Well Well (Oct 29 2009)


log in

Advanced search

Message boards : Technical News : Well Well (Oct 29 2009)

Author Message
Profile Matt Lebofsky
Volunteer moderator
Project administrator
Project developer
Project scientist
Avatar
Send message
Joined: 1 Mar 99
Posts: 1388
Credit: 74,079
RAC: 0
United States
Message 943690 - Posted: 29 Oct 2009, 21:35:05 UTC
Last modified: 29 Oct 2009, 21:35:16 UTC

As predicted the data well temporarily ran dry overnight, but I'm trying my best to keep up with demand today (and set it up for over the weekend).

Weird thing today - I've been noticing intermittent problems connecting to the science database to make the most trivial queries. We thought this, and the assimilator queue backing up, were probably due to Bob's recent configuration changes to the informix database engine perhaps not helping so much. But then I noticed one of assimilators was inserting thousands and thousands of signals as fast as it possibly could from a single result file... since 7:40am yesterday morning!

This is not normal. Result files usually contain a handful of signals, maybe a few dozen tops. If they reach 30K in size they are automatically "cut off" and sent back to us. I tracked down the result file with all the signals - it was 1.6 gigabytes in size! Not sure how this happened, nor how it passed validation (though I have my theories), but it sure contained a lot of signals repeated over and over and over again. I moved that out of the way and hopefully that'll improve performance in general around here.

- Matt
____________
-- BOINC/SETI@home network/web/science/development person
-- "Any idiot can have a good idea. What is hard is to do it." - Jeanne-Claude

Profile [seti.international] Dirk Sadowski
Volunteer tester
Avatar
Send message
Joined: 6 Apr 07
Posts: 7016
Credit: 59,117,698
RAC: 20,600
Germany
Message 943697 - Posted: 29 Oct 2009, 22:02:21 UTC


You remember the movie 'Contact'?

Hey, maybe we got now an instructions for to make a machine for to visit extraterrestrial intelligence?

;-D


BTW.
Thanks for the daily TNews.

____________
BR



>Das Deutsche Cafe. The German Cafe.<

Profile 52 Aces
Avatar
Send message
Joined: 7 Jan 02
Posts: 497
Credit: 12,866,706
RAC: 6,888
United States
Message 943699 - Posted: 29 Oct 2009, 22:03:18 UTC - in response to Message 943690.
Last modified: 29 Oct 2009, 22:08:33 UTC

Someone !AND! their wingman sent back a 1.6 gig result file? wow.

(Not that you can fit 1.6 gig into either a 366k WU or 8meg AP, at least not without a loop bug)

Profile [seti.international] Dirk Sadowski
Volunteer tester
Avatar
Send message
Joined: 6 Apr 07
Posts: 7016
Credit: 59,117,698
RAC: 20,600
Germany
Message 943700 - Posted: 29 Oct 2009, 22:08:00 UTC


For the ones who missed the movie..

[http://en.wikipedia.org/wiki/Contact_(film)]

:-)

____________
BR



>Das Deutsche Cafe. The German Cafe.<

Josef W. Segur
Volunteer developer
Volunteer tester
Send message
Joined: 30 Oct 99
Posts: 4195
Credit: 1,028,567
RAC: 285
United States
Message 943804 - Posted: 30 Oct 2009, 4:13:58 UTC - in response to Message 943699.

Someone !AND! their wingman sent back a 1.6 gig result file? wow.

(Not that you can fit 1.6 gig into either a 366k WU or 8meg AP, at least not without a loop bug)

The way the data is analyzed involves going through it many times at different chirp rates and FFT lengths, or dedispersions for AP, so it is definitely possible a task could generate billions of signals. But as Matt noted the signals were duplicates, and that also explains why the result could validate against a normal result since the validator simply checks whether there's a matching signal in the wingmate result. IOW, the validator doesn't do sanity checks, simply signal comparison.

How that many duplicates were generated without the overflow limit of 31 for MB or 60 for AP being tripped remains a puzzle, as is the failure of BOINC to refuse to upload the file since it exceeded both the 64 KB Enhanced size limit and the 640 KB AP size limit. It almost seems like the file system looped while writing a signal to the outfile, rather than any problem with BOINC or the S@H application.
Joe

Profile ML1
Volunteer tester
Send message
Joined: 25 Nov 01
Posts: 8250
Credit: 4,070,696
RAC: 516
United Kingdom
Message 943834 - Posted: 30 Oct 2009, 12:08:38 UTC - in response to Message 943699.

Someone !AND! their wingman sent back a 1.6 gig result file? wow. ...

I very much doubt that.

Much more likely is that locally a s@h backend spasm magically made the 1.6GB 'appear'.


Time to run a fsck?...

Regards,
Martin

____________
See new freedom: Mageia4
Linux Voice See & try out your OS Freedom!
The Future is what We make IT (GPLv3)

DJStarfox
Send message
Joined: 23 May 01
Posts: 1040
Credit: 532,447
RAC: 19
United States
Message 943886 - Posted: 30 Oct 2009, 17:14:44 UTC - in response to Message 943697.

That would be really funny if the instructions for a machine to visit ET were in that WU. Do you have a link to this WU in question?

Yes, time to run fsck as a preventative measure.

Jim Herod
Volunteer tester
Send message
Joined: 23 Apr 99
Posts: 18
Credit: 3,410,981
RAC: 3,466
United States
Message 943962 - Posted: 30 Oct 2009, 23:01:43 UTC

Change that 1.6GB file's extension to .pdf and see if Acrobat Reader will open it. Should be about the right size for a set of blueprints.
____________

zpm
Volunteer tester
Avatar
Send message
Joined: 25 Apr 08
Posts: 284
Credit: 1,458,391
RAC: 3,292
United States
Message 943988 - Posted: 30 Oct 2009, 23:47:14 UTC - in response to Message 943962.

