Panic Mode On (78) Server Problems?


log in

Advanced search

Message boards : Number crunching : Panic Mode On (78) Server Problems?

Previous · 1 . . . 18 · 19 · 20 · 21 · 22 · Next
Author Message
Mark Lybeck
Send message
Joined: 9 Aug 99
Posts: 209
Credit: 106,199,864
RAC: 59,695
Finland
Message 1306880 - Posted: 16 Nov 2012, 19:40:55 UTC - in response to Message 1306810.

Hey,

With a proxy you can get higher throughput than without. It is totally different now compared to a direct connection. Previously on congestion at SETI the calculation was faster than Download or upload.


Ok the first Proxy I used had me blocked out. It did not take too many hours for the Proxy to think I was overusing it.

Changed to another one. Seems ok now so far.

____________

Profile S@NL - XP_Freak
Send message
Joined: 10 Jul 99
Posts: 99
Credit: 4,919,397
RAC: 1,395
Netherlands
Message 1306925 - Posted: 16 Nov 2012, 22:09:02 UTC

How do I get rid of the ghost files?

My wife has less then 10 CPU wu's to crunch (0 GPU)
But there are about 1180 ghost WU's on her account, so she doesn't get any new work.

http://setiathome.berkeley.edu/results.php?hostid=5936358


____________

Goodbye Seti Classic

Profile [seti.international] Dirk Sadowski
Volunteer tester
Avatar
Send message
Joined: 6 Apr 07
Posts: 7124
Credit: 61,641,989
RAC: 15,634
Germany
Message 1306930 - Posted: 16 Nov 2012, 22:21:09 UTC - in response to Message 1306925.

S@NL - XP_Freak wrote:
How do I get rid of the ghost files?

My wife has less then 10 CPU wu's to crunch (0 GPU)
But there are about 1180 ghost WU's on her account, so she doesn't get any new work.

http://setiathome.berkeley.edu/results.php?hostid=5936358


If the WUs are in the online overview as *in progress* but not in BOINC, then you would need to do nothing.
BOINC would D/L the WUs, every work request 20 re-sent WUs (if successful scheduler contact).

If the request always time out, maybe you could press *from time to time* *update* in BOINC Manager.
You wouldn't need to wait the 303 seconds between requests.


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

SETI@home Needs your Help ... $10 & U get a Star!

Team seti.international

Das Deutsche Cafe. The German Cafe.

Profile S@NL - XP_Freak
Send message
Joined: 10 Jul 99
Posts: 99
Credit: 4,919,397
RAC: 1,395
Netherlands
Message 1306933 - Posted: 16 Nov 2012, 22:27:05 UTC - in response to Message 1306930.

Thanks for the response.

I see on my own PC I don't get any new work either.
Only 32 left to crunch.


____________

Goodbye Seti Classic

ClaggyProject donor
Volunteer tester
Send message
Joined: 5 Jul 99
Posts: 4249
Credit: 34,988,542
RAC: 21,020
United Kingdom
Message 1306934 - Posted: 16 Nov 2012, 22:27:13 UTC - in response to Message 1306930.

S@NL - XP_Freak wrote:
How do I get rid of the ghost files?

My wife has less then 10 CPU wu's to crunch (0 GPU)
But there are about 1180 ghost WU's on her account, so she doesn't get any new work.

http://setiathome.berkeley.edu/results.php?hostid=5936358


If the WUs are in the online overview as *in progress* but not in BOINC, then you would need to do nothing.
BOINC would D/L the WUs, every work request 20 re-sent WUs (if successful scheduler contact).

If the request always time out, maybe you could press *from time to time* *update* in BOINC Manager.
You wouldn't need to wait the 303 seconds between requests.

Lowering the cache settings so only one or two Wu's are sent at a time will increase the likelihood of a successful scheduler contact,
or he can try a proxy,

Claggy

Profile [seti.international] Dirk Sadowski
Volunteer tester
Avatar
Send message
Joined: 6 Apr 07
Posts: 7124
Credit: 61,641,989
RAC: 15,634
Germany
Message 1306939 - Posted: 16 Nov 2012, 22:42:11 UTC

Just a note because of the PROXY usage ..

