| Author |
Message |
|---|
|
|
|
PRPNet server is now up.
Current Projects
server = "pgllr.mine.nu"
port 7171: PPS Test on 3000<k<4000 360k<n<363K (k*2^n+1
port 8181: Sophie Germain Prime Search (SGS) (k*2^666666-1)
port 6868: Generalized Woodall & Cullen; b=13 (n*b^n-/+1)
The client file and all you need to run our server is in here.
http://pgllr.mine.nu/PRPNet/
There is a new PRPclient file 1.0.5
To windows and Linux
NEW Mac client is now avalible
You can run one on each core like LLRNet.
Right now we have work on PPS n~360k
If you find a new prime it will make TOP5000 :)
Here is some guidelines to get started
This is prpclient.ini
// email= is a REQUIRED field (either here, or on the command line)
// Comment the next line out, and fill in your email address.
// *** Where to send Found Prime Mail !! ***
email=user@domain.com
// This value differtiates clients using the same e-mail ID
// ***Enter your PG username here !! ***
clientid=username
// server= are the servers that you want to get work from. The first
// number is a percentage of work to get from that server. The second
// parameter is a unique identifier appended to file names created by
// the client for that server. The third is the server IP. The fourth
// is the port for that server. It is normally 7101. Using a percentage
// of 0 will get work from that server only if the client could not
// connect to any other servers. Please read the prpnet_servers.txt
// in this directory for information on some known PRPNet servers.
server=100:PGPRPNet:pgllr.mine.nu:7171
// maxworkunits= is the maximum number of workunits that a client can
// request at a time. If a client requests too many the chances that
// some workunits timeout is greater. The default is 5. The server
// also has a limit, so the actual limit will be the less of this
// value and the value on the server.
maxworkunits=3
// This is the name of LLR executable. On Windows, this needs to be
// the LLR console application, not the GUI application. The GUI
// application does not terminate when the PRP test is done.
// On some systems you will need to put a "./" in front of the executable
// name so that it looks in the current directory for it rather than
// in the system path.
// LLR can be downloaded from http://jpenne.free.fr/index2.html
llrexe=./sllr
// This is the name of Phrot executable.
// On some systems you will need to put a "./" in front of the executable
// name so that it looks in the current directory for it rather than
// in the system path.
// Phrot can be downloaded from http://home.roadrunner.com/~mrodenkirch/
phrotexe=./phrot
// Default idle status is 1 (idle)
idle=1
// Timeout on communications errors
// (default is 60 minutes, minimum is 1 minute if not specified here...)
// Note that the actual used in the client is anywhere from 90% to 110% of this value
errortimeout=2
// Size limit for the prpclient.log file...
// 0 means no limit.
// -1 means no log.
loglimit=5000000
// Set the debug level for the client
// 0 - no debug messages
// 1 - all debug messages
// 2 - output debug messages from socket communication
debuglevel=0
/Lennart |
|
|
John Honorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
                 
|
|
This will eventually replace our PGLLRNet. PRPNet allows use of the latest software. This means that the latest version of LLR is used which is about 10% faster than the one used in LLRNet. Also, PRPNet makes use of Phrot which is much faster than LLR at non base 2 tests. This will help with the generalized Cullen/Woodall search at base 13.
Please give it a test run. Right now Lennart has loaded it with some "live" test work.
Report back here your experience.
____________
|
|
|
|
|
|
I had a quick go...
The log shows that the app started 21:44 and I stopped it at 20:05
In that time it ran 36 tests. That's 2 mins 45 each.
Looking through the log, the minimum time was 155 and maximum 240.
PPS LLR under BOINC on the same machine takes 4 mins 40.
Q6600 @ stock 2.4 GHz Ubuntu 64bit
Peter
P.S. I'm a Linux nub.
____________
35 x 2^3587843+1 is prime! |
|
|
dh1sajVolunteer tester
Send message
Joined: 13 Jul 08
Posts: 49
ID: 25532
Credit: 3,600,543,070
RAC: 141,986
                          
|
That's 2 mins 45 each.
Looking through the log, the minimum time was 155 and maximum 240.
PPS LLR under BOINC on the same machine takes 4 mins 40.
Second that.
Machine here is a 'bit' slower, but relation is identical.
4.5mins here, 9+mins on BOINC.
AMD4200+ @2.3GHz, 64-Bit Ubuntu 8.04
Joe |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
That's 2 mins 45 each.
Looking through the log, the minimum time was 155 and maximum 240.
PPS LLR under BOINC on the same machine takes 4 mins 40.
Second that.
Machine here is a 'bit' slower, but relation is identical.
4.5mins here, 9+mins on BOINC.
AMD4200+ @2.3GHz, 64-Bit Ubuntu 8.04
Joe
Thanks for the feedback!
FYI...the "live" data is from a much lower n range. n~350K as opposed to n~483K in BOINC.
