Panic Mode On (79) Server Problems?


log in

Advanced search

Message boards : Number crunching : Panic Mode On (79) Server Problems?

Previous · 1 . . . 11 · 12 · 13 · 14 · 15 · 16 · 17 . . . 23 · Next
Author Message
N9JFE David SProject donor
Volunteer tester
Avatar
Send message
Joined: 4 Oct 99
Posts: 11998
Credit: 14,661,510
RAC: 12,131
United States
Message 1311573 - Posted: 6 Dec 2012, 0:28:53 UTC

Since this thread went to its 4th page a couple days before the outage, it's been causing my employer's content filter to barf (I work in a school). I finally remembered to read it here at home and I don't see anything obviously offensive, so I'm guessing it has to do with all that .bat file coding. Anyway, I just hope it clears up when the thread goes to page 5.

Meanwhile, I'll just let my machines plug along on Einstein and whatever Seti they can get. The Einstein work does wonders for my overall Boinc stats. I do note that the one new task my i7 got downloaded in 10 seconds.

____________
David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.


zoom314Project donor
Avatar
Send message
Joined: 30 Nov 03
Posts: 46519
Credit: 36,865,313
RAC: 5,104
United States
Message 1311625 - Posted: 6 Dec 2012, 6:01:10 UTC
Last modified: 6 Dec 2012, 6:02:43 UTC

Now 'No Headers, No Data'... Uploads are ok, reporting is a problem.

727 SETI@home 12/5/2012 9:57:31 PM Scheduler request failed: Server returned nothing (no headers, no data)


Oops now I got 15 new wu's, weird, traffic maybe?
____________
My Facebook, War Commander, 2015

Grant (SSSF)
Send message
Joined: 19 Aug 99
Posts: 5868
Credit: 60,618,280
RAC: 47,488
Australia
Message 1311649 - Posted: 6 Dec 2012, 7:21:12 UTC - in response to Message 1311625.
Last modified: 6 Dec 2012, 7:56:12 UTC

While i was at work the the project came back up my systems were able to get some work, but most Scheduler requests result in timeouts again.


EDIT- and when i finally managed to make contact, i got a new message.
6/12/2012 17:11:23 SETI@home Message from server: Didn't resend lost task 01jn12ab.19033.99684.140733193388039.10.255.vlar_1 (expired)

That was in response to a request for more GPU work.
____________
Grant
Darwin NT.

Profile Alex Storey
Volunteer tester
Avatar
Send message
Joined: 14 Jun 04
Posts: 553
Credit: 1,667,736
RAC: 575
Greece
Message 1311732 - Posted: 6 Dec 2012, 13:20:37 UTC - in response to Message 1311695.

Meowghost.


I set to NNT when I started getting ghosted up but then someone mentioned that they do get through. So I did get them overnight.

I guess the good news is that they aren't getting stuck in limbo like last time!

Lionel
Send message
Joined: 25 Mar 00
Posts: 576
Credit: 236,123,801
RAC: 231,631
Australia
Message 1311990 - Posted: 7 Dec 2012, 0:19:41 UTC


uploads stuck ...
____________

Lionel
Send message
Joined: 25 Mar 00
Posts: 576
Credit: 236,123,801
RAC: 231,631
Australia
Message 1311993 - Posted: 7 Dec 2012, 0:28:05 UTC - in response to Message 1311990.

uploads stuck ...


edit: only stuck via the proxy I was using ... uurrggghhhh

____________

rob smithProject donor
Volunteer tester
Send message
Joined: 7 Mar 03
Posts: 8535
Credit: 59,492,229
RAC: 87,051
United Kingdom
Message 1312050 - Posted: 7 Dec 2012, 6:28:24 UTC

You've just hit one of the snags of using a proxy, they have a habit of stopping working if you hit them too hard too often with one target address.
____________
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

Profile S@NL Etienne Dokkum
Volunteer tester
Avatar
Send message
Joined: 11 Jun 99
Posts: 165
Credit: 17,062,935
RAC: 20,804
Netherlands
Message 1312058 - Posted: 7 Dec 2012, 7:09:49 UTC

What has happened ???!!!

