Panic Mode On (82) Server Problems?

Message boards : Number crunching : Panic Mode On (82) Server Problems?
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 9 · 10 · 11 · 12 · 13 · 14 · 15 . . . 24 · Next

AuthorMessage
natolars
Avatar

Send message
Joined: 27 Jul 00
Posts: 6
Credit: 11,345,260
RAC: 0
Norway
Message 1346727 - Posted: 15 Mar 2013, 1:32:33 UTC

I ran TCP Optimizer...

I disabled TCP Auto-Tuning in Win 8... So, everything should be "optimized".

Still, DL speed is painfully slow, stalls, backs off, transient HTTP errors; you know the story...

AP units can take more than a day to download.... So my GTX 690 is actually spending more time idle than working...

But I'm trying to look on the bright side of life. I'm not starving, and Cherry Coke is here (Norway) to stay!


All those moments will be lost in time... Like tears in rain...
ID: 1346727 · Report as offensive
Profile Wiggo
Avatar

Send message
Joined: 24 Jan 00
Posts: 34744
Credit: 261,360,520
RAC: 489
Australia
Message 1346735 - Posted: 15 Mar 2013, 2:53:43 UTC - in response to Message 1346727.  

I ran TCP Optimizer...

I disabled TCP Auto-Tuning in Win 8... So, everything should be "optimized".

Still, DL speed is painfully slow, stalls, backs off, transient HTTP errors; you know the story...

AP units can take more than a day to download.... So my GTX 690 is actually spending more time idle than working...

But I'm trying to look on the bright side of life. I'm not starving, and Cherry Coke is here (Norway) to stay!


Have you restarted it and checked to make sure that the settings took?

Cheers.
ID: 1346735 · Report as offensive
natolars
Avatar

Send message
Joined: 27 Jul 00
Posts: 6
Credit: 11,345,260
RAC: 0
Norway
Message 1346744 - Posted: 15 Mar 2013, 4:22:56 UTC - in response to Message 1346735.  
Last modified: 15 Mar 2013, 4:23:36 UTC

I ran TCP Optimizer...

I disabled TCP Auto-Tuning in Win 8... So, everything should be "optimized".

Still, DL speed is painfully slow, stalls, backs off, transient HTTP errors; you know the story...

AP units can take more than a day to download.... So my GTX 690 is actually spending more time idle than working...

But I'm trying to look on the bright side of life. I'm not starving, and Cherry Coke is here (Norway) to stay!


Have you restarted it and checked to make sure that the settings took?

Cheers.


I've tried different settings/tweaks and restarted every time, but with no luck. I've also tried different firewall settings, but no luck there either.

A question: I've chosen maintaining enough tasks for 5 days, and up to 2 additional days. I never get tasks for anything near 5 days. Also tried 5 days, up to 0.1 days. It didn't make any difference. Does anyone know why I never get as many tasks as I've opted for?

If relevant; here are my specs:

Intel(R) Core(TM) i7-3770K CPU @ 3.50 GHz
16GB RAM
250GB SSD HDD
1TB SATA 3.0 HDD (Data are stored on this drive, with 100 GB available to BOINC)
Nvidia GeForce GTX 690 4GB PhysX CUDA (Driver 314.07; CUDA works fine)
Windows 8 Professional x64 Edition
Internet bandwidth: 10 Mbit

Obviously, I don't have the computer knowledge as many on this forum, so I apologise in advance for any ignorant questions. ;-) Any specific config file suggestions for my system?

Cheers.
All those moments will be lost in time... Like tears in rain...
ID: 1346744 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13715
Credit: 208,696,464
RAC: 304
Australia
Message 1346776 - Posted: 15 Mar 2013, 6:40:52 UTC - in response to Message 1346744.  
Last modified: 15 Mar 2013, 6:45:34 UTC

Any specific config file suggestions for my system?

Wait for whatever is wrong with Seti to be sorted out.
I got back from work to find downloads queued up & project backoffs in force.
Hit Retry now several times, and after about 2 minutes some of the WUs started to download- at less than .5kB/s. They had been downloading at 4-5kB/s.
Something, somewhere has jammed up in the network pipe, even more so than at any time in the past.


EDIT-
Just had a look at the graphs- AP ready to send is climbing like a rocket- so that will keep the network tied up niceley. Database queries numbers haven't updateed for hours- spomthing there is borked. There are more MB splitters running than there were, but still not enough to meet demand when the longer running WUs become scarce again.
It's not looking like a good weekend at this stage.
Grant
Darwin NT
ID: 1346776 · Report as offensive
Profile Bernie Vine
Volunteer moderator
Volunteer tester
Avatar

Send message
Joined: 26 May 99
Posts: 9954
Credit: 103,452,613
RAC: 328
United Kingdom
Message 1346794 - Posted: 15 Mar 2013, 8:37:54 UTC

Has about 30 downloads in back-off this morning, hit retry and they all downloaded without pause at same sort of speeds as before.

