Panic Mode On (103) Server Problems?

Message boards : Number crunching : Panic Mode On (103) Server Problems?
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 5 · 6 · 7 · 8 · 9 · 10 · 11 . . . 34 · Next

AuthorMessage
Profile betreger Project Donor
Avatar

Send message
Joined: 29 Jun 99
Posts: 11360
Credit: 29,581,041
RAC: 66
United States
Message 1810422 - Posted: 19 Aug 2016, 16:57:58 UTC - in response to Message 1810417.  

the ones I made yesterday and today.

Mark that should get you 2 stars but I don't think you'll get a toaster.
ID: 1810422 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51468
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1810424 - Posted: 19 Aug 2016, 17:02:35 UTC - in response to Message 1810422.  

the ones I made yesterday and today.

Mark that should get you 2 stars but I don't think you'll get a toaster.

LOL....been waiting for that one for a looooooooooooooong time.

Meow!
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1810424 · Report as offensive
Profile Zalster Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 27 May 99
Posts: 5517
Credit: 528,817,460
RAC: 242
United States
Message 1810449 - Posted: 19 Aug 2016, 18:24:13 UTC - in response to Message 1810424.  

Mark doesn't want any old toaster, he wants the limited edition flying toaster!


ID: 1810449 · Report as offensive
Profile betreger Project Donor
Avatar

Send message
Joined: 29 Jun 99
Posts: 11360
Credit: 29,581,041
RAC: 66
United States
Message 1810455 - Posted: 19 Aug 2016, 18:34:29 UTC - in response to Message 1810449.  

Does the limited edition flying toaster have a setting to char bread?
ID: 1810455 · Report as offensive
Profile Zalster Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 27 May 99
Posts: 5517
Credit: 528,817,460
RAC: 242
United States
Message 1810458 - Posted: 19 Aug 2016, 19:00:28 UTC - in response to Message 1810455.  

It burns the seti@home symbol into the toast!
ID: 1810458 · Report as offensive
Profile Jeff Buck Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester

Send message
Joined: 11 Feb 00
Posts: 1441
Credit: 148,764,870
RAC: 0
United States
Message 1810512 - Posted: 19 Aug 2016, 22:00:46 UTC

It just occurred to me that I've not seen a single guppi come down the pipe today that wasn't a resend, not in at least the last 15+ hours. It appears as if those small MESSIER031 files that started getting split yesterday afternoon might not actually be producing any usable WUs. Has anybody gotten any tasks from those files?
ID: 1810512 · Report as offensive
Profile Zalster Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 27 May 99
Posts: 5517
Credit: 528,817,460
RAC: 242
United States
Message 1810516 - Posted: 19 Aug 2016, 22:13:00 UTC - in response to Message 1810512.  

Just check, nope not seeing any of those from that tape.
ID: 1810516 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14649
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1810518 - Posted: 19 Aug 2016, 22:13:19 UTC - in response to Message 1810512.  

It just occurred to me that I've not seen a single guppi come down the pipe today that wasn't a resend, not in at least the last 15+ hours. It appears as if those small MESSIER031 files that started getting split yesterday afternoon might not actually be producing any usable WUs. Has anybody gotten any tasks from those files?

Yes, I'd been thinking the same thing. No sign of a Messier here either, assigned to either CPU or GPU. And I'm running a larger cache than usual during WOW!, so I'm fetching regularly and bouncing off the 'max in progress' limits.

With approx 450K - 500K results ready to send, and a turnover of ~70K results per hour, you'd expect newly split work to reach the top of the pile some 6 to 7 hours after splitting starts - I think we're well past that point now.

Less easy to speak with certainty about Guppi HIPs using the display tools I have here, but I'd agree there seem to have been resends only for a while. Maybe all this is to do with the test the other Jeff (in the lab) mentioned to Mark, when he restarted the Arecibo flow - lucky we've got that to fall back on.
ID: 1810518 · Report as offensive
Profile Jeff Buck Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester

Send message
Joined: 11 Feb 00
Posts: 1441
Credit: 148,764,870
RAC: 0
United States
Message 1810530 - Posted: 19 Aug 2016, 22:27:37 UTC - in response to Message 1810518.  

With approx 450K - 500K results ready to send, and a turnover of ~70K results per hour, you'd expect newly split work to reach the top of the pile some 6 to 7 hours after splitting starts - I think we're well past that point now.

I think that, for the last several months anyway, the RTS buffer has stayed pretty consistently at or above 500K, except following the Tuesday outage. It's currently down to about 436K, which makes me think that only the pfb splitters are actually feeding it and they can't quite keep up if nothing's coming from the gbt splitters.

