mersenneforum.org PRPnet beta test server
 Register FAQ Search Today's Posts Mark Forums Read

 2009-05-02, 19:59 #45 IronBits I ♥ BOINC!     Oct 2002 Glendale, AZ. (USA) 3×7×53 Posts Whether the Servers are running or not has zero impact to the copying off of the files. Servers do not need to be stopped for that to happen because, it's being run on Linux not Windows My Servers run 24/7 unless you give me more k/n pairs to add to knpairs.txt at which time I stop, add the pairs, and then re-start the server. Last fiddled with by IronBits on 2009-05-02 at 20:02
 2009-05-02, 21:26 #46 Flatlander I quite division it     "Chris" Feb 2005 England 31×67 Posts I bet Gary used to jump on his toys when he was small, to see if they would break. I agree though. After the bug with Phrot we don't want one here too.
2009-05-02, 22:47   #47
gd_barnes

May 2007
Kansas; USA

276E16 Posts

Quote:
 Originally Posted by IronBits Whether the Servers are running or not has zero impact to the copying off of the files. Servers do not need to be stopped for that to happen because, it's being run on Linux not Windows My Servers run 24/7 unless you give me more k/n pairs to add to knpairs.txt at which time I stop, add the pairs, and then re-start the server.
Let me restate:

We need to see if the copying off still fully works IF the server is stopped in the middle of it's copying process for any reason whatsoever...since that can happen for any reason at any time in the real world.

I'm not saying that they need to be stopped for the copying off to work.

In other words, we want to test the server in all kinds of wierd situations that it might encounter in the real world while we are beta testing. In this case, we want to make sure that it completes its copying process if it has started but not finished it when it is stopped and then restarted. I'm more concerned about the interfacing between the PRPnet server and the stats/results process than anything. I do not want us to lose or have duplicate results because it can be such a mess to straighten out.

Funny you should mention that Chris. No, I didn't like to "stress test" my toys (lol) as a kid but as an adult and a mainframe programmer, I used to like to find ways to break programs. I'd do all kinds of things to them: Feed them empty files, put alphas in numeric fields, run huge files through them, run files through that I knew would cause monetary overflow (or underflow) in some fields, etc. When you're running a back office (i.e. accounting and billing) system for an insurance co., you better make sure you fully test new programs or changes to existing programs before putting them in the production environment.

Gary

Last fiddled with by gd_barnes on 2009-05-02 at 22:53

 2009-05-03, 04:21 #48 Razor_FX_II     Jan 2009 4310 Posts Has anyone experienced a client not saying in the window that its returning or getting new work but get new work anyway. I restarted it and then it ran fine. Not sure if these that ran one right after another were returned or not. I checked the log file and here is the results. Code: [2009-05-03 03:46:40 GMT] User name Razor_FX_II at email address is razorfxii@sio.midco.net [2009-05-03 03:49:05 GMT] G2000: 3099*2^428793-1 is not prime. Residue 21E42EC297B0A5E3 [2009-05-03 03:52:28 GMT] G2000: 3123*2^428794-1 is not prime. Residue A2129B9C15D2D9B1 [2009-05-03 03:55:52 GMT] G2000: 3145*2^428793-1 is not prime. Residue CFE6B12A0E73F935 [2009-05-03 03:59:16 GMT] G2000: 3149*2^428796-1 is not prime. Residue 631725BCAB5164B6 [2009-05-03 04:02:51 GMT] G2000: 3183*2^428795-1 is not prime. Residue 48237763EB8D9A16 [2009-05-03 04:06:16 GMT] G2000: 3189*2^428796-1 is not prime. Residue 7F9255F6C955D2F7 [2009-05-03 04:09:40 GMT] G2000: 3195*2^428793-1 is not prime. Residue FC1A5ADB3F461158 [2009-05-03 04:12:14 GMT] G2000: Could not open file [lresults.txt] for reading. Assuming user stopped with ^C [2009-05-03 04:12:58 GMT] G2000: Attempting to send previously completed work [2009-05-03 04:12:59 GMT] G2000: Returning work to server nplb-gb1.no-ip.org at port 2000 [2009-05-03 04:13:03 GMT] G2000: INFO: Test for candidate 3099*2^428793-1 accepted [2009-05-03 04:13:03 GMT] PRPNet Client application v2.0.4 started [2009-05-03 04:13:03 GMT] User name Razor_FX_II at email address is razorfxii@sio.midco.net [2009-05-03 04:13:04 GMT] G2000: Getting work from server nplb-gb1.no-ip.org at port 2000 [2009-05-03 04:16:33 GMT] G2000: 3151*2^432301-1 is not prime. Residue 9584289E3B7DF743 [2009-05-03 04:16:34 GMT] Total Time: 0:03:35 Total Tests: 1 Total PRPs Found: 0 [2009-05-03 04:16:34 GMT] G2000: Returning work to server nplb-gb1.no-ip.org at port 2000 [2009-05-03 04:16:37 GMT] G2000: INFO: Test for candidate 3151*2^432301-1 accepted [2009-05-03 04:16:37 GMT] G2000: Getting work from server nplb-gb1.no-ip.org at port 2000 Last fiddled with by Razor_FX_II on 2009-05-03 at 05:01
2009-05-03, 06:28   #49
mdettweiler
A Sunny Moo