Also a machine I had set for NNT on Einstein I switched to SETI-Beta and it downloaded 45 tasks with no pause.

So it seems even with the "fix" some people have more problems that others!!
ID: 1346794 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13715
Credit: 208,696,464
RAC: 304
Australia
Message 1346800 - Posted: 15 Mar 2013, 9:30:02 UTC - in response to Message 1346794.  


Well, after getting them to download & the next few eventually downloading by themselves, they're all starting to backup again. Retries, disable & re-enable network access & the TCP tweak all having no effect.
It's borked.
Grant
Darwin NT
ID: 1346800 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14644
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1346801 - Posted: 15 Mar 2013, 9:42:51 UTC - in response to Message 1346800.  


Well, after getting them to download & the next few eventually downloading by themselves, they're all starting to backup again. Retries, disable & re-enable network access & the TCP tweak all having no effect.
It's borked.

It might be worth checking which IP address is being contacted, with http_debug. I'm seeing (and cricket is showing) data flowing normally some of the time, but other downloads stalling. Sounds like one broken download server out of two?
ID: 1346801 · Report as offensive
Profile Wiggo
Avatar

Send message
Joined: 24 Jan 00
Posts: 34744
Credit: 261,360,520
RAC: 489
Australia
Message 1346804 - Posted: 15 Mar 2013, 9:51:11 UTC - in response to Message 1346794.  

While AP's are coming out in such an uncontrolled fashion we'll just have to put up with the pipeline being overwhelmed (hopefully they will not load any new files for the AP splitters to feast on until the MB splitters catch up with those all ready loaded).

But my earlier worries in the day here so far have not happened with the upgrade that I did and both my rigs here have been on the limits all day.

Cheers.
ID: 1346804 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13715
Credit: 208,696,464
RAC: 304
Australia
Message 1346806 - Posted: 15 Mar 2013, 9:59:21 UTC - in response to Message 1346801.  
Last modified: 15 Mar 2013, 10:02:32 UTC


Well, after getting them to download & the next few eventually downloading by themselves, they're all starting to backup again. Retries, disable & re-enable network access & the TCP tweak all having no effect.
It's borked.

It might be worth checking which IP address is being contacted, with http_debug. I'm seeing (and cricket is showing) data flowing normally some of the time, but other downloads stalling. Sounds like one broken download server out of two?

I've had my hosts file set to use just .13 for a while now.


EDIT- since my earlier post all those downloads that had built up have cleared themselves. Downloads are still really, really crappy (1-3kB/s where they had been around 5kB/s before) but they are eventually downloading again- at the moment.
Grant
Darwin NT
ID: 1346806 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13715
Credit: 208,696,464
RAC: 304
Australia
Message 1346807 - Posted: 15 Mar 2013, 10:04:10 UTC - in response to Message 1346801.  

Sounds like one broken download server out of two?

Or maybe the load sharing is working better now, so the good server is occasionally being overwhelmed on those occasions it's been given more of the load?

Grant
Darwin NT
ID: 1346807 · Report as offensive
Bruce
Volunteer tester

Send message
Joined: 15 Mar 02
Posts: 123
Credit: 124,955,234
RAC: 11
United States
Message 1346866 - Posted: 15 Mar 2013, 15:10:41 UTC - in response to Message 1346744.  

I ran TCP Optimizer...

I disabled TCP Auto-Tuning in Win 8... So, everything should be "optimized".

Still, DL speed is painfully slow, stalls, backs off, transient HTTP errors; you know the story...

AP units can take more than a day to download.... So my GTX 690 is actually spending more time idle than working...

But I'm trying to look on the bright side of life. I'm not starving, and Cherry Coke is here (Norway) to stay!


Have you restarted it and checked to make sure that the settings took?

Cheers.


I've tried different settings/tweaks and restarted every time, but with no luck. I've also tried different firewall settings, but no luck there either.

A question: I've chosen maintaining enough tasks for 5 days, and up to 2 additional days. I never get tasks for anything near 5 days. Also tried 5 days, up to 0.1 days. It didn't make any difference. Does anyone know why I never get as many tasks as I've opted for?

If relevant; here are my specs:

Intel(R) Core(TM) i7-3770K CPU @ 3.50 GHz
16GB RAM
250GB SSD HDD
1TB SATA 3.0 HDD (Data are stored on this drive, with 100 GB available to BOINC)
Nvidia GeForce GTX 690 4GB PhysX CUDA (Driver 314.07; CUDA works fine)
Windows 8 Professional x64 Edition
Internet bandwidth: 10 Mbit

Obviously, I don't have the computer knowledge as many on this forum, so I apologise in advance for any ignorant questions. ;-) Any specific config file suggestions for my system?

Cheers.


