http://setiathome.berkeley.edu/workunit.php?wuid=154621326
Got and sent at September 5th

I just had a quick look .... my oldest pending dates back to september 12, which I returned september 20. Wingman should have reported on november 8, but didn't. The wu was then sent to somebody who now has until january 8. He has reported his last result on november 6, and now has some 103 wu's 'in progress', though in his case I think that term should be used with some caution ;-)

I guess the wu finds the idea of hibernation appealing. With a bit of luck it might get awake somewhere in spring :-)

Regards,
John.

Ok, now that we have a pretty good list of old pending wus to work with, lets see if we can determine a reason for them. It appears to me by glancing through the posts here that non-responsive clients is the main reason for these wus to be still pending. Would you agree with this?
I'm not saying that this is not a problem, but it appears by the small number of posts here that it is less of a problem than first imagined. The system seems to be working properly by reissuing wus which have been assigned to these non-responsive computers, but the longer deadlines causes longer delays in getting the wus validated.

I'll leave this stickied for a few more days for comments, suggestions and last minute additions to the wu list, then unless Eric or Matt request that it be left up longer, I will unsticky this as it will have served the purpose for which I stickied it.

---

Jim

Some people plan their life out and look back at the wealth they've had.
Others live life day by day and look back at the wealth of experiences and enjoyment they've had.

Ok, now that we have a pretty good list of old pending wus to work with, lets see if we can determine a reason for them. It appears to me by glancing through the posts here that non-responsive clients is the main reason for these wus to be still pending. Would you agree with this?
I'm not saying that this is not a problem, but it appears by the small number of posts here that it is less of a problem than first imagined. The system seems to be working properly by reissuing wus which have been assigned to these non-responsive computers, but the longer deadlines causes longer delays in getting the wus validated.

I'll leave this stickied for a few more days for comments, suggestions and last minute additions to the wu list, then unless Eric or Matt request that it be left up longer, I will unsticky this as it will have served the purpose for which I stickied it.

My Einstein@home deadlines are two weeks long. My QMC deadlines can reach three weeks. My latest SETI deadline goes to January 18 2008. This is simply ridiculous, even if my CPU is a 400 MHz PII. It has already crunched 1/3 of the WU, even if its share is one third of Einstein's.
Tullio

---

Another problem with these WUs are hosts looking to see why they are pending not always wingmen. I have a problem with -9 Vista overflow usually 31 pulses found while wingmen are clear, These I usually report to Joe as I get a fair few on Beta. Anyone else getting these. Cannot recall having seen them reported over here.

Take WU 154621326 -9 overflow problem 18 spikes
13 gaussian

---

Jim,

I think your diagnosis is quite correct. As far as I can tell, many of the long-living pendings were initially caused by clients that disembarked due to the problems seti was having with their servers in the september/october period. If I glance through some of my own pendings, I get the distinct feeling that a lot of people starting with seti in that period were not pleased with the technical problems seti had at that time and simply gave up after a relatively short time.

I guess they haven't pulled out in an abrupt way, not by detaching or aborting wu's but by simply deleting stuff or suspending the project. Might even have removed boinc completely from their setup.

Of course, the pendings have a tendency to survive quite long due to the relatively long deadlines imposed. They'll validate in the long run, and the buffer of pendings, although not affecting the long-term RAC, has a nice effect of smoothing the RAC if a machine stops crunching for a short time due to malfunction or whatever. Downside is that the same smoothing effects make it take longer to reach the new 'final' value of RAC when you have cranked up your processor some tens of MHz.

Or as Johan Cruyff (a Dutch soccer player) tended to say: every disadvantage also has an advantage.

Regards,
John.

Does anyone still have August/September pendings???

---

Does anyone still have August/September pendings???

Yes, e.g. this wu is pending since 28th August and will be for another 3 weeks (unless wingmen...) .
And a few more from september are also pending.

