Panic Mode On (81) Server Problems?


log in

Advanced search

Message boards : Number crunching : Panic Mode On (81) Server Problems?

Previous · 1 . . . 3 · 4 · 5 · 6 · 7 · 8 · 9 . . . 21 · Next
Author Message
juan BFB
Volunteer tester
Avatar
Send message
Joined: 16 Mar 07
Posts: 4939
Credit: 269,676,323
RAC: 368,312
Brazil
Message 1334416 - Posted: 3 Feb 2013, 21:20:05 UTC - in response to Message 1334413.
Last modified: 3 Feb 2013, 21:20:28 UTC

They don't normally swap tapes until they have been split for both MB & AP, so we could have some wait given the current rate of (not) splitting the APs :-(

The question is, that is doing automaticaly (wait of few hours) or manualy (wait almost a day)?
____________

Profile Mike
Volunteer tester
Avatar
Send message
Joined: 17 Feb 01
Posts: 23377
Credit: 31,811,997
RAC: 24,432
Germany
Message 1334423 - Posted: 3 Feb 2013, 21:26:42 UTC - in response to Message 1334416.

They don't normally swap tapes until they have been split for both MB & AP, so we could have some wait given the current rate of (not) splitting the APs :-(

The question is, that is doing automaticaly (wait of few hours) or manualy (wait almost a day)?


If some are still in the pipe automatically.
If not ..........

____________

msattler
Volunteer tester
Avatar
Send message
Joined: 9 Jul 00
Posts: 38322
Credit: 560,054,833
RAC: 652,999
United States
Message 1334424 - Posted: 3 Feb 2013, 21:29:36 UTC

My faster rigs are already running out of GPU tasks. The slower ones will hang in there for some hours yet.
My guess is we won't see any new MB work until maybe tomorrow, but I can still be hopeful that will change.
____________
*********************************************
Embrace your inner kitty...ya know ya wanna!

I have met a few friends in my life.
Most were cats.

juan BFB
Volunteer tester
Avatar
Send message
Joined: 16 Mar 07
Posts: 4939
Credit: 269,676,323
RAC: 368,312
Brazil
Message 1334429 - Posted: 3 Feb 2013, 21:41:23 UTC - in response to Message 1334424.
Last modified: 3 Feb 2013, 21:42:20 UTC

Im running on SETI drayland too, 100WU on a 2x690 host takes less than 2hr to finish. I realy hate the 100WU limit, they need to change it soon at least to 100WU per GPU.
____________

rob smith
Volunteer tester
Send message
Joined: 7 Mar 03
Posts: 8137
Credit: 52,705,729
RAC: 75,093
United Kingdom
Message 1334432 - Posted: 3 Feb 2013, 21:45:44 UTC

Before doing that they need to sort the ability to deliver the >100 work units per hour that many of the more modern crunchers manage.
(For the last couple of weeks my GTX690 based cruncher has bee attacking Malaria, Einstein, Beta, but not the number of S@H I would like...
____________
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

msattler
Volunteer tester
Avatar
Send message
Joined: 9 Jul 00
Posts: 38322
Credit: 560,054,833
RAC: 652,999
United States
Message 1334433 - Posted: 3 Feb 2013, 21:48:27 UTC - in response to Message 1334432.

Before doing that they need to sort the ability to deliver the >100 work units per hour that many of the more modern crunchers manage.
(For the last couple of weeks my GTX690 based cruncher has bee attacking Malaria, Einstein, Beta, but not the number of S@H I would like...

Well, I do find it rather frustrating when I can't cache enough work to ride out anything much more than a few hours of server problems on my better crunchers.
____________
*********************************************
Embrace your inner kitty...ya know ya wanna!

I have met a few friends in my life.
Most were cats.

ExchangeMan
Volunteer tester
Send message
Joined: 9 Jan 00
Posts: 108
Credit: 125,192,137
RAC: 151,555
United States
Message 1334454 - Posted: 3 Feb 2013, 23:17:22 UTC - in response to Message 1334392.

Actually I just received 3 GPU work units. Someone must be working on it at the lab. Results ready to send now shows as 1.

just resends from WU's who missed their deadline...

You can't be sure of that, his computers are hidden,

I've just received GPU workunits too, they aren't resends, they just happen to be Astropulse GPU Wu's, GPU Wu's none the less.

Claggy

Thanks Claggy. Ya, they weren't resends.

Just curious ..

Which endings the WUs had? x_0, x_1, or x_2, x_3 .. ?


* Best regards! :-) * Sutaru Tsureku, team seti.international founder. * Optimize your PC for higher RAC. * SETI@home needs your help. *

Here's the name of one of the work units. The others had similar long names. No x_0 or x_1 or x_2 or x_3.

