i am getting lmh tfs regularly. hopefully.

Client-server sync oddity
 

I just updated my completion dates, and I noticed an oddity with an exponent:

[code]
kossy@leela:~/mprime256-linux64> grep 46567721 worktodo.txt
Factor=00000000000000000000000102C69129,46567721,63,68
kossy@leela:~/mprime256-linux64> grep 46567721 primenet_report.pl.html
46567721 F 68 0.0 52.0 61.0 07-Feb-08 08:43 quad-pro-quo 2399 v19/v20
[/code]So the client wants to trial factor the exponent from 63 to 68 bits, but the server says it *has* already been trial factored to 68 bits. Plus, the server assigned the exponent to me when I sent the updated completion date, whereas prime.log states:

[code]
[Wed Jan 30 07:16:10 2008 - ver 25.6]
...
Registering assignment: Trial factor M46567721
Assignment registered as: 00000000000000000000000102C69129
...
[Wed Feb 6 21:11:23 2008 - ver 25.6]
...
Sending expected completion date for M46567721: Mar 28 2008
...
[Thu Feb 7 09:38:49 2008 - ver 25.6]
...
Sending expected completion date for M46567721: Mar 30 2008
...
[/code]Which obviously means that

(a) the assignment dates back to at least January,
(b) the completion date for the exponent was updated successfully just yesterday,
(c) somehow the exponent was taken away from me overnight (it wasn't unreserved from my end and it was expected to finish well within my DaysOfWork=66 anyway)
(d) it was assigned to somebody else and finally
(e) they already finished it!

What's going on here? :confused:

Same problem with all these exponents:

[code]
kossy@leela:~/mprime256-linux64> grep " 68 " primenet_report.pl.html
46033901 F 68 1.5 24.0 60.0 07-Feb-08 08:40 06-Feb-08 20:13 quad-pro-quo 2399 v19/v20
46033963 F 68 1.5 24.0 60.0 07-Feb-08 08:36 06-Feb-08 20:18 quad-pro-quo 2399 v19/v20
46034069 F 68 1.5 24.0 60.0 07-Feb-08 08:40 06-Feb-08 20:13 quad-pro-quo 2399 v19/v20
46034137 F 68 1.5 24.0 60.0 07-Feb-08 08:42 06-Feb-08 20:15 quad-pro-quo 2399 v19/v20
46037339 F 68 1.5 25.0 60.0 07-Feb-08 08:36 06-Feb-08 20:18 quad-pro-quo 2399 v19/v20
46545349 F 68 11.9 51.0 60.0 07-Feb-08 08:38 27-Jan-08 10:23 quad-pro-quo 2399 v19/v20
46567721 F 68 1.0 51.0 60.0 07-Feb-08 08:43 quad-pro-quo 2399 v19/v20
kossy@leela:~/mprime256-linux64> cat worktodo.txt
...
Factor=00000000000000000000000102BE6BED,46033901,63,68
Factor=00000000000000000000000102BE6C2B,46033963,63,68
Factor=00000000000000000000000102BE6C95,46034069,63,68
Factor=00000000000000000000000102BE6CD9,46034137,63,68
Factor=00000000000000000000000102BE795B,46037339,63,68
Factor=00000000000000000000000102C639C5,46545349,66,68
...
kossy@leela:~/mprime256-linux64>[/code]mprime wants to factor to 68 bits though my status page says somebody did that already. I'll simply unreserve the lot of them, but I'd still love to see an explanation.

[QUOTE=ckdo;125147]Same problem with all these exponents:

Factor=00000000000000000000000102BE6BED,46033901,63,68
Factor=00000000000000000000000102BE6C2B,46033963,63,68
Factor=00000000000000000000000102BE6C95,46034069,63,68
Factor=00000000000000000000000102BE6CD9,46034137,63,68
Factor=00000000000000000000000102BE795B,46037339,63,68
Factor=00000000000000000000000102C639C5,46545349,66,68[/QUOTE]

Can you email the prime.log file? My records show that these have not been factored to 2^68, but the v4 server seems to think they were.

Never mind: TPR factored M46033901 to 2^68 on 2/6/08. I'll try to see if they had a valid reservation.