This morning I look at my tasks and Seti tells me I aborted all 200 tasks last night at 18.47 UTC although they are still running, uploading and reporting. No credit for any tasks since then and they don't show as pending or validated even after I reported them.

Anyone else got this problem yet ?
____________

zoom314Project donor
Avatar
Send message
Joined: 30 Nov 03
Posts: 46519
Credit: 36,865,313
RAC: 5,104
United States
Message 1312059 - Posted: 7 Dec 2012, 7:12:56 UTC - in response to Message 1312058.

Nor here, earlier I had a Win7 BSOD, totally weird, that was a driver problem...
____________
My Facebook, War Commander, 2015

Profile Khangollo
Avatar
Send message
Joined: 1 Aug 00
Posts: 245
Credit: 36,410,524
RAC: 0
Slovenia
Message 1312081 - Posted: 7 Dec 2012, 7:50:59 UTC - in response to Message 1312058.
Last modified: 7 Dec 2012, 7:56:31 UTC

What has happened ???!!!

This morning I look at my tasks and Seti tells me I aborted all 200 tasks last night at 18.47 UTC although they are still running, uploading and reporting. No credit for any tasks since then and they don't show as pending or validated even after I reported them.

Anyone else got this problem yet ?

Yes, this is normal seti@home function now. Auto abandoning all tasks at random.
Apparently, I got full cache of 22 astropulses abandoned yesterday (again!). This is the third time this happened to me. Usually it happens when scheduler connection times out a lot.
Fun.

You should manually abort all those tasks now.
____________

Grant (SSSF)
Send message
Joined: 19 Aug 99
Posts: 5868
Credit: 60,618,280
RAC: 47,488
Australia
Message 1312082 - Posted: 7 Dec 2012, 7:52:39 UTC - in response to Message 1312081.


Add to that, neither the MB or AP assimilators appear to be working. They both show green on the server status page, but the number waiting to be assimilated continues to climb for both.
____________
Grant
Darwin NT.

Grant (SSSF)
Send message
Joined: 19 Aug 99
Posts: 5868
Credit: 60,618,280
RAC: 47,488
Australia
Message 1312084 - Posted: 7 Dec 2012, 7:58:41 UTC - in response to Message 1312081.

What has happened ???!!!

This morning I look at my tasks and Seti tells me I aborted all 200 tasks last night at 18.47 UTC although they are still running, uploading and reporting. No credit for any tasks since then and they don't show as pending or validated even after I reported them.

Anyone else got this problem yet ?

Yes, this is normal seti@home function now. Auto abandoning all tasks at random.
Apparently, I got full cache of 22 astropulses abandoned yesterday (again!). This is the third time this happened to me. Usually it happens when scheduler connection times out a lot.
Fun.

You should manually abort all those tasks now.

And to add to that, since the last outage i've had a whole bunch of tasks timeout- within minutes of getting them (one took a couple of hours but most were within 10 minutes).
____________
Grant
Darwin NT.

Grant (SSSF)
Send message
Joined: 19 Aug 99
Posts: 5868
Credit: 60,618,280
RAC: 47,488
Australia
Message 1312299 - Posted: 7 Dec 2012, 21:08:00 UTC - in response to Message 1312082.

Add to that, neither the MB or AP assimilators appear to be working. They both show green on the server status page, but the number waiting to be assimilated continues to climb for both.

MB & AP assimilator backlogs continue to climb.
____________
Grant
Darwin NT.

Tom*
Send message
Joined: 12 Aug 11
Posts: 114
Credit: 4,815,274
RAC: 248
United States
Message 1312375 - Posted: 8 Dec 2012, 3:19:55 UTC
Last modified: 8 Dec 2012, 3:22:46 UTC

Since the grass needs to be cut, everything seems to be running optimally.

I know the really heavy consumers of internet traffic are busy elsewhere,

but limiting internet downloads seems to help the schedular and the download

speeds. Now if we can switch over to AP's with no Shorties or MB's that would

be an excellent test of increasing the bandwidth from 100 Mbts to 200 Mbts.

Posing a question - Can 100Mbits handle confortably just AP's with no MB's

as 100 Mbits seem to be able to handle just MB's currently????

