Panic Mode On (74) Server problems?


log in

Advanced search

Message boards : Number crunching : Panic Mode On (74) Server problems?

Previous · 1 . . . 3 · 4 · 5 · 6 · 7 · 8 · 9 . . . 11 · Next
Author Message
Cosmic_Ocean
Avatar
Send message
Joined: 23 Dec 00
Posts: 2830
Credit: 10,577,308
RAC: 1,025
United States
Message 1225373 - Posted: 30 Apr 2012, 17:51:58 UTC
Last modified: 30 Apr 2012, 18:00:25 UTC

I have my max transfers set for 5, though I hardly ever get more than 1-2 tasks at a time anyway.

About the only thing you can do to get the work that you have been assigned is to either deal with the back-offs, or use a build that isn't so bad with the back-offs. 6.10.58/60 gets recommended frequently.

Personally, since I don't need/use GPUs, I use 6.2.19. Back-offs are per transfer and not global. If there is one transfer in back-off, scheduler request is postponed until at least one transfer starts flowing, and the maximum back-off is 3:59:59.

Just throwing options out there.

[edit] Speaking of the download pipe.. just got an AP and it is downloading at 175KB/sec. [/edit]
____________
Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)

Grant (SSSF)
Volunteer tester
Send message
Joined: 19 Aug 99
Posts: 7294
Credit: 86,505,334
RAC: 41,646
Australia
Message 1225396 - Posted: 30 Apr 2012, 18:41:17 UTC - in response to Message 1225373.


Well there was a burst of network traffic for a while thre, but untill the validators & assimilators clear the backlog i don't think we're going to see much work.
4 hours ago was the last time i managed to get some work- 2 WUs.
____________
Grant
Darwin NT.

Profile Area 51
Avatar
Send message
Joined: 31 Jan 04
Posts: 965
Credit: 42,193,520
RAC: 0
United Kingdom
Message 1225407 - Posted: 30 Apr 2012, 19:02:26 UTC - in response to Message 1225274.


Or perhaps has anyone a tip for getting a more reliable DL of WU's?

Thanks
Mowo

[BOINC] > cat cc_config.xml <cc_config> <log_flags> </log_flags> <options> <max_file_xfers>50</max_file_xfers> <max_file_xfers_per_project>50</max_file_xfers_per_project> </options> </cc_config>

You'll be buying them 25 extra download servers, then? :P

Seriously, this is a community, distributed project, with tens of thousands of users jostling for a scarce resource. I would prefer to see solutions based on collaboration, rather than competition and exclusion.


I have to agree - and looking at how many hosts this person is responsible for, I sure hope that they don't all have that sort of setting. Good grief - talk about selfishness.
____________

tbret
Volunteer tester
Avatar
Send message
Joined: 28 May 99
Posts: 3364
Credit: 246,136,707
RAC: 270,253
United States
Message 1225414 - Posted: 30 Apr 2012, 19:25:18 UTC - in response to Message 1224356.
Last modified: 30 Apr 2012, 19:26:13 UTC

Cliff,

You can't go. That's part of the contract you signed before you downloaded your first WU.

I'm thinking, "laptop." I'm thinking, "some way to recharge a big 12v battery."

David SProject Donor
Volunteer tester
Avatar
Send message
Joined: 4 Oct 99
Posts: 16450
Credit: 20,270,131
RAC: 8,324
United States
Message 1226359 - Posted: 2 May 2012, 21:07:10 UTC

In 20 minutes, it will have been a whole 48 hours since the last post to this thread. Other people have started threads about problems they're having, but they don't seem to be general problems affecting everyone. The crickets are maxed, but things seem to be going smoothly.

What is this world coming to?

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


Profile Gary CharpentierCrowdfunding Project Donor
Volunteer tester
Avatar
Send message
Joined: 25 Dec 00
Posts: 17838
Credit: 19,443,318
RAC: 14,549
United States
Message 1226372 - Posted: 2 May 2012, 21:24:57 UTC - in response to Message 1226359.

In 20 minutes, it will have been a whole 48 hours since the last post to this thread. Other people have started threads about problems they're having, but they don't seem to be general problems affecting everyone. The crickets are maxed, but things seem to be going smoothly.

What is this world coming to?

Give it a couple days to the weekend.

____________

Wembley
Volunteer tester
Avatar
Send message
Joined: 16 Sep 09
Posts: 429
Credit: 1,509,143
RAC: 1,544
United States
Message 1226436 - Posted: 2 May 2012, 23:21:59 UTC - in response to Message 1226359.

In 20 minutes, it will have been a whole 48 hours since the last post to this thread. Other people have started threads about problems they're having, but they don't seem to be general problems affecting everyone. The crickets are maxed, but things seem to be going smoothly.

What is this world coming to?

As soon as the last person leaves the office on Friday evening, the gremlins will come out.

____________

Profile ivan
Volunteer tester
Avatar
Send message
Joined: 5 Mar 01
Posts: 780
Credit: 223,795,074
RAC: 106,381
United Kingdom
Message 1226800 - Posted: 3 May 2012, 19:38:38 UTC - in response to Message 1226436.
Last modified: 3 May 2012, 19:39:12 UTC

In 20 minutes, it will have been a whole 48 hours since the last post to this thread. Other people have started threads about problems they're having, but they don't seem to be general problems affecting everyone. The crickets are maxed, but things seem to be going smoothly.

What is this world coming to?