You know that you use a PC or server of someone else for to reach S@h then?

If it's a *bad* person, maybe he read all this what you send/receive.

No?

I could imagine that this could happen.

With every scheduler contact you send your *account key* to S@h.

With help of this *account key* this person could log in in your account here then.


Only for advanced members:
At least ..
I would edit the account_setiathome.berkeley.edu.xml file.
In the line <authenticator> I would insert the *weak account key*.


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

SETI@home Needs your Help ... $10 & U get a Star!

Team seti.international

Das Deutsche Cafe. The German Cafe.

ClaggyProject donor
Volunteer tester
Send message
Joined: 5 Jul 99
Posts: 4249
Credit: 34,988,542
RAC: 21,020
United Kingdom
Message 1306952 - Posted: 16 Nov 2012, 23:17:03 UTC
Last modified: 16 Nov 2012, 23:18:55 UTC

I've just had my 200 tasks marked up as abandoned, all at 16 Nov 2012 | 22:28:46 UTC, strange as this was between scheduler contacts:

16/11/2012 22:25:19 SETI@home [sched_op_debug] Starting scheduler request
16/11/2012 22:25:19 SETI@home Sending scheduler request: To fetch work.
16/11/2012 22:25:19 SETI@home Requesting new tasks for CPU and GPU
16/11/2012 22:25:19 SETI@home [sched_op_debug] CPU work request: 228077.80 seconds; 0.00 CPUs
16/11/2012 22:25:19 SETI@home [sched_op_debug] NVIDIA GPU work request: 255253.39 seconds; 0.00 GPUs
16/11/2012 22:25:19 SETI@home [sched_op_debug] ATI GPU work request: 0.00 seconds; 0.00 GPUs
16/11/2012 22:25:30 SETI@home Scheduler request completed: got 0 new tasks
16/11/2012 22:25:30 SETI@home [sched_op_debug] Server version 701
16/11/2012 22:25:30 SETI@home Message from server: No tasks sent
16/11/2012 22:25:30 SETI@home Message from server: No tasks are available for AstroPulse v6
16/11/2012 22:25:30 SETI@home Message from server: Tasks for AMD/ATI GPU are available, but your preferences are set to not accept them
16/11/2012 22:25:30 SETI@home Message from server: This computer has reached a limit on tasks in progress
16/11/2012 22:25:30 SETI@home Message from server: Project has no tasks available
16/11/2012 22:25:30 SETI@home Project requested delay of 303 seconds
16/11/2012 22:25:30 SETI@home [sched_op_debug] Deferring communication for 5 min 3 sec
16/11/2012 22:25:30 SETI@home [sched_op_debug] Reason: requested by project


16/11/2012 22:30:37 SETI@home [sched_op_debug] Starting scheduler request
16/11/2012 22:30:37 SETI@home Sending scheduler request: To fetch work.
16/11/2012 22:30:37 SETI@home Requesting new tasks for CPU and GPU
16/11/2012 22:30:37 SETI@home [sched_op_debug] CPU work request: 222742.75 seconds; 0.00 CPUs
16/11/2012 22:30:37 SETI@home [sched_op_debug] NVIDIA GPU work request: 255610.56 seconds; 0.00 GPUs
16/11/2012 22:30:37 SETI@home [sched_op_debug] ATI GPU work request: 0.00 seconds; 0.00 GPUs
16/11/2012 22:30:59 SETI@home Scheduler request completed: got 0 new tasks
16/11/2012 22:30:59 SETI@home [sched_op_debug] Server version 701
16/11/2012 22:30:59 SETI@home Message from server: Not sending work - last request too recent: 128 sec
16/11/2012 22:30:59 SETI@home Project requested delay of 303 seconds
16/11/2012 22:30:59 SETI@home [sched_op_debug] Deferring communication for 5 min 3 sec
16/11/2012 22:30:59 SETI@home [sched_op_debug] Reason: requested by project

I've reset the project to get rid of all the abandoned work, and now have a fresh 200 tasks to crunch.

Claggy

Grant (SSSF)
Send message
Joined: 19 Aug 99
Posts: 5955
Credit: 62,509,568
RAC: 40,744
Australia
Message 1306955 - Posted: 16 Nov 2012, 23:20:53 UTC - in response to Message 1306952.