---

_\|/_
U r s

Does anyone still have August/September pendings???

Yes, I still have three Augusts
reported Aug 14th
reported Aug 18th
reported Aug 31st

I also have 9 reported in Sept.

-Dave

---

Does anyone still have August/September pendings???

Yes, My 11 August issue / Reported 13th August is still there [ An AWOL, then checked but no consensus, then another AWOL]. conceivably for much longer yet, though it should be reissued in a couple of days I believe. (from memory 2nd place title for oldest pending reported here ? [2nd Place By about an hour I think, :D] )

Jason

---

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.

Sounds like old WU's hanging in pending status is still an issue. I suspect people just stopped posting them here.

We're just a few days from December so the August pendings are between 3 and 4 months old - that is a loooooooooong time for a pending to report.

---

Does anyone still have August/September pendings???

Yes, My 11 August issue / Reported 13th August is still there [ An AWOL, then checked but no consensus, then another AWOL]. conceivably for much longer yet, though it should be reissued in a couple of days I believe. (from memory 2nd place title for oldest pending reported here ? [2nd Place By about an hour I think, :D] )

Jason

I see that one just got validated today !

588577147 3676307 11 Aug 2007 14:28:24 UTC 3 Oct 2007 10:43:01 UTC Over No reply New 0.00 --- ---
626013482 3528722 3 Oct 2007 14:27:22 UTC 6 Oct 2007 4:22:32 UTC Over Success Done 12,416.25 54.12 54.12
628221308 3640976 6 Oct 2007 4:22:44 UTC 28 Nov 2007 0:37:21 UTC Over No reply New 0.00 --- ---
588577146 3106419 11 Aug 2007 14:28:24 UTC 13 Aug 2007 4:02:44 UTC Over Success Done 16,262.48 54.12 54.12
668756324 3746296 28 Nov 2007 3:13:40 UTC 28 Nov 2007 12:42:30 UTC Over Success Done 10,706.05 54.12 54.12

Well done to "Þórgnýr Thoroddsen" for returning the _4 result quickly.

---

Sir Arthur C Clarke 1917-2008

Reported 13th August

Yes :D This one (long standing) crunched immediately on reissue it seems :D Thanks "Þórgnýr Thoroddsen"!

---

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.

And thanks to CrazyBear, my pending from 14 Aug just got credit...
Waiting until at least early Dec. for the other 2 August ones.

-Dave

---

My 25 August just got reissued today. New wingman has an 11 day turnround, but at least he's crunching (and donating). Thanks Jay.

My oldest is from September 3rd - one which overflowed (so not many points at stake!). My wingman for that has 45 tasks in hand and a average turnaround of 78 days: he times out tomorrow!

---

My oldest is from September 3rd - one which overflowed (so not many points at stake!). My wingman for that has 45 tasks in hand and a average turnaround of 78 days: he times out tomorrow!

Ouch!

---

Looks like this thread has out lived its life.

It doesn't appear long pendings are an issue at the moment.

---

OK - not so old as some but this one could last longer I suspect...

I got it on 13 Sept and returned it on 14 Sept (74.78 Credits). My wingman got it on 12 Sept and timed out on 8 Nov. The new wingman is Anon and is showing 30 WU's issued between 2nd and 8th Nov (non returned) and is due to time out on 4th Jan. It would be just my luck for it to be re-issued to a 3.xx and I will get zilch!! (My one and only March 2008 WU so far (100+ claim) was paired up with a 4.45 and received 68).

Hey-ho...

F.

---

Ok, well here's an oldy but goody - I've got

http://setiweb.ssl.berkeley.edu/workunit.php?wuid=147107700
(sorry, haven't studied this BBtag stuff)

which I got 13 Aug
wingman timed out 5 Oct
next wingman timed out 28 Nov

newest wingman looks to be live, and likely to return
in only a couple more days.

Salivating over finally getting those 54 creds!

Dave