Rejected k/n pairs

em99010pepe · 2008-02-08, 22:57

Bryan from RieselSieve project said this about the rejected pairs:

Quote:

Rejected is any k/n pair that comes back that is not either in the jobs.txt or jobs table (file vs mysql setup), whether it's too late, the server received the result, but the client did not receive the acknowledgment, the client never had the k/n pair (someone put it in, basically), or any number of reasons. Basically, the server didn't think that user should have had the k/n pair they returned.

I'm getting a bunch of rejected files from michaf and I think he should be credit for them even if others later receive the pairs to re-test. Meanwhile I increased the job prune to each two days instead of one, maybe this was the problem. Thoughts please.

Carlos

EDIT: I'm going to sleep. Just checked again the file and for the last two hours the server has been receiving all k/n pairs from michaf as rejected ones...

ET_ · 2008-02-08, 23:19

Golden slumbers...

Luigi

michaf · 2008-02-09, 06:47

Hmm... nothing I did wrong hopefully?
Everything looks ok here, albeit a too low prime content.
As far as I can tell from the lresults.txt file, all results have been sent successfully to the server (manual scan of it, might have missed a few)

And don't bother too much to give the credit, I don't mind being last in the race :)

Quote:

Originally Posted by em99010pepe

Bryan from RieselSieve project said this about the rejected pairs:

I'm getting a bunch of rejected files from michaf and I think he should be credit for them even if others later receive the pairs to re-test. Meanwhile I increased the job prune to each two days instead of one, maybe this was the problem. Thoughts please.

Carlos

EDIT: I'm going to sleep. Just checked again the file and for the last two hours the server has been receiving all k/n pairs from michaf as rejected ones...

em99010pepe · 2008-02-09, 08:32

I really don't know what's happening. I'll tell Karsten to count the rejected files because they are good, the residues are correct.

AES · 2008-02-09, 19:44

michaf, may i ask which os you are running?

My results seem to be going to the rejected file as well as the lresults file. The residues all appear to be correct. I'm seeing strange client behavior in 64-bit Vista, but the results and log files are all ok.

em99010pepe · 2008-02-09, 19:58

You both are getting rejected pairs...

Quote:

Originally Posted by AES

michaf, may i ask which os you are running?

I was not receiving rejected pairs when he was running for CRUS project.

michaf · 2008-02-10, 00:57

I have a desktop running XP-home
and a laptop running Vista Home Premium.
both running on 32bits

edit:
Hey... I just noticed that my laptop AND my desktop were crunching the SAME number:
887*2^328654-1

Any reason as why I get two of the same?

Quote:

Originally Posted by AES

michaf, may i ask which os you are running?

My results seem to be going to the rejected file as well as the lresults file. The residues all appear to be correct. I'm seeing strange client behavior in 64-bit Vista, but the results and log files are all ok.

kar_bon · 2008-02-10, 01:21

Quote:

Originally Posted by michaf

edit:
Hey... I just noticed that my laptop AND my desktop were crunching the SAME number:
887*2^328654-1

Any reason as why I get two of the same?

look up your llr-clientconfig.txt on both:
get sure they different services and guiport adresses.

compare the workfile.txt's! equal?

kar_bon · 2008-02-10, 10:15

all 1136 rejected k/n-pairs i got from Carlos are double work of the same user.
seems that 2 clients did same pairs and sent their results, but 2nd was rejected.

1136 pairs:
346 from AES
10 from paratima
780 from michaf

attachment contains table with:
- first 9 columns contains data from rejected file
- next comment and index from result-file
- next 8 columns data from result-file

Karsten

michaf · 2008-02-10, 10:45

I think I'm the dumb one here...
I made four dir's accessible from both my machines, each containing a llr-net
I intended to start each core in a different dir, but I apparently ended up using the same dir's twice.

If this was the problem, I'll be sending no double results from now on...

Ah well... at least we have some double-checks done then...

em99010pepe · 2008-02-10, 11:19

Karsten,

What about adding a column at the score table for double-check pairs?

michaf,

I already stop receiving rejected pairs. I think we all learned by your mistake.

Carlos

2008-02-09, 19:44	#5
AES Jul 2007 Tennessee 2⁵·19 Posts	michaf, may i ask which os you are running? My results seem to be going to the rejected file as well as the lresults file. The residues all appear to be correct. I'm seeing strange client behavior in 64-bit Vista, but the results and log files are all ok. Last fiddled with by AES on 2008-02-09 at 19:52

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Search for all amicable pairs up to 2^64	Sergei Chernykh	Aliquot Sequences	37	2018-01-27 19:23
Why was this rejected?	petrw1	PrimeNet	2	2015-10-01 17:11
GPU LL result rejected	mdettweiler	PrimeNet	4	2010-10-04 20:03
Tony Reix, I'm feeling rejected :)	Prime95	Lounge	2	2007-02-26 08:05
Double-checks come in pairs?	BigRed	Software	1	2002-10-20 05:29

2008-02-08, 23:19	#2
ET_ Banned "Luigi" Aug 2002 Team Italia 3²·5·107 Posts	Golden slumbers... Luigi

2008-02-09, 08:32	#4
em99010pepe Sep 2004 2×5×283 Posts	I really don't know what's happening. I'll tell Karsten to count the rejected files because they are good, the residues are correct.

2008-02-10, 10:45	#10
michaf Jan 2005 111011111₂ Posts	I think I'm the dumb one here... I made four dir's accessible from both my machines, each containing a llr-net I intended to start each core in a different dir, but I apparently ended up using the same dir's twice. If this was the problem, I'll be sending no double results from now on... Ah well... at least we have some double-checks done then...

2008-02-10, 11:19	#11
em99010pepe Sep 2004 2·5·283 Posts	Karsten, What about adding a column at the score table for double-check pairs? michaf, I already stop receiving rejected pairs. I think we all learned by your mistake. Carlos