Message boards :
Number crunching :
Panic Mode On (103) Server Problems?
Message board moderation
Previous · 1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 . . . 34 · Next
Author | Message |
---|---|
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
And when the server processes are stopped and restarted - say for Tuesday maintenance - the splitters appear to start working on the most recently loaded batch, rather than resuming older batches. Within each batch, yes. But you need to be aware of when each collection of 'tapes' started to appear on the page: the partially-split blc7 tapes have been there the longest, and the partially-split blc5 tapes have been there less long. The active blc3 tapes are the most recent arrivals. A one-time glance at the SSP isn't enough to give you the full picture. (I must get out more...) |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
Interesting arrival of a bunch of small MESSIER031 tapes on the SSP. I imagine I won't have to worry about those at least until the morning. |
Jeff Buck Send message Joined: 11 Feb 00 Posts: 1441 Credit: 148,764,870 RAC: 0 |
Interesting arrival of a bunch of small MESSIER031 tapes on the SSP. I imagine I won't have to worry about those at least until the morning. Seems to be from observations a day or two earlier than all those MESSIER031 files we got back in April. This group is all from the 57396 date, while all the ones in that first "messy" group were either 57397 or 57398. I wonder if these will be just as noisy or if they've tried cleaning them up at all before splitting them. |
Speedy Send message Joined: 26 Jun 04 Posts: 1643 Credit: 12,921,799 RAC: 89 |
Interesting arrival of a bunch of small MESSIER031 tapes on the SSP. I imagine I won't have to worry about those at least until the morning. If they are noisy we will burn through them reasonably quickly. I am assuming by noisy you mean tasks that run fast? |
Jeff Buck Send message Joined: 11 Feb 00 Posts: 1441 Credit: 148,764,870 RAC: 0 |
If they are noisy we will burn through them reasonably quickly. I am assuming by noisy you mean tasks that run fast? Yeah, I think about 70% of them resulted in -9 overflows. I remember posting the following at the time: Was just checking my machines before I went to bed and watched one blow through 7 consecutive messier031 tasks in 41 seconds. They weren't VLARs, either, so even if they weren't overflows, the NVIDIA GPUs handled them just fine, IIRC. |
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 |
I wonder if these will be just as noisy or if they've tried cleaning them up at all before splitting them. This combining with different AR than BLC ones could mean not correctly chosen thresholds for such GBT data... If new bunch will have excessive noise too worth to attract Eric attention to this issue. SETI apps news We're not gonna fight them. We're gonna transcend them. |
betreger Send message Joined: 29 Jun 99 Posts: 11361 Credit: 29,581,041 RAC: 66 |
Where did we go? |
rob smith Send message Joined: 7 Mar 03 Posts: 22189 Credit: 416,307,556 RAC: 380 |
Third door on the left, the invisible one without a handle or name plate.... Bob Smith Member of Seti PIPPS (Pluto is a Planet Protest Society) Somewhere in the (un)known Universe? |
Zalster Send message Joined: 27 May 99 Posts: 5517 Credit: 528,817,460 RAC: 242 |
So it looks like seti is reporting negative credits to free DC and BoincStats. Anyone else seeing this? Is this related to outage the other day? |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
So it looks like seti is reporting negative credits to free DC and BoincStats. BOINC projects don't export stats 'to' anywhere at all: they just make the data available at a public place, and say "come and get them" - so if two different sites show the same problem, it's a sure guess that there's something wrong with the exported data, rather than the stats sites' interpretation of them. In our case, the public place is http://setiathome.berkeley.edu/stats/ Looking at the 12 August version of those files, there's something wrong with tables.xml - there should be summary lines for users, hosts, teams and project total credit at the top, under the 'update time'. The server status page was also squiffy for a while after the server outage, and I'm guessing you're right - they are related. Also note that the replica database is currently about 25K seconds behind the master, which can affect values too. I'd suggest we wait and see if the files return to normal today or tomorrow, and drop a note to the staff on Monday if they're still out of kilter. |
Zalster Send message Joined: 27 May 99 Posts: 5517 Credit: 528,817,460 RAC: 242 |
Cool, thanks Richard |
WezH Send message Joined: 19 Aug 99 Posts: 576 Credit: 67,033,957 RAC: 95 |
I'd suggest we wait and see if the files return to normal today or tomorrow, and drop a note to the staff on Monday if they're still out of kilter. Hopefully files do get fixed before SETI@Home Wow Event 2016 starts Monday 16.00 UTC... |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
I'd suggest we wait and see if the files return to normal today or tomorrow, and drop a note to the staff on Monday if they're still out of kilter. The Wow stats are updated hourly, so they must use a different source from the public dump, which is only updated once every 24 hours. |
Cosmic_Ocean Send message Joined: 23 Dec 00 Posts: 3027 Credit: 13,516,867 RAC: 13 |
I'm wondering how long those 71 MBv7 results are going to be waiting for DB purging. Normally, they stick around for ~24 hours after being assimilated, but it's been a few days now. Linux laptop: record uptime: 1511d 20h 19m (ended due to the power brick giving-up) |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
Looks like BOINCstats is finding some new credits in today's dump, but we won't see the full picture until tomorrow's 'daily update' is complete. |
arkayn Send message Joined: 14 May 99 Posts: 4438 Credit: 55,006,323 RAC: 0 |
I'd suggest we wait and see if the files return to normal today or tomorrow, and drop a note to the staff on Monday if they're still out of kilter. I think they are scanning the actual account pages for the participants. |
Wedge009 Send message Joined: 3 Apr 99 Posts: 451 Credit: 431,396,357 RAC: 553 |
I did notice the negative credit in BOINCStats and I wonder if it's related to a situation I've been experiencing since at least about 21:00 UTC on 13th August. From my home network I have had no access to S@h - or anywhere else on berkeley.edu for that matter. But checking at my parents' place that same day (about 02:00 UTC 14th August) and also at my work office just now (20:00 UTC 14th August), I see no issues with berkeley.edu so I'm wondering if there's some routing problem somewhere between my home network and Berkeley. Just to be clear, I have no other Internet issues on my home network and even gave the router a restart to get a fresh connection with my ISP. Am I really the only one with this issue? Soli Deo Gloria |
Zalster Send message Joined: 27 May 99 Posts: 5517 Credit: 528,817,460 RAC: 242 |
No problems here at my different locations. Might want to your antivirus program and make sure it isn't blocking your access |
Keith Myers Send message Joined: 29 Apr 01 Posts: 13164 Credit: 1,160,866,277 RAC: 1,873 |
You might want to do a traceroute to the Seti servers. http://boinc2.ssl.berkeley.edu http://georgem.ssl.berkeley.edu http://vader.ssl.berkeley.edu You might also add them to your allowed URL "whitelist" in your AV program. I was getting download errors in mine till I whitelisted the servers for Seti and Einstein. Seti@Home classic workunits:20,676 CPU time:74,226 hours A proud member of the OFA (Old Farts Association) |
Wedge009 Send message Joined: 3 Apr 99 Posts: 451 Credit: 431,396,357 RAC: 553 |
The issue is at the network level, so I really doubt anti-virus is involved. Access is blocked to all services on berkeley.edu across all devices on my home network - Windows and Linux. I did try a traceroute last night (at home) but I was using an on-line service (which was successful) and didn't think to try it from one of my computers. I'll try that when I get home. Soli Deo Gloria |
©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.