I don't remember for sure, but it seems to me that the lowest number MESSIER031 file before the splitters reached them yesterday afternoon was about "_0018". Now I see that "_0037" is the lowest one still listed, so it would seem as if the splitters are burning through them quickly, but just not producing anything useful. Probably just as well, since the first go 'round with MESSIER031 produced mostly -9 overflows, anyway.
ID: 1810530 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13720
Credit: 208,696,464
RAC: 304
Australia
Message 1810533 - Posted: 19 Aug 2016, 22:32:39 UTC - in response to Message 1810530.  

I've gone from all Guppies in my cache, to mostly Arecibo with the odd Guppie resend, to all Arecibo.
Grant
Darwin NT
ID: 1810533 · Report as offensive
Profile arkayn
Volunteer tester
Avatar

Send message
Joined: 14 May 99
Posts: 4438
Credit: 55,006,323
RAC: 0
United States
Message 1810550 - Posted: 19 Aug 2016, 23:46:43 UTC

Same for me, all resends and Acecibo on my system as well.

ID: 1810550 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1810551 - Posted: 19 Aug 2016, 23:46:44 UTC

I thought WUs got 10 tries before being given up on. This one seems to have been killed after just 6 Hosts?
http://setiathome.berkeley.edu/workunit.php?wuid=2237726991
5098684501  6796479   15 Aug 2016, 16:32:49 UTC 16 Aug 2016, 2:41:52 UTC 	Completed, can't validate     893.77 	873.24 	0.00  SETI@home v8 Anonymous platform (NVIDIA GPU)
5098684502  7273253   15 Aug 2016, 16:32:48 UTC 15 Aug 2016, 21:27:37 UTC 	Completed, can't validate  15,961.96 15,668.30  0.00  SETI@home v8 v8.05 i686-pc-linux-gnu
5099962552  7443139   16 Aug 2016, 9:41:03 UTC 	16 Aug 2016, 9:46:09 UTC 	Completed, can't validate       5.59 	3.94 	0.00  SETI@home v8 v8.00 windows_intelx86
5100467037  6872183   16 Aug 2016, 20:43:07 UTC 17 Aug 2016, 1:42:52 UTC 	Completed, can't validate     141.93 	133.95 	0.00  SETI@home v8 v8.03 x86_64-apple-darwin
5101580267  7764122   17 Aug 2016, 8:51:31 UTC 	18 Aug 2016, 6:01:04 UTC 	Completed, can't validate   7,382.78 	143.24 	0.00  SETI@home v8 v8.12 (opencl_atiapu_SoG) windows_intelx86
5103654397  8069725   18 Aug 2016, 13:17:03 UTC 19 Aug 2016, 23:11:58 UTC 	Completed, can't validate  10,426.21 	256.07 	0.00  SETI@home v8 v8.00 (opencl_nvidia_mac) x86_64-apple-darwin

Is this a GUPPI thing?
ID: 1810551 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13720
Credit: 208,696,464
RAC: 304
Australia
Message 1810559 - Posted: 20 Aug 2016, 0:04:53 UTC - in response to Message 1810551.  

I thought WUs got 10 tries before being given up on. This one seems to have been killed after just 6 Hosts?
http://setiathome.berkeley.edu/workunit.php?wuid=2237726991
5098684501  6796479   15 Aug 2016, 16:32:49 UTC 16 Aug 2016, 2:41:52 UTC 	Completed, can't validate     893.77 	873.24 	0.00  SETI@home v8 Anonymous platform (NVIDIA GPU)
5098684502  7273253   15 Aug 2016, 16:32:48 UTC 15 Aug 2016, 21:27:37 UTC 	Completed, can't validate  15,961.96 15,668.30  0.00  SETI@home v8 v8.05 i686-pc-linux-gnu
5099962552  7443139   16 Aug 2016, 9:41:03 UTC 	16 Aug 2016, 9:46:09 UTC 	Completed, can't validate       5.59 	3.94 	0.00  SETI@home v8 v8.00 windows_intelx86
5100467037  6872183   16 Aug 2016, 20:43:07 UTC 17 Aug 2016, 1:42:52 UTC 	Completed, can't validate     141.93 	133.95 	0.00  SETI@home v8 v8.03 x86_64-apple-darwin
5101580267  7764122   17 Aug 2016, 8:51:31 UTC 	18 Aug 2016, 6:01:04 UTC 	Completed, can't validate   7,382.78 	143.24 	0.00  SETI@home v8 v8.12 (opencl_atiapu_SoG) windows_intelx86
5103654397  8069725   18 Aug 2016, 13:17:03 UTC 19 Aug 2016, 23:11:58 UTC 	Completed, can't validate  10,426.21 	256.07 	0.00  SETI@home v8 v8.00 (opencl_nvidia_mac) x86_64-apple-darwin

Is this a GUPPI thing?