Update: I found that you reserved 46033901 on or about 12/27/07. I could not find a reservation by TPR.

My prime.log starts Jan 30th, when I switched to 25.6. Those exponents were all originally reserved using 24.14. Can't help you there.

[QUOTE=ckdo;125420]My prime.log starts Jan 30th, when I switched to 25.6. Those exponents were all originally reserved using 24.14. Can't help you there.[/QUOTE]

24.14 connects to the v4 server, 25.6 connects to the v5 server. The v5 server doesn't know what you have reserved at the v4 server.

[quote=Prime95;121281]3) PM or email to get the secret line to add to prime.txt so that this version contacts the new server.
[/quote]

No, Andi47, 25.6 contacts the v4 server as well unless told otherwise, which I didn't.

This time I witnessed client and server get out of sync. From my terminal window:

[code]
[Worker #3 Feb 11 08:02] Trial factoring of M45091727 to 2^68 is 99.59% complete. Time: 220.266 sec.
[Worker #3 Feb 11 08:03] M45091727 no factor from 2^67 to 2^68, Wd8: B0CB2C1B

[Comm thread Feb 11 08:03] Sending result to server: UID: ckdo/quad-pro-quo, M45091727 no factor from 2^67 to 2^68, Wd8: B0CB2C1B
[Comm thread Feb 11 08:03]
[Comm thread Feb 11 08:03] URL: http://mersenne.org/cgi-bin/pnHttp.exe?ar&4&88&4&10001&15369&ckdo&thisisnotmypassword&quad-pro-quo&45091727&0&68.0
[Comm thread Feb 11 08:04] RESPONSE:
[Comm thread Feb 11 08:04] pnResult=0

[Comm thread Feb 11 08:04] URL: http://mersenne.org/cgi-bin/pnHttp.exe?rm&4&126&4&16015&33606&ckdo&thisisnotmypassword&quad-pro-quo&UID:+ckdo/quad-pro-quo,+M45091727+no+factor+from+2^67+to+2^68,+Wd8:+B0CB2C1B%0A
[Comm thread Feb 11 08:04] RESPONSE:
[Comm thread Feb 11 08:04] pnResult=0

[Comm thread Feb 11 08:04] Getting assignment from server
[Comm thread Feb 11 08:04] URL: http://mersenne.org/cgi-bin/pnHttp.exe?ga&4&64&4&2179&7510&ckdo&thisisnotmypassword&quad-pro-quo&1&2&64.0
[Comm thread Feb 11 08:04] RESPONSE:
[Comm thread Feb 11 08:04] pnResult=7
[Comm thread Feb 11 08:04] PrimeNet error 20007: Unknown error code
[Comm thread Feb 11 08:04] Visit http://mersenneforum.org for help.
[Comm thread Feb 11 08:04] Will try contacting server again in 1 minute.

[Comm thread Feb 11 08:05] Getting assignment from server
[Comm thread Feb 11 08:05] URL: http://mersenne.org/cgi-bin/pnHttp.exe?ga&4&64&4&23203&462&ckdo&thisisnotmypassword&quad-pro-quo&1&2&64.0
[Comm thread Feb 11 08:06] RESPONSE:
[Comm thread Feb 11 08:06] pnResult=0
[Comm thread Feb 11 08:06] requestType=1
[Comm thread Feb 11 08:06] exponent=46941733
[Comm thread Feb 11 08:06] how_factored=63.0
[Comm thread Feb 11 08:06] Got assignment 00000000000000000000000002CC4625: Trial factor M46941733
[Comm thread Feb 11 08:06] Sending expected completion date for M46941733: Mar 12 2008
[Comm thread Feb 11 08:06] URL: http://mersenne.org/cgi-bin/pnHttp.exe?cd&4&68&4&21075&25001&ckdo&thisisnotmypassword&quad-pro-quo&46941733&30&0&2&0&1
[Comm thread Feb 11 08:06] RESPONSE:
[Comm thread Feb 11 08:06] pnResult=0