Its a question that begs IMHO to be answered.

Thanks

Bill aka Tom

Profile HAL9000
Volunteer tester
Avatar
Send message
Joined: 11 Sep 99
Posts: 4447
Credit: 119,194,137
RAC: 140,191
United States
Message 1312379 - Posted: 8 Dec 2012, 3:31:06 UTC - in response to Message 1312375.
Last modified: 8 Dec 2012, 3:33:19 UTC

Since the grass needs to be cut, everything seems to be running optimally.

I know the really heavy consumers of internet traffic are busy elsewhere,

but limiting internet downloads seems to help the schedular and the download

speeds. Now if we can switch over to AP's with no Shorties or MB's that would

be an excellent test of increasing the bandwidth from 100 Mbts to 200 Mbts.

Posing a question - Can 100Mbits handle confortably just AP's with no MB's

as 100 Mbits seem to be able to handle just MB's currently????

Its a question that begs IMHO to be answered.

Thanks

Bill aka Tom

AP tasks are about 20 times more data to transfer, but only require 10 times the amount of time to process. At leas for CPU. I imagine GPU processing would make the ratio even worse.
However if that was the only work going out it would be hard to say. Not all the current machines running can complete an AP tasks in time & some choose not to do them at all.
____________
SETI@home classic workunits: 93,865 CPU time: 863,447 hours

Join the BP6/VP6 User Group today!

Profile ivan
Volunteer tester
Avatar
Send message
Joined: 5 Mar 01
Posts: 625
Credit: 143,578,837
RAC: 150,127
United Kingdom
Message 1312535 - Posted: 8 Dec 2012, 12:38:47 UTC - in response to Message 1312375.

Posing a question - Can 100Mbits handle confortably just AP's with no MB's

as 100 Mbits seem to be able to handle just MB's currently????

Its a question that begs IMHO to be answered.

Well, as I see it, an AP WU is 8 MB; our download bandwidth peaks at 94 Mb/s, being generous that's 10 MB/s -- so we max out at 1.25 AP WUs/sec! Any more pressure than that and we get into the realm of lost ACKs and ghost workunits.
____________

juan BFBProject donor
Volunteer tester
Avatar
Send message
Joined: 16 Mar 07
Posts: 5414
Credit: 306,632,149
RAC: 330,511
Brazil
Message 1312565 - Posted: 8 Dec 2012, 13:41:31 UTC
Last modified: 8 Dec 2012, 13:42:51 UTC

It´s simple question of numbers...

If with MB only, the graphs shows about 75% average usage of the avaiable bandwith and AP uses aproximately the same band... need to say anything else?

What they realy needs is split the work in two diferent lines (100 MBps each) or... finaly put the 1GBps link to work... The actual configuration with MB/AP running at the same time reaches it´s full capacity a long time ago.

In bussines (ok SETI is not bussines but the rule apply in the same manner), every IT men knows, you could never use an average off more than 3/4 of your total bandwidth or... your will start to loose clients. Besides mantain your clients you need to have the capacity to allow new ones.

Any other solution, if works, will work for a few amount of time, too many new powerfull GPU´s are expected to come to crunch in the next months/years...
____________

Grant (SSSF)
Send message
Joined: 19 Aug 99
Posts: 5868
Credit: 60,618,280
RAC: 47,488
Australia
Message 1313177 - Posted: 9 Dec 2012, 18:08:59 UTC - in response to Message 1312565.


Assimilator backlogs continue to grow.
____________
Grant
Darwin NT.

zoom314Project donor
Avatar
Send message
Joined: 30 Nov 03
Posts: 46519
Credit: 36,865,313
RAC: 5,104
United States
Message 1313242 - Posted: 9 Dec 2012, 19:56:45 UTC - in response to Message 1313177.


Assimilator backlogs continue to grow.

That's probably cause the Borg Drones the project got are experiencing traffic, maybe. ;) Merry XMAS Grant.
____________
My Facebook, War Commander, 2015

Previous · 1 . . . 11 · 12 · 13 · 14 · 15 · 16 · 17 . . . 23 · Next

Message boards : Number crunching : Panic Mode On (79) Server Problems?

Copyright © 2014 University of California