____________
|
|
|
 pschoeferVolunteer developer
Volunteer tester
Send message
Joined: 20 Sep 05
Posts: 664
ID: 845
Credit: 2,329,565,917
RAC: 2,936,618
                       
|
|
P4M 1,8GHz, openSuSE 11.1 i586, about 350 seconds per WU.
No problems with the client so far. :)
____________
  |
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1198
ID: 18646
Credit: 566,086,232
RAC: 939,121
               
|
|
1563*2^350601+1 is not prime. Proth RES64: BA9BE54E79C959EE Time : 247.844 sec.
[2009-01-01 18:58:48 GMT] PGPRPNet: 1563*2^350601+1 is not prime. Residue BA9BE54E79C959EE Time : 247
[2009-01-01 18:58:48 GMT] Total Time: 0:04:08 Total Tests: 1 Total PRPs Found: 0
[2009-01-01 18:58:49 GMT] PGPRPNet: Returning work to server pgllr.mine.nu at port 7171
[2009-01-01 18:58:49 GMT] PGPRPNet: INFO: All 1 test results were accepted
let em run on Ubuntu-Linux 64-bit AMD Phenom 9550 Quad-Core
got to install zip-package (I dont know why i m not need it before ;) )
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
 |
|
|
|
|
|
1577*2^350947+1 is not prime. Proth RES64: B2A0595153672045 Time : 157.713 sec.
[2009-01-01 19:54:15 GMT] PGPRPNet: 1577*2^350947+1 is not prime. Residue B2A0595153672045 Time : 157
[2009-01-01 19:54:15 GMT] Total Time: 0:08:31 Total Tests: 3 Total PRPs Found: 0
[2009-01-01 19:54:15 GMT] PGPRPNet: Returning work to server pgllr.mine.nu at port 7171
[2009-01-01 19:54:17 GMT] PGPRPNet: INFO: All 3 test results were accepted
Looks like approx 157 seconds per WU here.
Running on a Intel C2D E6600 @ 2.4ghz on 64bit Gentoo |
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1198
ID: 18646
Credit: 566,086,232
RAC: 939,121
|
let em run on Ubuntu-Linux 64-bit AMD Phenom 9550 Quad-Core
it still runs
last message:
[2009-01-02 13:39:58 GMT] Total Time: 18:21:21 Total Tests: 251 Total PRPs Found: 0
i will let em run, until you inform me to stop ...
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
|
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 793
ID: 18447
Credit: 355,699,359
RAC: 1,449,978
           
|
|
Seems to be running fine in general - I will also leave running until you say to stop/upgrade/whatever.
A couple of comments:
1) if you hit ctrl-C, the client stops ok, but it doesn't resume any already downloaded WUs when it next starts up. On one occasion, it left one of the work_PGPRPNet.* files behind (can't remember which one) and the client dumped core on startup until this file was removed. Don't have full details anymore. I'll try some more tests to get the exact behaviour explained fully.
2) is there any way of telling the client to download N WUs, run them, upload the results and then exit? This would be equivalent to setting refill = 0 in the PG_LLRNetlinux_Batch24 client.
____________
 |
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 23,566,369
RAC: 73,396
      
|
|
will there be manual credit?
And is changing the e-mailaddres enough?
EDIT: I also found a litle bug:
this error-message:
warddr@warddr-laptop:~/Bureaublad/PRPclient$ ./prpclient
Client ID supplied (ward.ubuntu@gmail.com) is not valid as it has an embedded '@', ':' or ' '.
the error-message tels me that client ID supplied is ward.ubuntu@gmail.com, but that's the email supplied, the id supplied ("Ward De Ridder") was wrong, I changed it to Ward_De_Ridder and now everything works fine.
Ward
____________
 |
|
|
|
|
|
NEWS !!!!!!!!!!!!!! NEWS
In the first post i have added a client for Windows !!!
Released today.
Report all bugs here :)
/Lennart |
|
|
|
|
will there be manual credit?
And is changing the e-mailaddres enough?
EDIT: I also found a litle bug:
this error-message:
warddr@warddr-laptop:~/Bureaublad/PRPclient$ ./prpclient
Client ID supplied (ward.ubuntu@gmail.com) is not valid as it has an embedded '@', ':' or ' '.
the error-message tels me that client ID supplied is ward.ubuntu@gmail.com, but that's the email supplied, the id supplied ("Ward De Ridder") was wrong, I changed it to Ward_De_Ridder and now everything works fine.
Ward
I have reported the bug.
Thanks. I'll take care of it. (from Mark)
/Lennart |
|
|
|
|
|
Hi Lennart.
It's Carlos from Free-DC Team.
I added 4 cores to the test. Meanwhile PPRNet 1.0.3 is out.