Aug 2007
USA (GMT-5)

186916 Posts

Quote:
 Originally Posted by Razor_FX_II Has anyone experienced a client not saying in the window that its returning or getting new work but get new work anyway. I restarted it and then it ran fine. Not sure if these that ran one right after another were returned or not. I checked the log file and here is the results. Code: [2009-05-03 03:46:40 GMT] User name Razor_FX_II at email address is razorfxii@sio.midco.net [2009-05-03 03:49:05 GMT] G2000: 3099*2^428793-1 is not prime. Residue 21E42EC297B0A5E3 [2009-05-03 03:52:28 GMT] G2000: 3123*2^428794-1 is not prime. Residue A2129B9C15D2D9B1 [2009-05-03 03:55:52 GMT] G2000: 3145*2^428793-1 is not prime. Residue CFE6B12A0E73F935 [2009-05-03 03:59:16 GMT] G2000: 3149*2^428796-1 is not prime. Residue 631725BCAB5164B6 [2009-05-03 04:02:51 GMT] G2000: 3183*2^428795-1 is not prime. Residue 48237763EB8D9A16 [2009-05-03 04:06:16 GMT] G2000: 3189*2^428796-1 is not prime. Residue 7F9255F6C955D2F7 [2009-05-03 04:09:40 GMT] G2000: 3195*2^428793-1 is not prime. Residue FC1A5ADB3F461158 [2009-05-03 04:12:14 GMT] G2000: Could not open file [lresults.txt] for reading. Assuming user stopped with ^C [2009-05-03 04:12:58 GMT] G2000: Attempting to send previously completed work [2009-05-03 04:12:59 GMT] G2000: Returning work to server nplb-gb1.no-ip.org at port 2000 [2009-05-03 04:13:03 GMT] G2000: INFO: Test for candidate 3099*2^428793-1 accepted [2009-05-03 04:13:03 GMT] PRPNet Client application v2.0.4 started [2009-05-03 04:13:03 GMT] User name Razor_FX_II at email address is razorfxii@sio.midco.net [2009-05-03 04:13:04 GMT] G2000: Getting work from server nplb-gb1.no-ip.org at port 2000 [2009-05-03 04:16:33 GMT] G2000: 3151*2^432301-1 is not prime. Residue 9584289E3B7DF743 [2009-05-03 04:16:34 GMT] Total Time: 0:03:35 Total Tests: 1 Total PRPs Found: 0 [2009-05-03 04:16:34 GMT] G2000: Returning work to server nplb-gb1.no-ip.org at port 2000 [2009-05-03 04:16:37 GMT] G2000: INFO: Test for candidate 3151*2^432301-1 accepted [2009-05-03 04:16:37 GMT] G2000: Getting work from server nplb-gb1.no-ip.org at port 2000
Well, based on some discussions I've had with Mark via email, you definitely shouldn't be losing any results. Whenever the client says "Returning work to server x at port x" it always tries to return all queued results for that server. So, you should be OK.