I've just had my 200 tasks marked up as abandoned, all at 16 Nov 2012 | 22:28:46 UTC, strange as this was between scheduler contacts:


15 Nov 2012 | 18:09:02 UTC the Scheduler decalred 200 of mine abandoned also, not long after i started using a proxy.
So far it's been just that one instance, but apparently some people have had multiple instances of WUs been declared abandoned.
____________
Grant
Darwin NT.

Profile Tron
Send message
Joined: 16 Aug 09
Posts: 180
Credit: 2,236,055
RAC: 0
United States
Message 1307005 - Posted: 17 Nov 2012, 3:19:59 UTC
Last modified: 17 Nov 2012, 4:18:19 UTC

I have found that when my scheduler requests time out or downloads hang...if I suspend networking for about 20 seconds then re-enable,
I am then connected right away and see average DL speeds of 30kbs ... done it five times now just to be sure I was not self conjuring a pattern.

Edit : I forgot to mention ,You must also dance around your host whilst chanting ancient Aztec blessings. j,k on that last part , the other thing does work for me though.

Grant (SSSF)
Send message
Joined: 19 Aug 99
Posts: 5955
Credit: 62,509,568
RAC: 40,744
Australia
Message 1307018 - Posted: 17 Nov 2012, 5:21:25 UTC - in response to Message 1307005.
Last modified: 17 Nov 2012, 5:42:10 UTC

OK, over the last hour on one of my systems the Scheduler has timed out on all but 2 requests- both of those got resends. Prior to that all work received for the last 24hrs or so had been new work.
EDIT- even requests for work, but not reporting, are timing out.

On the other system, about 50% of the requests over the last hour and a half have timed out.
____________
Grant
Darwin NT.

Grant (SSSF)
Send message
Joined: 19 Aug 99
Posts: 5955
Credit: 62,509,568
RAC: 40,744
Australia
Message 1307019 - Posted: 17 Nov 2012, 5:23:26 UTC - in response to Message 1307005.
Last modified: 17 Nov 2012, 5:41:23 UTC

I have found that when my scheduler requests time out or downloads hang...if I suspend networking for about 20 seconds then re-enable,
I am then connected right away and see average DL speeds of 30kbs ... done it five times now just to be sure I was not self conjuring a pattern.

That's been the case for downloads for several years now, and sometimes with uploads as well. If they stall, or don't actually start even once the timer starts counting then disabling & re-enabling network access usually gets them going again.
I haven't had any luck with the Scheduler timeouts though.
____________
Grant
Darwin NT.

Mark Lybeck
Send message
Joined: 9 Aug 99
Posts: 209
Credit: 106,199,864
RAC: 59,695
Finland
Message 1307029 - Posted: 17 Nov 2012, 6:19:12 UTC - in response to Message 1306939.

Just a note because of the PROXY usage ..

You know that you use a PC or server of someone else for to reach S@h then?

If it's a *bad* person, maybe he read all this what you send/receive.

No?

I could imagine that this could happen.

With every scheduler contact you send your *account key* to S@h.

With help of this *account key* this person could log in in your account here then.


Only for advanced members:
At least ..
I would edit the account_setiathome.berkeley.edu.xml file.
In the line <authenticator> I would insert the *weak account key*.




How would the weak key help security? What is the overall sequence of messages here with the account key?
____________

Grant (SSSF)
Send message
Joined: 19 Aug 99
Posts: 5955
Credit: 62,509,568
RAC: 40,744
Australia
Message 1307046 - Posted: 17 Nov 2012, 8:27:43 UTC - in response to Message 1307029.


Getting more & more Scheduler timeouts, then eventually it responds & resends the lost tasks.
____________
Grant
Darwin NT.

Profile Brother Frank
Send message
Joined: 10 Dec 11
Posts: 26
Credit: 15,142,410
RAC: 0
United States
Message 1307053 - Posted: 17 Nov 2012, 9:15:36 UTC - in response to Message 1307046.
Last modified: 17 Nov 2012, 9:16:42 UTC

My desktops and newer notebooks with Nvidia graphics cards are getting work now after changing them all over to proxy servers. My little laptops with integrated graphics are not getting new work anymore and timing out. My i3 notebook is timing out, told there is no work, and has reached it's limits for jobs in progress. Lots of ghosts. The integrated graphic card notebooks started to have big issues with downloading work today and are fast running out of work. I'll try to keep my desktops and new notebook going, but I am not going to intervene anymore manually with the upload and download problems with my slightly older notebooks.