Change that 1.6GB file's extension to .pdf and see if Acrobat Reader will open it. Should be about the right size for a set of blueprints.


or make it a powerpoint. instructions to the Enterprise-E... LOL!

____________

I recommend Secunia PSI: http://secunia.com/vulnerability_scanning/personal/
Go Georgia Tech.

CryptokiD
Avatar
Send message
Joined: 2 Dec 00
Posts: 134
Credit: 2,814,936
RAC: 0
United States
Message 943997 - Posted: 31 Oct 2009, 0:01:10 UTC

thanks for the update matt. i am glad i decided to queue up a few thousand workunits. i had a feeling we would be running out before halloween.

is there going to be much work units left? im kind of wondering how long it's going to be that we are super low on seti work. thanks !

Chris
Send message
Joined: 7 Oct 09
Posts: 1
Credit: 21,269
RAC: 0
United States
Message 944037 - Posted: 31 Oct 2009, 1:35:47 UTC - in response to Message 943690.

Matt-

This is not the result of a pattern is it?

Profile T. Bell
Send message
Joined: 27 Mar 06
Posts: 20
Credit: 229,303
RAC: 0
Australia
Message 944121 - Posted: 31 Oct 2009, 4:43:38 UTC

Hi,

Only just managed to get a new task from SETI now. It's been coming up empty for the last 12 or so hours. Least now I know why this is so :)

Cheers,

-- Tom :)

Eoin
Volunteer tester
Send message
Joined: 21 Oct 09
Posts: 1
Credit: 13,068
RAC: 0
Message 944465 - Posted: 1 Nov 2009, 18:41:04 UTC - in response to Message 943962.
Last modified: 1 Nov 2009, 18:42:19 UTC

Change that 1.6GB file's extension to .pdf and see if Acrobat Reader will open it. Should be about the right size for a set of blueprints.


Just had to say... lol :D

Cheers :)

Erik
Send message
Joined: 27 Oct 09
Posts: 1
Credit: 398
RAC: 0
Denmark
Message 944510 - Posted: 1 Nov 2009, 23:00:47 UTC

Hi everyone,

I know this might be the wrong place for my query.

But I have been wondering. A couple of days ago I installed BOINC and got a few packages of data to analyze. But the last few days e.g. friday, saturday and sunday. I cant get any more data. It says that the project has no jobs available.

Is SETI@home ending??

cheers

Erik

1mp0£173
Volunteer tester
Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 944514 - Posted: 1 Nov 2009, 23:54:34 UTC - in response to Message 944510.

Is SETI@home ending??

No, it is not ending.

Most projects, including SETI@Home, do not promise that they will always have enough work.

It's a combination of maintenance being done at the telescope and the fact that it takes longer than they'd like to radar-blank some of the old recordings.

It's sunday afternoon in Berkeley. I suspect they'll be working hard on this tomorrow (if they haven't had machines blanking radar all weekend).

____________

Profile Julie
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 28 Oct 09
Posts: 18370
Credit: 3,324,238
RAC: 6,292
Belgium
Message 944660 - Posted: 2 Nov 2009, 18:57:19 UTC - in response to Message 944121.
Last modified: 2 Nov 2009, 19:01:07 UTC

I ran empty for 2 days!! Now I got someting from Astropulse. Really glad, thought SETI forgot about me:(
Does this happen a lot? Next time I don't have to worry then. just wanna be useful:)
____________


rOZZ

Cosmic_Ocean
Avatar
Send message
Joined: 23 Dec 00
Posts: 2233
Credit: 8,419,341
RAC: 3,919
United States
Message 944665 - Posted: 2 Nov 2009, 19:23:22 UTC - in response to Message 944660.

I ran empty for 2 days!! Now I got someting from Astropulse. Really glad, thought SETI forgot about me:(
Does this happen a lot? Next time I don't have to worry then. just wanna be useful:)

Some would argue that this sort of thing happens more frequently in the relative "recent past" than before.

There have been some work availability issues over the past 6-12 months where there are either dry spells or the bandwidth is fully saturated and therefore, hardly anyone gets new work.

The current issue is that the telescope shut down for the entire month of October for maintenance, and the recorded data from September did not last the entire duration. There is some old data in off-site storage, but it needed a new piece of in-house software for it to be pre-processed and then split into WUs. I believe the pre-processing software is not able to go anywhere near real-time, and takes ~5 times longer to pre-process than it takes to split into WUs, so new tasks do come along every couple of hours, but it's a slow process.
____________

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

Profile Julie
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 28 Oct 09
Posts: 18370
Credit: 3,324,238
RAC: 6,292
Belgium
Message 944675 - Posted: 2 Nov 2009, 19:57:07 UTC - in response to Message 944665.

I ran empty for 2 days!! Now I got someting from Astropulse. Really glad, thought SETI forgot about me:(
Does this happen a lot? Next time I don't have to worry then. just wanna be useful:)

Some would argue that this sort of thing happens more frequently in the relative "recent past" than before.

There have been some work availability issues over the past 6-12 months where there are either dry spells or the bandwidth is fully saturated and therefore, hardly anyone gets new work.

The current issue is that the telescope shut down for the entire month of October for maintenance, and the recorded data from September did not last the entire duration. There is some old data in off-site storage, but it needed a new piece of in-house software for it to be pre-processed and then split into WUs. I believe the pre-processing software is not able to go anywhere near real-time, and takes ~5 times longer to pre-process than it takes to split into WUs, so new tasks do come along every couple of hours, but it's a slow process.

Thankx, that was some useful information:)

Message boards : Technical News : Well Well (Oct 29 2009)

Copyright © 2014 University of California