TCP Optimizer 3.0.8, which I believe is the latest, is not compatible with Win8. They are working on one that will work with Win8, but it might be awhile. This means that you will have to open Regedit and make the TCP1323 Opts change manually by changing the DWORD to a value of 3, save and reboot. Read this message http://setiathome.berkeley.edu/forum_thread.php?id=71002&postid=1344159 to find the actual location. Hope this helps a little.

As for your cache, there are currently limits in place, 100 work units for the CPU and 100 work units for the GPU.

Bruce
Bruce
ID: 1346866 · Report as offensive
Profile Donald L. Johnson
Avatar

Send message
Joined: 5 Aug 02
Posts: 8240
Credit: 14,654,533
RAC: 20
United States
Message 1346874 - Posted: 15 Mar 2013, 15:23:05 UTC - in response to Message 1346866.  

As for your cache, there are currently limits in place, 100 work units for the CPU and 100 work units for the GPU.

Bruce

He's only got about 40 in progress on each of his machines, mostly cpu tasks. He's not even close to the limits.
Donald
Infernal Optimist / Submariner, retired
ID: 1346874 · 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 1346875 - Posted: 15 Mar 2013, 15:27:37 UTC - in response to Message 1346874.  
Last modified: 15 Mar 2013, 15:28:58 UTC

As for your cache, there are currently limits in place, 100 work units for the CPU and 100 work units for the GPU.

Bruce

He's only got about 40 in progress on each of his machines, mostly cpu tasks. He's not even close to the limits.

There is nothing server side right now that should prevent anybody from getting what they need.
Even with all difficulties involved, I have 9 rigs online, with 1,796 WUs cached out of a possible 1,800.
The TCP opt is important, as it has worked to allow download of the big AP tasks.
Those will kill workfetch if they are eternally in backoff status.

Otherwise, I suspect there are other problems in play.

Don't blame the servers just now.
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1346875 · Report as offensive
juan BFP Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 16 Mar 07
Posts: 9786
Credit: 572,710,851
RAC: 3,799
Panama
Message 1346974 - Posted: 15 Mar 2013, 19:09:47 UTC

Can´t get any work today, TCP fix still on.

15/03/2013 15:51:01 | SETI@home | Scheduler request failed: Couldn't connect to server
15/03/2013 15:51:15 | | Project communication failed: attempting access to reference site
15/03/2013 15:51:18 | | Internet access OK - project servers may be temporarily down.

ID: 1346974 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13715
Credit: 208,696,464
RAC: 304
Australia
Message 1347079 - Posted: 15 Mar 2013, 22:29:49 UTC - in response to Message 1346974.  
Last modified: 15 Mar 2013, 22:30:16 UTC

Can´t get any work today, TCP fix still on.

15/03/2013 15:51:01 | SETI@home | Scheduler request failed: Couldn't connect to server

I had 2 of those errors in a row a day or 2 ago, 3rd attempt no problem & been OK since as far as i can see.
Grant
Darwin NT
ID: 1347079 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1347955 - Posted: 18 Mar 2013, 2:09:19 UTC

Are we still Panicking over a Shorty Storm? Because if we are, now might be a good time...
ID: 1347955 · Report as offensive
Lionel

Send message
Joined: 25 Mar 00
Posts: 680
Credit: 563,640,304
RAC: 597
Australia
Message 1347969 - Posted: 18 Mar 2013, 4:22:14 UTC - in response to Message 1347955.  

There was a shower this morning (my time - Melbourne Australia) but it backed off a bit during the day ... most of the wus now are 2:31 as opposed to 1:51 earlier ... I am seeing quite a few wus come through though and having implemented the TCP tweak I really haven't had a problem across any box ... However, I am deliberately driving myself into shaping to see what the impact will be with the tweak in place ...




ID: 1347969 · Report as offensive
Profile Akio
Avatar

Send message
Joined: 18 May 11
Posts: 375
Credit: 32,129,242
RAC: 0
United States
Message 1348317 - Posted: 19 Mar 2013, 0:37:47 UTC

Slow going with the downloads the last couple of days. Tons of work, just having lots of difficulty getting them fully downloaded.
ID: 1348317 · Report as offensive
Lionel

Send message
Joined: 25 Mar 00
Posts: 680
Credit: 563,640,304
RAC: 597
Australia
Message 1348319 - Posted: 19 Mar 2013, 0:52:33 UTC - in response to Message 1348317.  

Have you implemented the TCP tweak?
ID: 1348319 · Report as offensive
Profile Akio
Avatar

Send message
Joined: 18 May 11
Posts: 375
Credit: 32,129,242
RAC: 0
United States
Message 1348321 - Posted: 19 Mar 2013, 1:05:31 UTC - in response to Message 1348319.  

Have you implemented the TCP tweak?


Negative. The explanations I've read here on the forums gave me a headache trying to understand them. I gave up.
ID: 1348321 · Report as offensive
Previous · 1 . . . 9 · 10 · 11 · 12 · 13 · 14 · 15 . . . 24 · Next

Message boards : Number crunching : Panic Mode On (82) 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.