28no12ad.22646.124880.140733193388038.10.113

____________

Claggy
Volunteer tester
Send message
Joined: 5 Jul 99
Posts: 4048
Credit: 32,693,237
RAC: 546
United Kingdom
Message 1334458 - Posted: 4 Feb 2013, 0:06:00 UTC - in response to Message 1334454.

Actually I just received 3 GPU work units. Someone must be working on it at the lab. Results ready to send now shows as 1.

just resends from WU's who missed their deadline...

You can't be sure of that, his computers are hidden,

I've just received GPU workunits too, they aren't resends, they just happen to be Astropulse GPU Wu's, GPU Wu's none the less.

Claggy

Thanks Claggy. Ya, they weren't resends.

Just curious ..

Which endings the WUs had? x_0, x_1, or x_2, x_3 .. ?


* Best regards! :-) * Sutaru Tsureku, team seti.international founder. * Optimize your PC for higher RAC. * SETI@home needs your help. *

Here's the name of one of the work units. The others had similar long names. No x_0 or x_1 or x_2 or x_3.

28no12ad.22646.124880.140733193388038.10.113

That's the Workunit name, what about the task name? There are Multiple tasks to a workunit, from 2 to 10.

Claggy

ExchangeMan
Volunteer tester
Send message
Joined: 9 Jan 00
Posts: 108
Credit: 125,192,137
RAC: 151,555
United States
Message 1334460 - Posted: 4 Feb 2013, 0:14:44 UTC - in response to Message 1334458.

Actually I just received 3 GPU work units. Someone must be working on it at the lab. Results ready to send now shows as 1.

just resends from WU's who missed their deadline...

You can't be sure of that, his computers are hidden,

I've just received GPU workunits too, they aren't resends, they just happen to be Astropulse GPU Wu's, GPU Wu's none the less.

Claggy

Thanks Claggy. Ya, they weren't resends.

Just curious ..

Which endings the WUs had? x_0, x_1, or x_2, x_3 .. ?


* Best regards! :-) * Sutaru Tsureku, team seti.international founder. * Optimize your PC for higher RAC. * SETI@home needs your help. *

Here's the name of one of the work units. The others had similar long names. No x_0 or x_1 or x_2 or x_3.

28no12ad.22646.124880.140733193388038.10.113

That's the Workunit name, what about the task name? There are Multiple tasks to a workunit, from 2 to 10.

Claggy

OK, this should be what you're looking for.

1140189392

____________

Profile arkayn
Volunteer tester
Avatar
Send message
Joined: 14 May 99
Posts: 3594
Credit: 47,338,244
RAC: 241
United States
Message 1334483 - Posted: 4 Feb 2013, 1:15:43 UTC - in response to Message 1334460.


That's the Workunit name, what about the task name? There are Multiple tasks to a workunit, from 2 to 10.

Claggy

OK, this should be what you're looking for.

1140189392


I'm going to guess that this is your computer.
http://setiathome.berkeley.edu/show_host_detail.php?hostid=6896017
____________

ExchangeMan
Volunteer tester
Send message
Joined: 9 Jan 00
Posts: 108
Credit: 125,192,137
RAC: 151,555
United States
Message 1334484 - Posted: 4 Feb 2013, 1:23:21 UTC - in response to Message 1334483.


That's the Workunit name, what about the task name? There are Multiple tasks to a workunit, from 2 to 10.

Claggy

OK, this should be what you're looking for.

1140189392


I'm going to guess that this is your computer.
http://setiathome.berkeley.edu/show_host_detail.php?hostid=6896017

Yep. I'm still in the process of putting this cruncher together and need more parts. It's running with the mobo sitting on a wooden table and fans blowing on it.

____________

Cosmic_Ocean
Avatar
Send message
Joined: 23 Dec 00
Posts: 2236
Credit: 8,449,136
RAC: 4,081
United States
Message 1334545 - Posted: 4 Feb 2013, 8:19:32 UTC

Well I did just notice that with 5 AP splitters running and no new MB work being produced.. cricket shows that we are not at max capacity.

So if we are MB-only, we run about 60-70mbit, and if we are AP-only, we run about 60-70mbit. If we could get our pipe to be increased to 150mbit (since my understanding is that it is connected via a gigabit connection, but we're only allowed to use 100mbit of it), that should pretty much "fix" the throughput problem.
____________

Linux laptop uptime: 1484d 22h 42m
Ended due to UPS failure, found 14 hours after the fact

Grant (SSSF)
Send message
Joined: 19 Aug 99
Posts: 5695
Credit: 56,375,799
RAC: 48,730
Australia
Message 1334556 - Posted: 4 Feb 2013, 9:51:15 UTC - in response to Message 1334545.

If we could get our pipe to be increased to 150mbit (since my understanding is that it is connected via a gigabit connection, but we're only allowed to use 100mbit of it), that should pretty much "fix" the throughput problem.