BTW, please do not change your llrnet server with this one because llrnet is based on llr3.5.0 and this one is still faster, for small numbers, than llrnet or prpnet based on llr3.7.1c. If I am not mistaken with n=500k llr3.7.1c is already faster than llr3.5.0. Now I can't remember what happens between 400k<n<500k. On time I tested all these things but I can't recall the conclusion.
This comparison is only for standard clients and not BOINC. BOINC will always be slower than the standard clients because of the the way the program behaves. This means, if you have BOINC based on llr3.7.1c and llr3.7.1c standard client testing the same kn pair the latter is still faster. What happens here is simple: at least one core is wasted when downloading new tasks, Boinc.exe gets busy, and also there's a delay of 3-4 seconds everytime a new test is started because llr application and the wrapper program have to shutdown and restart.
Please take this as someone that has been around on prime hunting for the last 5 years. Hope this helps.
How many k's are on the server? And what ranges? How many candidates? I need numbers...lol
Carlos |
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 793
ID: 18447
Credit: 355,699,359
RAC: 1,449,978
|
1) if you hit ctrl-C, the client stops ok, but it doesn't resume any already downloaded WUs when it next starts up. On one occasion, it left one of the work_PGPRPNet.* files behind (can't remember which one) and the client dumped core on startup until this file was removed. Don't have full details anymore. I'll try some more tests to get the exact behaviour explained fully.
I've been unable to reproduce this since. Every time it interrupts, it keeps all 3 work_PGPRPNet.* files and a zNNNNNNN cache, and restarts the downloaded work correctly. I guess this must have been a corner case where the llr process was just starting or ending.
____________
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1902
ID: 352
Credit: 3,602,484,350
RAC: 5,414,201
                          
|
2) is there any way of telling the client to download N WUs, run them, upload the results and then exit? This would be equivalent to setting refill = 0 in the PG_LLRNetlinux_Batch24 client.
Would changing maxworkunits=n in prpclient.ini do?
____________
My stats
Badge score: 1*1 + 5*1 + 8*3 + 9*11 + 10*1 + 11*1 + 12*3 = 186 |
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1902
ID: 352
Credit: 3,602,484,350
RAC: 5,414,201
|
|
Run ~500 tests on Windows server, no problem so far....nor a prime...lol.
Lennart, how much tests do you want us to run?
And as Ward De Ridder asked for manual credit - should we set e-mail address same as the one on BOINC?
____________
My stats
Badge score: 1*1 + 5*1 + 8*3 + 9*11 + 10*1 + 11*1 + 12*3 = 186 |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
1) if you hit ctrl-C, the client stops ok, but it doesn't resume any already downloaded WUs when it next starts up. On one occasion, it left one of the work_PGPRPNet.* files behind (can't remember which one) and the client dumped core on startup until this file was removed. Don't have full details anymore. I'll try some more tests to get the exact behaviour explained fully.
I've been unable to reproduce this since. Every time it interrupts, it keeps all 3 work_PGPRPNet.* files and a zNNNNNNN cache, and restarts the downloaded work correctly. I guess this must have been a corner case where the llr process was just starting or ending.
This is one of the issues that has been resolved in later versions.
2) is there any way of telling the client to download N WUs, run them, upload the results and then exit? This would be equivalent to setting refill = 0 in the PG_LLRNetlinux_Batch24 client.
Would changing maxworkunits=n in prpclient.ini do?
The only way to do this now is to set maxworkunits to n, start client and let n WU's download, change maxworkunits to 0, then stop and restart client for changes to be applied. Once the last WU is completed, the client will upload and close.
Lennart, how much tests do you want us to run?
And as Ward De Ridder asked for manual credit - should we set e-mail address same as the one on BOINC?
Run as much as you like...it's all "live" data. As for manual credit, we can do this...but please remember it may take a while for the credit to be applied. It will be done in batches.
We think that with the latest version released, v1.0.3, that the system is stable enough to start moving production from LLRNet to PRPNet. Yes, there may still be some minor bugs that haven't shown themselves yet, but from what we've tested, things look good and stable.
We'll provide more information very soon.
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
New Port Information
Server: pgllr.mine.nu
Port: 6868 for the Generalized Cullen/Woodall Prime Search base=13
Port: 7171 for the Proth Prime Search (extended)
Port: 8181 for the Sophie Germain Prime Search
This is the information you'll provide in the prpclient.ini file. Actually, a nice feature is that you can participate in all the projects while running just one instance of the client. For example, you could list:
server=10:gcw13:pgllr.mine.nu:6868
server=60:pps:pgllr.mine.nu:7171
server=30:sg:pgllr.mine.nu:8181
This setup would provide 10% work from GCW13, 60% work from PPS, and 30% work from SG. Or you can make any combinations you like. If you only want to participate in PPS, then the numbers would be 0, 100, 0. Just make sure all numbers add up to 100. :)
You can copy the following over the current server information in your prpclient.ini file and adjust the numbers to what you like.
server=0:gcw13:pgllr.mine.nu:6868
server=0:pps:pgllr.mine.nu:7171
server=0:sg:pgllr.mine.nu:8181
If you are currently running LLRNet, please start making the transition to PRPNet. We'll be phasing out LLRNet very soon.