I hope Seti at Home can work these networking/scheduler and ghost issues out. Am switching the 2 and 3 year old notebooks over to other BOINC projects like world community grid and a cosmology project or two. These issues have to be affecting a huge number of volunteers, as in hundreds of thousands. Many of us are making hundreds and thousands of dollars of in-kind donations. Very important and critical good will is being lost here by letting these upload and download issues to remain unaddressed and low priority. I hope project leaders understand how large an impact this loss of good will is going to be. Please listen to what many of us are saying here. We are among your most dedicated.

Brother Frank

Profile Link
Avatar
Send message
Joined: 18 Sep 03
Posts: 841
Credit: 1,578,326
RAC: 52
Germany
Message 1307057 - Posted: 17 Nov 2012, 9:32:19 UTC - in response to Message 1307029.

How would the weak key help security? What is the overall sequence of messages here with the account key?

See here. With the normal authenticator you can log into your account and steal it by changing the email adress and password, with the weak one you can just crunch for this account. It's not a big issue since criminals probably have better things to do than stealing other people's BOINC accounts, but why not improve the security if there's an easy way?

Are there some disadvantages when using the weak account key for your computers?
____________
.

Profile Gary CharpentierProject donor
Volunteer tester
Avatar
Send message
Joined: 25 Dec 00
Posts: 13201
Credit: 7,980,007
RAC: 16,206
United States
Message 1307102 - Posted: 17 Nov 2012, 14:58:54 UTC - in response to Message 1307071.

Not much to say except the kitties are now starting to do Einstein as their GPUs run dry of Seti work until this mess is fixed.

Not a happy time for the kittyman.

And Einstein is in a planned outage ...

____________

juan BFBProject donor
Volunteer tester
Avatar
Send message
Joined: 16 Mar 07
Posts: 5489
Credit: 316,514,007
RAC: 133,716
Brazil
Message 1307105 - Posted: 17 Nov 2012, 15:02:21 UTC
Last modified: 17 Nov 2012, 15:03:57 UTC

Did anybody else see the validating pendings are rising to an very high value?

Mine are at (18804) and rissing > 100/hr.
____________

Richard HaselgroveProject donor
Volunteer tester
Send message
Joined: 4 Jul 99
Posts: 8830
Credit: 53,651,244
RAC: 49,135
United Kingdom
Message 1307111 - Posted: 17 Nov 2012, 15:51:08 UTC - in response to Message 1307102.

Not much to say except the kitties are now starting to do Einstein as their GPUs run dry of Seti work until this mess is fixed.

Not a happy time for the kittyman.

And Einstein is in a planned outage ...

Due to end 2 PM Milwaukee time (CST, UTC-6), if anybody's wondering.

http://www4.uwm.edu/technology/itstatus/index.cfm?a1=details&item_id=17216

ClaggyProject donor
Volunteer tester
Send message
Joined: 5 Jul 99
Posts: 4249
Credit: 34,988,542
RAC: 21,020
United Kingdom
Message 1307112 - Posted: 17 Nov 2012, 15:56:27 UTC - in response to Message 1307111.

Not much to say except the kitties are now starting to do Einstein as their GPUs run dry of Seti work until this mess is fixed.

Not a happy time for the kittyman.

And Einstein is in a planned outage ...

Due to end 2 PM Milwaukee time (CST, UTC-6), if anybody's wondering.

http://www4.uwm.edu/technology/itstatus/index.cfm?a1=details&item_id=17216

They were planning to keep the core servers running on UPS, but it's failed, so they've had to shut them down.

http://boinc.berkeley.edu/dev/forum_thread.php?id=7102&nowrap=true#46333

Claggy

Previous · 1 . . . 18 · 19 · 20 · 21 · 22 · Next

Message boards : Number crunching : Panic Mode On (78) Server Problems?

Copyright © 2014 University of California