2009-05-03, 09:32   #50
MyDogBuster

May 2008
Wilmington, DE

2×13×109 Posts

Quote:
 Funny you should mention that Chris. No, I didn't like to "stress test" my toys (lol) as a kid but as an adult and a mainframe programmer, I used to like to find ways to break programs. I'd do all kinds of things to them: Feed them empty files, put alphas in numeric fields, run huge files through them, run files through that I knew would cause monetary overflow (or underflow) in some fields, etc. When you're running a back office (i.e. accounting and billing) system for an insurance co., you better make sure you fully test new programs or changes to existing programs before putting them in the production environment.
I can relate to Gary's testing ideas. I was a mainframe back office programmer also (before I got old and cute). The easiest way to debug or test a program is to make it fail. Feed it junk, pull the plug, run it with other programs that share the same DB. And when you think you have it perfect and put it into production, be sure to keep your evenings free for the first couple months of production runs. The phone will ring, and you will be going in to fix it anyway. Before Murphy wrote all his laws, he was a programmer. LOL

2009-05-03, 14:27   #51
rogue

"Mark"
Apr 2003
Between here and the

7×11×73 Posts

Quote:
 Originally Posted by mdettweiler Well, based on some discussions I've had with Mark via email, you definitely shouldn't be losing any results. Whenever the client says "Returning work to server x at port x" it always tries to return all queued results for that server. So, you should be OK.
This is not an issue. The client is not capturing the other INFO messages from the server and outputting them. It will be fixed in the next release. I hope to release a beta tomorrow.

 2009-05-03, 14:58 #52 gd_barnes     May 2007 Kansas; USA 2×72×103 Posts Can someone please define the word "binary" as it is used in this thread? Before I got into DC 2 years ago, I never heard the term used in the way that it is frequently used on this forum. It appears that it means some sort of special program but can someone be more specific? Perhaps a Linux or math-related program? Thanks, Gary Last fiddled with by gd_barnes on 2009-05-03 at 14:58
2009-05-03, 15:06   #53
Mini-Geek
Account Deleted

"Tim Sorbera"
Aug 2006
San Antonio, TX USA

426710 Posts

Quote:
 Originally Posted by gd_barnes Can someone please define the word "binary" as it is used in this thread? Before I got into DC 2 years ago, I never heard the term used in the way that it is frequently used on this forum. It appears that it means some sort of special program but can someone be more specific? Perhaps a Linux or math-related program? Thanks, Gary
In short: It means an executable, I don't know why we call it a binary.
In long:
It means a compiled program (as opposed to a program still in source code form). a.k.a. executable. (this is not a Linux- or math-specific term) On Windows this is a .exe, on other systems it's...um...not that simple...no extension, I think, but somehow the OS knows it's an executable. I don't really know precisely.
It's probably used here so much more than other places because the source code is (almost?) always available, and there is much variation in OSs and bits (e.g. Win32, Win64, Linux32, Linux64, etc. would all have the same source code but different binaries).

Last fiddled with by Mini-Geek on 2009-05-03 at 15:07 Reason: added in short

2009-05-03, 21:34   #54
mdettweiler
A Sunny Moo

Aug 2007
USA (GMT-5)

3·2,083 Posts

Quote:
 Originally Posted by Mini-Geek In short: It means an executable, I don't know why we call it a binary. In long: It means a compiled program (as opposed to a program still in source code form). a.k.a. executable. (this is not a Linux- or math-specific term) On Windows this is a .exe, on other systems it's...um...not that simple...no extension, I think, but somehow the OS knows it's an executable. I don't really know precisely. It's probably used here so much more than other places because the source code is (almost?) always available, and there is much variation in OSs and bits (e.g. Win32, Win64, Linux32, Linux64, etc. would all have the same source code but different binaries).
Yep, pretty much as Mini-Geek said.

 2009-05-03, 23:13 #55 IronBits I ♥ BOINC!     Oct 2002 Glendale, AZ. (USA) 3·7·53 Posts Basically, a non-text file vs. a TEXT based file.

 Similar Threads Thread Thread Starter Forum Replies Last Post johnadam74 Software 2 2016-01-01 15:58 ltd Sierpinski/Riesel Base 5 15 2013-03-19 18:03 ltd Prime Sierpinski Project 9 2011-03-15 04:58 Joe O Sierpinski/Riesel Base 5 6 2010-12-25 20:45 mdettweiler No Prime Left Behind 40 2010-01-30 18:05

All times are UTC. The time now is 22:49.

Sat Apr 4 22:49:02 UTC 2020 up 10 days, 20:22, 0 users, load averages: 1.29, 1.22, 1.35