As always, please provide your feedback here.
____________
|
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 793
ID: 18447
Credit: 355,699,359
RAC: 1,449,978
|
|
New 1.0.3 build does not include prpclient_32 any more. Please can we continue to ship both 32 and 64 bit binaries?
Even the old 1.0.1 build depends on libstdc++.so, so won't run on lots of old systems. Please can we attempt to reduce the number of dynamic dependancies (especially modern ones) lest we curtail the number of platforms that can participate in this?
____________
|
|
|
|
|
New 1.0.3 build does not include prpclient_32 any more. Please can we continue to ship both 32 and 64 bit binaries?
Even the old 1.0.1 build depends on libstdc++.so, so won't run on lots of old systems. Please can we attempt to reduce the number of dynamic dependancies (especially modern ones) lest we curtail the number of platforms that can participate in this?
Ok i will.
/Lennart
|
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 23,566,369
RAC: 73,396
|
|
you placed the linux prpclient.ini in the windows directory,
llrexe=./sllr should be llrexe=clls.exe and the other one should also be changed.
____________
|
|
|
|
|
you placed the linux prpclient.ini in the windows directory,
llrexe=./sllr should be llrexe=clls.exe and the other one should also be changed.
Thanks Soon fixed
/Lennart
EDIT: Fixed |
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 23,566,369
RAC: 73,396
|
|
I used pspad when I changed the settings, but now I see that notepad (the one who is default to open .ini files) replaces the new line breakes with a square, I'd recommend you to use \n for a new line.
____________
|
|
|
|
|
I used pspad when I changed the settings, but now I see that notepad (the one who is default to open .ini files) replaces the new line breakes with a square, I'd recommend you to use \n for a new line.
I have change it to dos format
/Lennart |
|
|
|
|
|
John, could you clarify something for me please?
You have said that on the PRPNet we are processing 'live' data under PPS.
But with n at c. 350k these numbers are around 40k decimal digits shorter than the numbers I'm crunching via BOINC. The PPS End of year goals thread shows that we are near n=500k.
So are we just double checking?
Thanks and sorry if I've missed something.
Peter
____________
35 x 2^3587843+1 is prime! |
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 793
ID: 18447
Credit: 355,699,359
RAC: 1,449,978
|
|
But the k value is different - boinc is testing 5-1199 - prpnet is testing larger values, where the n value is just about in the top5000.
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
John, could you clarify something for me please?
You have said that on the PRPNet we are processing 'live' data under PPS.
But with n at c. 350k these numbers are around 40k decimal digits shorter than the numbers I'm crunching via BOINC. The PPS End of year goals thread shows that we are near n=500k.
So are we just double checking?
Thanks and sorry if I've missed something.
Peter
As noted in this post: New Port Information
"Proth Prime Search (extended)"
When the Proth Prime Search was first started back in Feb 2008, two sieves were created...one that matched the current ProthSearch range k<1200 and an extended one for k>1200. The "live" test data came from the "extended" sieve file. It's only being used for testing purposes since these were much shorter testing times than SG and GCW13.
BTW, Vato is correct. Current PPS in BOINC is testing odd k 5-1199. k=3 is part of the 321 Prime Search.
____________
|
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 793
ID: 18447
Credit: 355,699,359
RAC: 1,449,978
|
|
BTW Lennart
If possible, please can you do: chmod a+x prpclient prpclient_32 phrot phrot.p3-linux sllr prior to zipping up the next release - it'll save a few support questions down the line.
For anyone interested, I use this as prpclient.sh to autoselect the right Nbit client:
#!/bin/sh
UNAMEM=`uname -m`
if [ "x${UNAMEM}x" = "xx86_64x" ]
then
PRPCLIENT='./prpclient'
else
PRPCLIENT='./prpclient_32'
fi
exec ${PRPCLIENT} ${1:+"$@"}
____________
|
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 793
ID: 18447
Credit: 355,699,359
RAC: 1,449,978
|
|
Doesn't seem to like doing the SG tests. Stopped and exited at this point instead of returning the result.
385604505*2^666673-1, iteration : 660000 / 666673 [98.99%]. Time per iteration
385604505*2^666673-1 is not prime. LLR Res64: 416EFB8033C73D57 Time : 1436.480 sec.
[2009-01-04 01:02:09 GMT] sg: Could not find RES64 residue [lresults.txt]. Is llr broken?
And if I fire it up again, it starts retesting the same WU.
____________
|
|
|
|
|
Doesn't seem to like doing the SG tests. Stopped and exited at this point instead of returning the result.
385604505*2^666673-1, iteration : 660000 / 666673 [98.99%]. Time per iteration
385604505*2^666673-1 is not prime. LLR Res64: 416EFB8033C73D57 Time : 1436.480 sec.