Result files-
Pulse count: 6
Pulse count: 4
Workunit error - check skipped
Autocorr count: 30
Pulse count: 5

Spike count: 2
Autocorr count: 0
Pulse count: 3

In the WU information it has "At the max # of error/total/success tasks 5, 10, 5"
I think this falls in to the Maximum number of errors: 5.
5 different machines got 5 completely different results, so the 6th result was ignored & the WU declared dead (and the 6th result was even more different than the other results).

Probably would be good if someone who is able could get a copy of that WU for an autopsy by those than know what they're doing...
Grant
Darwin NT
ID: 1810559 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51468
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1810564 - Posted: 20 Aug 2016, 0:26:41 UTC - in response to Message 1810458.  

It burns the seti@home symbol into the toast!

YES!!!

Meow.
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1810564 · Report as offensive
Profile betreger Project Donor
Avatar

Send message
Joined: 29 Jun 99
Posts: 11360
Credit: 29,581,041
RAC: 66
United States
Message 1810566 - Posted: 20 Aug 2016, 0:31:33 UTC - in response to Message 1810564.  

It burns the seti@home symbol into the toast!

YES!!!

Meow.

Mark no toaster yet and the green stars you deserve aren't there either. Is this the end of the world?
ID: 1810566 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 1810571 - Posted: 20 Aug 2016, 0:51:36 UTC - in response to Message 1810559.  
Last modified: 20 Aug 2016, 0:59:52 UTC

Probably would be good if someone who is able could get a copy of that WU for an autopsy by those than know what they're doing...


Cursory look says 'perfect storm' of cluster-clowns.
- First result is Tbar running 'special' code, known to have some issues
- Second is a Linux host, CPU turning in more invalids than valids
- Third is Stock Windows CPU, usually reference, AMD CPU massively overclocked or broken (looking at its tasks)
- fourth is a stock Mac App (nuff said, lots of invalids)
- Fifth, AMD APU, looks fine to me (maybe, maybe not, TBC)
- sixth, Stock mac OpenCL

So my bet is that the workunit itself is 'fine', and you're just seeing a statistical inevitably of combining 5 kinds of turds, and one poor little APU that seems possibly legit, but we may never really know.

Will be happy if someone wants to dig and disprove or prove my assessment, which is: IMO this is not good enough [Edit: yes I'm starting to get salty, lmao]
"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 1810571 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51468
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1810573 - Posted: 20 Aug 2016, 0:57:43 UTC - in response to Message 1810566.  

It burns the seti@home symbol into the toast!

YES!!!

Meow.

Mark no toaster yet and the green stars you deserve aren't there either. Is this the end of the world?

Naw, me and the kitties will just keep on plodding along for the cause.

Meow.
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1810573 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13720
Credit: 208,696,464
RAC: 304
Australia
Message 1810586 - Posted: 20 Aug 2016, 1:34:12 UTC - in response to Message 1810550.  

Same for me, all resends and Acecibo on my system as well.

Just picked up a lonely Guppie resend.
Grant
Darwin NT
ID: 1810586 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51468
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1810601 - Posted: 20 Aug 2016, 2:50:49 UTC - in response to Message 1810586.  
Last modified: 20 Aug 2016, 2:53:10 UTC

Same for me, all resends and Acecibo on my system as well.

Just picked up a lonely Guppie resend.

The kitties have a lotta Guppies to crunch through first before the get to the coveted Arecibo kibbles.
Oh, for a steady diet of those again.
Makes me dreamy just thinking about it.
That, and dreams about flying toasters...........

Meowsigh.
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1810601 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1810606 - Posted: 20 Aug 2016, 3:12:07 UTC - in response to Message 1810571.  

Probably would be good if someone who is able could get a copy of that WU for an autopsy by those than know what they're doing...


Cursory look says 'perfect storm' of cluster-clowns.
- First result is Tbar running 'special' code, known to have some issues
- Second is a Linux host, CPU turning in more invalids than valids
- Third is Stock Windows CPU, usually reference, AMD CPU massively overclocked or broken (looking at its tasks)
- fourth is a stock Mac App (nuff said, lots of invalids)
- Fifth, AMD APU, looks fine to me (maybe, maybe not, TBC)
- sixth, Stock mac OpenCL

So my bet is that the workunit itself is 'fine', and you're just seeing a statistical inevitably of combining 5 kinds of turds, and one poor little APU that seems possibly legit, but we may never really know.

Will be happy if someone wants to dig and disprove or prove my assessment, which is: IMO this is not good enough [Edit: yes I'm starting to get salty, lmao]

But... that still doesn't explain why the task was Pronounced after a mere 5 tries.
It should have been able to survive another Intel iGPU, a Mac ATI HD4, possibly 1 other, and still been able to Validate on the 10th attempt.
ID: 1810606 · Report as offensive
Previous · 1 . . . 5 · 6 · 7 · 8 · 9 · 10 · 11 . . . 34 · Next

Message boards : Number crunching : Panic Mode On (103) Server Problems?


 
©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.