As soon as the last person leaves the office on Friday evening, the gremlins will come out.


snafu -- which the dictionary I had as a schoolkid translated as "Situation Normal, All Fouled (euphemism) Up"! When I was an instrument scientist at a large physics lab in Switzerland it was almost inevitable that something would fail after 1800 on Friday, while I was still there but after the more office-hours-minded IT and engineering staff had left.
____________

Grant (SSSF)
Volunteer tester
Send message
Joined: 19 Aug 99
Posts: 7294
Credit: 86,505,334
RAC: 41,646
Australia
Message 1227612 - Posted: 5 May 2012, 4:51:46 UTC
Last modified: 5 May 2012, 4:52:06 UTC

Not a good sign. Ready to Send continues to fall, but the splitters haven't cranked up the pace. Few more hours & work will be very scarce.
____________
Grant
Darwin NT.

Profile James SotherdenProject Donor
Avatar
Send message
Joined: 16 May 99
Posts: 10090
Credit: 62,176,569
RAC: 45,487
United States
Message 1227613 - Posted: 5 May 2012, 4:58:33 UTC

Im having problems with downloads not wanting to download.
____________
[/quote]

Old James

zoom314
Volunteer tester
Avatar
Send message
Joined: 30 Nov 03
Posts: 55959
Credit: 40,371,343
RAC: 3,659
United States
Message 1227622 - Posted: 5 May 2012, 5:20:40 UTC - in response to Message 1227613.

Im having problems with downloads not wanting to download.

Me too and currently I have 3 that are either in slow mode or are simply stuck.
____________
Pluto is still a planet.

Beep! Beep!

Profile Area 51
Avatar
Send message
Joined: 31 Jan 04
Posts: 965
Credit: 42,193,520
RAC: 0
United Kingdom
Message 1228425 - Posted: 6 May 2012, 20:26:44 UTC

No issues for me - until now. The cricket's got an unhealthy gap forming.....
____________

Grant (SSSF)
Volunteer tester
Send message
Joined: 19 Aug 99
Posts: 7294
Credit: 86,505,334
RAC: 41,646
Australia
Message 1228445 - Posted: 6 May 2012, 21:15:36 UTC - in response to Message 1228425.
Last modified: 6 May 2012, 21:57:16 UTC

Network traffic has dropped off, Ready to Send still has plenty to go, the last dozen or so requests for work have got "Project has no tasks available" messages.
There was a spike in database activity for a while there- almost 3,000 queries per second, usually it's aronound 600. Hopefully once it settles down again we'll be able to get more work.
____________
Grant
Darwin NT.

Profile Area 51
Avatar
Send message
Joined: 31 Jan 04
Posts: 965
Credit: 42,193,520
RAC: 0
United Kingdom
Message 1228458 - Posted: 6 May 2012, 21:47:00 UTC

....false alarm - back up to full tilt.
____________

Grant (SSSF)
Volunteer tester
Send message
Joined: 19 Aug 99
Posts: 7294
Credit: 86,505,334
RAC: 41,646
Australia
Message 1229469 - Posted: 9 May 2012, 9:25:24 UTC - in response to Message 1220675.


Hmm, there's a blockage somewhere.
Plenty of work available, but network traffic isn't maxed out. After an outage, and with the limits raised further (or even removed) it should remain maxed out probably until the next outage.
____________
Grant
Darwin NT.

LadyL
Volunteer tester
Avatar
Send message
Joined: 14 Sep 11
Posts: 1679
Credit: 5,230,097
RAC: 0
Message 1229475 - Posted: 9 May 2012, 9:51:14 UTC - in response to Message 1229469.


Hmm, there's a blockage somewhere.
Plenty of work available, but network traffic isn't maxed out. After an outage, and with the limits raised further (or even removed) it should remain maxed out probably until the next outage.

My last bunch came down without problems and so fast I went to see why...

Server Status Page shows no AP are being split. AP account for quite a bit of the bandwidth as we've seen in the past when there were none. Good news for filling the caches all round.
____________
I'm not the Pope. I don't speak Ex Cathedra!

Richard HaselgroveProject Donor
Volunteer tester
Send message
Joined: 4 Jul 99
Posts: 10831
Credit: 79,658,771
RAC: 76,142
United Kingdom
Message 1229476 - Posted: 9 May 2012, 9:57:37 UTC - in response to Message 1229475.


Hmm, there's a blockage somewhere.
Plenty of work available, but network traffic isn't maxed out. After an outage, and with the limits raised further (or even removed) it should remain maxed out probably until the next outage.

My last bunch came down without problems and so fast I went to see why...

Server Status Page shows no AP are being split. AP account for quite a bit of the bandwidth as we've seen in the past when there were none. Good news for filling the caches all round.

09/05/2012 10:46:59 | SETI@home | Scheduler request completed: got 36 new tasks

And not a single shorty among them.

Grant (SSSF)
Volunteer tester
Send message
Joined: 19 Aug 99
Posts: 7294
Credit: 86,505,334
RAC: 41,646
Australia
Message 1229478 - Posted: 9 May 2012, 10:04:58 UTC - in response to Message 1229475.

Server Status Page shows no AP are being split. AP account for quite a bit of the bandwidth as we've seen in the past when there were none. Good news for filling the caches all round.

True, but considering the network traffic before the outage & the raising of the limits, i wouldn't expect the network traffic to drop off for at least 3 days.
And when network traffic isn't at maximum resuests for work rarely result in "Project has no tasks available" messages, but 80% of my scheduler requests are resulting in that message indcates something isn't quite right.
____________
Grant
Darwin NT.

Previous · 1 . . . 3 · 4 · 5 · 6 · 7 · 8 · 9 . . . 11 · Next

Message boards : Number crunching : Panic Mode On (74) Server problems?

Copyright © 2016 University of California