[2009-01-04 01:02:09 GMT] sg: Could not find RES64 residue [lresults.txt]. Is llr broken?
And if I fire it up again, it starts retesting the same WU.
Can you tell me what you are running ?
Win 32/64 Linux 32/64 and what LLR you run ?
What client ?
Lennart
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
Doesn't seem to like doing the SG tests. Stopped and exited at this point instead of returning the result.
385604505*2^666673-1, iteration : 660000 / 666673 [98.99%]. Time per iteration
385604505*2^666673-1 is not prime. LLR Res64: 416EFB8033C73D57 Time : 1436.480 sec.
[2009-01-04 01:02:09 GMT] sg: Could not find RES64 residue [lresults.txt]. Is llr broken?
And if I fire it up again, it starts retesting the same WU.
This is confirmed several times. We are unable to resolve the problem right now.
ATTENTION
Please do not continue with SG WU's. Set project share to 0:
server=0:sg:pgllr.mine.nu:8181
and delete all work_sg.* files. We'll update when we know more.
We apologize for this.
____________
|
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 793
ID: 18447
Credit: 355,699,359
RAC: 1,449,978
|
|
latest PRPNetLinux 1.0.3 posted a few hours ago.
ubuntu 8.04 32bit on a P4 2.8GHz
____________
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1902
ID: 352
Credit: 3,602,484,350
RAC: 5,414,201
|
New 1.0.3 build does not include prpclient_32 any more. Please can we continue to ship both 32 and 64 bit binaries?
Ok i will.
On the other hand - can you provide Win 64-bit?
____________
My stats
Badge score: 1*1 + 5*1 + 8*3 + 9*11 + 10*1 + 11*1 + 12*3 = 186 |
|
|
|
|
|
How many candidates do we have on the prpnet server?
____________
|
|
|
|
|
|
I know this may sound a daft question. How do you run this over multiple cores and run it in the background (Slient, no window)?
Windows Vista 64bit
Many Thanks! |
|
|
|
|
|
I've been running in hidden mode under windows xp on my quad-core.
You need to create 4 separate folders and rename a few files.
I made this to follow the hidden clients on task manager.
1 - made multiple copies of the prpclientWindows client
2 - then I edited "prpclient.ini" and changed the following lines:
clientid=core0 (etc)
llrexe=cllr0.exe (etc)
3- then I renamed cllr.exe to cllr0.exe for core0, cllr1.exe fore core1, etc
4- on each folder I renamed prpclient.exe to prpclient0.exe, prpclient1.exe, prpclient2.exe, prpclient3.exe.
5-then I copy a program called HideItX.exe to each folder and created separate shortcuts to run the client in hidden mode:
C:\DC\prpclientWindows\core0\HideItX.exe -starthide C:\DC\prpclientWindows\core0\prpclient0.exe (for each core shortcut)
Here's a copy of my quad installation in case you didn't understand my explanation:
prpclientWindows.zip (link to download)
Of course the shortcuts will only work (files 0, 1, 2 and 3) if you have the client installed on c:\dc\prpclientWindows.
Hope that helps.
Carlos
____________
|
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1198
ID: 18646
Credit: 566,086,232
RAC: 939,121
|
New Port Information
Server: pgllr.mine.nu
Port: 6868 for the Generalized Cullen/Woodall Prime Search base=13
6868 seems not to run
[2009-01-04 14:31:37 GMT] pgllr.mine.nu:6868 connect to socket failed
anything to know about ???
EDIT:
now it seems to work again
[2009-01-04 14:53:48 GMT] gcw13: Getting work from server pgllr.mine.nu at port 6868
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
|
|
|
|
|
New Port Information
Server: pgllr.mine.nu
Port: 6868 for the Generalized Cullen/Woodall Prime Search base=13
6868 seems not to run
[2009-01-04 14:31:37 GMT] pgllr.mine.nu:6868 connect to socket failed
anything to know about ???
EDIT:
now it seems to work again
[2009-01-04 14:53:48 GMT] gcw13: Getting work from server pgllr.mine.nu at port 6868
Have the same problems
____________
 |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
How many candidates do we have on the prpnet server?
Currently in the server...
PPS - 50K WU's
SG - 940 WU's
GCW13 - 30K WU's
____________
|
|
|
|
|
How many candidates do we have on the prpnet server?
Currently in the server...
PPS - 50K WU's
SG - 940 WU's
GCW13 - 30K WU's
What's the range for PPS work?
Carlos
____________
|
|
|
|
|
How many candidates do we have on the prpnet server?
Currently in the server...
PPS - 50K WU's
SG - 940 WU's
GCW13 - 30K WU's
What's the range for PPS work?
Carlos
right now i have 1201<k<2999 350k<n<351k
/Lennart |
|
|
|
|
|
How many PPS candidates is the server getting per hour?