It'd be good to push for a minimum of 250Mb/s. No more traffic jams after outages, and it would give room for growth. Be a bit of a pain to finally get an extra 50Mb/s, only to have to argue for it all over again in a matter of months.
____________
Grant
Darwin NT.

juan BFB
Volunteer tester
Avatar
Send message
Joined: 16 Mar 07
Posts: 4939
Credit: 269,676,323
RAC: 368,312
Brazil
Message 1334560 - Posted: 4 Feb 2013, 10:23:55 UTC

DonĀ“t forget the limits, MB uses 70% of the avaiable BW with them, or more without.

BTW Still running out of new WU, seems like the change of tapes is done manualy and we need to wait someone arrive at the lab, Others porojects say "thanks to that"
____________

Profile Vipin Palazhi
Avatar
Send message
Joined: 29 Feb 08
Posts: 247
Credit: 103,080,895
RAC: 48,915
India
Message 1334590 - Posted: 4 Feb 2013, 14:42:28 UTC

I have been crunching for a few years now without ever actually bothering about how long it takes for the discs to be recorded and how/when it gets transported from Arecibo to Berkeley.

Grant (SSSF)
Send message
Joined: 19 Aug 99
Posts: 5695
Credit: 56,375,799
RAC: 48,730
Australia
Message 1334649 - Posted: 4 Feb 2013, 17:56:56 UTC - in response to Message 1334590.
Last modified: 4 Feb 2013, 18:21:00 UTC

Work is being split again, unfortunately most Scheduler requests result in "Couldn't connect to server" messages.
*deep sigh*


EDIT- make that all requests. The Scheduler is borked.
____________
Grant
Darwin NT.

msattler
Volunteer tester
Avatar
Send message
Joined: 9 Jul 00
Posts: 38322
Credit: 560,054,833
RAC: 652,999
United States
Message 1334655 - Posted: 4 Feb 2013, 18:26:46 UTC - in response to Message 1334649.

Work is being split again, unfortunately most Scheduler requests result in "Couldn't connect to server" messages.
*deep sigh*


EDIT- make that all requests. The Scheduler is borked.

Not ALL requests...
Very hard to get through right now, but the kitties have managed on a few rigs, and have started to get just a bit of GPU work here and there.
But, yes, the scheduler is tied pretty tight right now.
____________
*********************************************
Embrace your inner kitty...ya know ya wanna!

I have met a few friends in my life.
Most were cats.

Grant (SSSF)
Send message
Joined: 19 Aug 99
Posts: 5695
Credit: 56,375,799
RAC: 48,730
Australia
Message 1334658 - Posted: 4 Feb 2013, 18:37:02 UTC - in response to Message 1334655.

Work is being split again, unfortunately most Scheduler requests result in "Couldn't connect to server" messages.
*deep sigh*


EDIT- make that all requests. The Scheduler is borked.

Not ALL requests...
Very hard to get through right now, but the kitties have managed on a few rigs, and have started to get just a bit of GPU work here and there.
But, yes, the scheduler is tied pretty tight right now.

Must be the distance thing again then.
45min & not a single request has gotten through on either of my systems.
____________
Grant
Darwin NT.

msattler
Volunteer tester
Avatar
Send message
Joined: 9 Jul 00
Posts: 38322
Credit: 560,054,833
RAC: 652,999
United States
Message 1334659 - Posted: 4 Feb 2013, 18:42:06 UTC - in response to Message 1334658.

Work is being split again, unfortunately most Scheduler requests result in "Couldn't connect to server" messages.
*deep sigh*


EDIT- make that all requests. The Scheduler is borked.

Not ALL requests...
Very hard to get through right now, but the kitties have managed on a few rigs, and have started to get just a bit of GPU work here and there.
But, yes, the scheduler is tied pretty tight right now.

Must be the distance thing again then.
45min & not a single request has gotten through on either of my systems.

Well, I have 9 rigs making the attempt, so I would have a few more chances to connect. I tried the update button on a couple of them, but it's pretty useless to do so. Just have to let the kitties keep trying whilst I am gone to work and count on random chance to have some work get through.
____________
*********************************************
Embrace your inner kitty...ya know ya wanna!

I have met a few friends in my life.
Most were cats.

Profile Link
Avatar
Send message
Joined: 18 Sep 03
Posts: 823
Credit: 1,545,199
RAC: 288
Germany
Message 1334663 - Posted: 4 Feb 2013, 18:56:39 UTC

Hmm... so they dumped once again lots of unsplited AP. Is AP not worth crunching or how shall we understand that?
____________
.

Previous · 1 . . . 3 · 4 · 5 · 6 · 7 · 8 · 9 . . . 21 · Next

Message boards : Number crunching : Panic Mode On (81) Server Problems?

Copyright © 2014 University of California