[Comm thread Feb 11 08:06] Getting assignment from server
[Comm thread Feb 11 08:06] URL: http://mersenne.org/cgi-bin/pnHttp.exe?ga&4&64&4&23062&15861&ckdo&thisisnotmypassword&quad-pro-quo&1&2&64.0
[Comm thread Feb 11 08:06] RESPONSE:
[Comm thread Feb 11 08:06] pnResult=0
[Comm thread Feb 11 08:06] requestType=1
[Comm thread Feb 11 08:06] exponent=46941779
[Comm thread Feb 11 08:06] how_factored=63.0
[Comm thread Feb 11 08:06] Got assignment 00000000000000000000000002CC4653: Trial factor M46941779
[Comm thread Feb 11 08:06] Sending expected completion date for M46941779: Mar 12 2008
[Comm thread Feb 11 08:06] URL: http://mersenne.org/cgi-bin/pnHttp.exe?cd&4&68&4&2683&47966&ckdo&thisisnotmypassword&quad-pro-quo&46941779&30&0&2&0&1
[Comm thread Feb 11 08:07] RESPONSE:
[Comm thread Feb 11 08:07] pnResult=23
[Comm thread Feb 11 08:07] PrimeNet error 3: Server busy
[Comm thread Feb 11 08:07] Visit http://mersenneforum.org for help.
[Comm thread Feb 11 08:07] Will try contacting server again in 1 minute.

[Comm thread Feb 11 08:08] Getting assignment from server
[Comm thread Feb 11 08:08] URL: http://mersenne.org/cgi-bin/pnHttp.exe?ga&4&64&4&7705&58157&ckdo&thisisnotmypassword&quad-pro-quo&1&2&64.0
[Comm thread Feb 11 08:08] RESPONSE:
[Comm thread Feb 11 08:08] pnResult=7
[Comm thread Feb 11 08:08] PrimeNet error 20007: Unknown error code
[Comm thread Feb 11 08:08] Visit http://mersenneforum.org for help.
[Comm thread Feb 11 08:08] Will try contacting server again in 1 minute.

[Comm thread Feb 11 08:09] Getting assignment from server
[Comm thread Feb 11 08:09] URL: http://mersenne.org/cgi-bin/pnHttp.exe?ga&4&64&4&8204&21010&ckdo&thisisnotmypassword&quad-pro-quo&1&2&64.0
[Comm thread Feb 11 08:09] RESPONSE:
[Comm thread Feb 11 08:09] pnResult=0
[Comm thread Feb 11 08:09] requestType=1
[Comm thread Feb 11 08:09] exponent=46941787
[Comm thread Feb 11 08:09] how_factored=63.0
[Comm thread Feb 11 08:09] Got assignment 00000000000000000000000002CC465B: Trial factor M46941787
[Comm thread Feb 11 08:09] Sending expected completion date for M46941787: Mar 12 2008
[Comm thread Feb 11 08:09] URL: http://mersenne.org/cgi-bin/pnHttp.exe?cd&4&68&4&30630&46289&ckdo&thisisnotmypassword&quad-pro-quo&46941787&30&0&2&0&1
[Comm thread Feb 11 08:10] RESPONSE:
[Comm thread Feb 11 08:10] pnResult=0

[Comm thread Feb 11 08:10] Done communicating with server.
[/code](Irrelevant lines removed, empty lines added. Times are UTC.)

The problem is with M46941779, which was added to my worktodo file but because the server never received the expected completion date, the assignment wasn't registered server-side and didn't show up on my status page nor in status.txt, even after the 09:00 update.

Anyone could have reserved the exponent by the next time I sent completion dates for all my exponents. At 09:50 I finally decided to not take any risk any pasted the URL for the failed request into Firefox. Voila, both status.txt and my status page now show the reservation. With the 09:50 timestamp, of course, but I can live with that.

The root of all evil here is the client not retrying (only) the failed update request but instead requesting yet another assignment (which it probably wouldn't have done had the update request not failed).

Here is the TPR reservation line as of Feb 4th according to my records:
[CODE]46033901 F 63 4.0 2.9 62.9 02-Feb-08 09:07 31-Jan-08 12:57 bayanne_clv9 1993 v19/v20[/CODE]

v4 server or v5?

v4.