Sorry for the questions.
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
How many PPS candidates is the server getting per hour?
Sorry for the questions.
For PPS ~320/hr
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
ATTENTION
Please do not continue with SG WU's. Set project share to 0:
server=0:sg:pgllr.mine.nu:8181
and delete all work_sg.* files. We'll update when we know more.
We apologize for this.
This issue has been resolved. Please download the latest version 1.0.4.
http://pgllr.mine.nu/PRPNet/
There was a character limit that SG was running into. Because the k's for the other projects were relatively small, this did not affect them. The k's for SG are much larger.
On another note, we are experiencing odd issues with the percentages. This affects all three projects.
server=0:gcw13:pgllr.mine.nu:6868
server=0:pps:pgllr.mine.nu:7171
server=0:sg:pgllr.mine.nu:8181
One odd behaviour is when percentages are changed, the client does not respond to the changes. For example, when changed from 5, 90, 5 to 0, 0, 100, work should only come from SG but instead, work still comes from all three.
Please play around with this and let us know what you experience. NOTE: Remember your work files when making changes...allow time for the work to clear to note your experience.
Thank you for your help!
EDIT: If you're done testing the percentages and would just like to work on a single project, a work around is to just comment out the other two ports and set the one port you want work from to 100.
____________
|
|
|
|
|
How many PPS candidates is the server getting per hour?
Sorry for the questions.
For PPS ~320/hr
I though we were doing more! I'm making ~113/hr...we need more power in here.
BTW, don't forget to compile the windows 1.0.4 version.
____________
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1902
ID: 352
Credit: 3,602,484,350
RAC: 5,414,201
|
BTW, don't forget to compile the windows 1.0.4 version.
And Win x64 if there is performance benefit.
____________
My stats
Badge score: 1*1 + 5*1 + 8*3 + 9*11 + 10*1 + 11*1 + 12*3 = 186 |
|
|
|
|
BTW, don't forget to compile the windows 1.0.4 version.
And Win x64 if there is performance benefit.
The performance is gain by the llr client and not by the prpclient/server.
____________
|
|
|
|
|
How many candidates do we have on the prpnet server?
Currently in the server...
PPS - 50K WU's
SG - 940 WU's
GCW13 - 30K WU's
What's the range for PPS work?
Carlos
right now i have 1201<k<2999 350k<n<351k
/Lennart
I see some k's>2999, right?
____________
|
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 793
ID: 18447
Credit: 355,699,359
RAC: 1,449,978
|
|
Got a duplicate WU - did the server restart around this time?
[2009-01-04 18:54:53 GMT] pps: Getting work from server pgllr.mine.nu at port 7171
[2009-01-04 19:02:54 GMT] pps: 2051*2^350119+1 is not prime. Residue F3BBE3A9422B42F8 Time : 478
[2009-01-04 19:10:53 GMT] pps: 2289*2^350121+1 is not prime. Residue 8A9A1BC899F3456A Time : 479
[2009-01-04 19:10:53 GMT] Total Time: 15:50:53 Total Tests: 128 Total PRPs Found: 0
[2009-01-04 19:10:53 GMT] pps: Returning work to server pgllr.mine.nu at port 7171
[2009-01-04 19:10:53 GMT] pps: INFO: Server has no record of this test. The test result for 2051*2^350119+1 has been logged
[2009-01-04 19:10:54 GMT] pps: Getting work from server pgllr.mine.nu at port 7171
[2009-01-04 19:18:58 GMT] pps: 2289*2^350121+1 is not prime. Residue 8A9A1BC899F3456A Time : 478
[2009-01-04 19:26:58 GMT] pps: 2289*2^350123+1 is not prime. Residue BC4D59A914E41A3F Time : 479
____________
|
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 793
ID: 18447
Credit: 355,699,359
RAC: 1,449,978
|
|
I hate to be a party-pooper, but please could we have some consistency in naming each release?
The latest has just been announced, and following the link, I can't tell which file to download this time,
and it had changed plenty of times previously. As I write this, the filename mentioned in the top post isn't there either.
eg
prpclient_linux.zip isn't there
prpclient-1.0.4-linux-x86-64.zip is, but unzips into prpclient-1.0.3/
____________
|
|
|
|
|
I hate to be a party-pooper, but please could we have some consistency in naming each release?
The latest has just been announced, and following the link, I can't tell which file to download this time,
and it had changed plenty of times previously. As I write this, the filename mentioned in the top post isn't there either.
eg
prpclient_linux.zip isn't there
prpclient-1.0.4-linux-x86-64.zip is, but unzips into prpclient-1.0.3/
Sorry. Changed now.
We wait on new files.
/Lennart |
|
|
|
|
Got a duplicate WU - did the server restart around this time?
