iiNet outages I had here and yesterday's long outrage I've developed ghosts on my backup rig and received penalties from my main rig's GPU backup project. :-( It's been years since I've had either. :-D damn 6.10.60 is good, it took a combined effort to cause them :-p Cheers. ID: 1724186 ·

kittyman Volunteer tester Send message Joined: 9 Jul 00 Posts: 51468 Credit: 1,018,363,574 RAC: 1,004	Message 1724280 - Posted: 10 Sep 2015, 15:54:47 UTC Ruh roh, Astrokitty....the replica DB went offline about half an hour ago. It could be somebody in the lab at this time of the day, or........... Panic mode time. "Freedom is just Chaos, with better lighting." Alan Dean Foster ID: 1724280 ·

kittyman Volunteer tester Send message Joined: 9 Jul 00 Posts: 51468 Credit: 1,018,363,574 RAC: 1,004	Message 1724292 - Posted: 10 Sep 2015, 16:35:42 UTC - in response to Message 1724291. Last modified: 10 Sep 2015, 16:42:01 UTC Ruh roh, Astrokitty....the replica DB went offline about half an hour ago. It could be somebody in the lab at this time of the day, or........... Panic mode time. <Optimist ON> That's just for the preparation of an almost endless supply of AP's. <Optimist OFF> Ahhh.........hope springs eternal, eh? EDIT... And the replica is back, 741 seconds behind. "Freedom is just Chaos, with better lighting." Alan Dean Foster ID: 1724292 ·

kittyman Volunteer tester Send message Joined: 9 Jul 00 Posts: 51468 Credit: 1,018,363,574 RAC: 1,004	Message 1724308 - Posted: 10 Sep 2015, 17:12:30 UTC - in response to Message 1724296. Last modified: 10 Sep 2015, 17:13:23 UTC Ruh roh, Astrokitty....the replica DB went offline about half an hour ago. It could be somebody in the lab at this time of the day, or........... Panic mode time. <Optimist ON> That's just for the preparation of an almost endless supply of AP's. <Optimist OFF> Ahhh.........hope springs eternal, eh? EDIT... And the replica is back, 741 seconds behind. Very good. Now the optimist only waits for the "almost endless supply of AP's" :-) Alas, another AP dataset appears, but it is just another drive-by scanning. "Freedom is just Chaos, with better lighting." Alan Dean Foster ID: 1724308 ·

Jimbocous Volunteer tester Send message Joined: 1 Apr 13 Posts: 1853 Credit: 268,616,081 RAC: 1,349	Message 1724335 - Posted: 10 Sep 2015, 18:28:55 UTC - in response to Message 1724308. Alas, another AP dataset appears, but it is just another drive-by scanning. Amazingly, I got a whole 2 APs out of that. Warmed the GPU for almost an hour. Wow ... ID: 1724335 ·

kittyman Volunteer tester Send message Joined: 9 Jul 00 Posts: 51468 Credit: 1,018,363,574 RAC: 1,004	Message 1724336 - Posted: 10 Sep 2015, 18:30:13 UTC - in response to Message 1724335. Alas, another AP dataset appears, but it is just another drive-by scanning. Amazingly, I got a whole 2 APs out of that. Warmed the GPU for almost an hour. Wow ... Were they really from that dataset, or were they really resends? I got no new APs, just a couple of timed out resends. "Freedom is just Chaos, with better lighting." Alan Dean Foster ID: 1724336 ·

Jimbocous Volunteer tester Send message Joined: 1 Apr 13 Posts: 1853 Credit: 268,616,081 RAC: 1,349	Message 1724337 - Posted: 10 Sep 2015, 18:32:04 UTC - in response to Message 1724336. Amazingly, I got a whole 2 APs out of that. Warmed the GPU for almost an hour. Wow ... Were they really from that dataset, or were they really resends? I got no new APs, just a couple of timed out resends. You're right. Should have looked more closely. Just resends ... ID: 1724337 ·

©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.