[2009-01-04 18:54:53 GMT] pps: Getting work from server pgllr.mine.nu at port 7171
[2009-01-04 19:02:54 GMT] pps: 2051*2^350119+1 is not prime. Residue F3BBE3A9422B42F8 Time : 478
[2009-01-04 19:10:53 GMT] pps: 2289*2^350121+1 is not prime. Residue 8A9A1BC899F3456A Time : 479
[2009-01-04 19:10:53 GMT] Total Time: 15:50:53 Total Tests: 128 Total PRPs Found: 0
[2009-01-04 19:10:53 GMT] pps: Returning work to server pgllr.mine.nu at port 7171
[2009-01-04 19:10:53 GMT] pps: INFO: Server has no record of this test. The test result for 2051*2^350119+1 has been logged
[2009-01-04 19:10:54 GMT] pps: Getting work from server pgllr.mine.nu at port 7171
[2009-01-04 19:18:58 GMT] pps: 2289*2^350121+1 is not prime. Residue 8A9A1BC899F3456A Time : 478
[2009-01-04 19:26:58 GMT] pps: 2289*2^350123+1 is not prime. Residue BC4D59A914E41A3F Time : 479
I shall investigate this more.
/Lennart |
|
|
|
|
|
Lennart,
How deep is the current PPS work on prp server sieved?
Carlos
____________
|
|
|
|
|
Lennart,
How deep is the current PPS work on prp server sieved?
Carlos
This is only to 1T but we shall not do this yet It was only to test the new PRPNet.
We can do the small n n=50k up to 300k later so we not lose them :)
/Lennart
|
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 793
ID: 18447
Credit: 355,699,359
RAC: 1,449,978
|
We wait on new files.
/Lennart
I'll upgrade as soon as 1.0.4 is ready.
Thanks!
____________
|
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1219
ID: 12001
Credit: 18,565,548
RAC: 0
|
Got a duplicate WU - did the server restart around this time?
[2009-01-04 18:54:53 GMT] pps: Getting work from server pgllr.mine.nu at port 7171
[2009-01-04 19:02:54 GMT] pps: 2051*2^350119+1 is not prime. Residue F3BBE3A9422B42F8 Time : 478
[2009-01-04 19:10:53 GMT] pps: 2289*2^350121+1 is not prime. Residue 8A9A1BC899F3456A Time : 479
[2009-01-04 19:10:53 GMT] Total Time: 15:50:53 Total Tests: 128 Total PRPs Found: 0
[2009-01-04 19:10:53 GMT] pps: Returning work to server pgllr.mine.nu at port 7171
[2009-01-04 19:10:53 GMT] pps: INFO: Server has no record of this test. The test result for 2051*2^350119+1 has been logged
[2009-01-04 19:10:54 GMT] pps: Getting work from server pgllr.mine.nu at port 7171
[2009-01-04 19:18:58 GMT] pps: 2289*2^350121+1 is not prime. Residue 8A9A1BC899F3456A Time : 478
[2009-01-04 19:26:58 GMT] pps: 2289*2^350123+1 is not prime. Residue BC4D59A914E41A3F Time : 479
In this case the client sent the same result twice to the server, probably because the first result responded with that INFO message. The server will do nothing with the duplicate since it will already have the result, but won't report it to the client. |
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 793
ID: 18447
Credit: 355,699,359
RAC: 1,449,978
|
|
My reading of this was that the server sent 2289*2^350121+1 twice (presumably because it reset to a slightly old db, or failed to write the last updates prior to crash, or somesuch), as well as forgetting about 2051*2^350119+1 - but you're closer to the code than I :-) I used to hack C a long time ago, but I'm a dinosaur these days.
____________
|
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1198
ID: 18646
Credit: 566,086,232
RAC: 939,121
|
|
a new prime
pps: 3015*2^350278+1 is prime! Time : 202.488 sec.
the prime made it into the TOP 5.000 (4988 at the moment)
http://primes.utm.edu/primes/page.php?id=86158
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
|
|
|
|
|
|
3081*2^350319+1 is prime!
http://primes.utm.edu/primes/page.php?id=86161
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
Here are the builds for the latest 1.0.4 release:
* prpclient-1.0.4-linux-x86
* prpclient-1.0.4-linux-x86-64
* prpclient-1.0.4-mingw32-x86 (EIDT: has issues...moved to "old" folder)
* Windows 64 bit not yet available
* MacPPC not yet available
* MacIntel not yet available
You can download the zip archives here:
http://pgllr.mine.nu/PRPNet/
Please provide feedback. :)
------------------------------------
Here are some BETA instructions:
1. Download the client for your OS and extract. If you are running multiple cores, then each core will need its own folder and full set of files.
2. Open the prpclient.ini and modify the following settings:
- email= Your email address. E-mail will be sent to this address when a number is found to be PRP.
- clientid= This identifies a specific instance of the client running for the specified e-mail.
- server= This points to PrimeGrid's PRPNet server from which the client will get work. It is already prepared with default values. Please make adjustments to your preference.
- maxworkunits= The maximum number of workunits that a client can request at a time. The server also has a limit, so the actual limit will be the less of this value and the value on the server. PrimeGrid defaults the value to 1 but you can change it to your preference.
Save the prpclient.ini file
3. If in Windows, just doubleclick the prpclient executable to run. If in Linux, just doubleclick the prpclient.bat file and select "Run in Terminal" if you want to watch.
4. The client will select which "primality testing program" to use on the WU downloaded. It will operate automatically until interrupted.
5. To stop client, hit ctrl-C.
Additional Information
Server Information (server=AA:BB:CC:DD)
AA is a percentage of work to get from that server
BB is a unique identifier appended to work file names (work_BB.in, work_BB.out, work_BB.save) created by the client for that server
CC is the server IP
DD is the server port
Server: pgllr.mine.nu
Port: 6868 for the Generalized Cullen/Woodall Prime Search base=13
Port: 7171 for the Proth Prime Search (extended)
Port: 8181 for the Sophie Germain Prime Search
Default Server Settings
server=0:gcw13:pgllr.mine.nu:6868
server=0:pps:pgllr.mine.nu:7171
server=100:sg:pgllr.mine.nu:8181
This tells the client to only get work from the Sophie Germain Prime Search. However, you can select other combinations such as 10, 60, 30. This would provide 10% work from GCW13, 60% work from PPS, and 30% work from SG. You can make any combinations you like. Just make sure all numbers add up to 100. :)
Finishing the queue and closing the client
To finish the WU's in cache, upload the results and close the client, change maxworkunits=0. Stop the client and restart for the new setting to take effect.
____________
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1902
ID: 352
Credit: 3,602,484,350
RAC: 5,414,201
|
|
No luck with 1.0.4 Win client.
D:\prp>prpclient-x86-mingw32.exe
Client will run in idle class.
[2009-01-06 07:16:53 GMT] PRPNet Client application v1.0.4 started
[2009-01-06 07:16:53 GMT] User email address is abc@xyz.com
And client will exit immediately on my main box (W2K3 x64, 8GB RAM).
Another machine (W2K3 x86) pops up "Visual Studio Just in-time debugger" and throws an unhandled win32 exception occurred [1304] in prpclient-x86-mingw32.exe
____________
My stats
Badge score: 1*1 + 5*1 + 8*3 + 9*11 + 10*1 + 11*1 + 12*3 = 186 |
|
|
|
|
|
Got the same error with the windows version. The client only runs one wu and then exits.
Please check the size of prpclient-x86-mingw32.exe file.
Carlos
____________
|
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 23,566,369
RAC: 73,396
|
|
Same problem here, deactivate read-only and it should work fine.
____________
|
|
|
|
|
|
I have same problems under Vista x64
____________
|
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 793
ID: 18447
Credit: 355,699,359
RAC: 1,449,978
|
|
prpclient-1.0.4-linux-x86 runs properly for SG now. Default file permissions are good now. I'm pretty happy with this release in behaviour. Only enhancement I'd like is static executables for prpnet (though that may have other knock-on effects).
Lennart - If you're interested, the following replacement prpclient.bat would allow you to merge the x86 and x86_64 distributions into a single zipfile which will discover the right executables at runtime. Of course, some folks might not like the extra size consumed. Even if not merged, it would allow you to ship the same prpclient.bat in both zipfiles and only have to maintain one going forward.
#!/bin/sh
ARCH=`uname -m`
OS=`uname -s | awk '{print $1}' | tr 'A-Z' 'a-z'`
PRPCLIENT=./prpclient-x86-linux
PRPCLIENTOTHER=./prpclient-${ARCH}-${OS}
if [ -x ${PRPCLIENTOTHER} ]
then
PRPCLIENT=${PRPCLIENTOTHER}
fi
exec ${PRPCLIENT} ${1:+"$@"}
____________
|
|
|
|
|
|
How many candidates did you test since the beginning of the beta test?
____________
|
|
|
|
|
How many candidates did you test since the beginning of the beta test?
There was about 50k test and that *2 double check is 100k
about 80% done.
/Lennart |
|
|
|
|
|
I new Windows Client can be downloaded here.
http://pgllr.mine.nu/PRPNet/
Release 1.0.5
I will soon add Linux 1.0.5
/Lennart
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1902
ID: 352
Credit: 3,602,484,350
RAC: 5,414,201
|
I new Windows Client can be downloaded here. I have been able to make 1.0.5 running with no problem.
EDIT: It seems to me that there is some redundant/duplicate info in screen output: not prime, Residue and Time are repeated on two lines.
This is the new PRPNet Server version 1.0.5
PrimeGrid PRP Test Server PPS
Starting Proth prime test of 2403*2^350888+1
Using FFT length 24K, a = 17
2403*2^350888+1 is not prime. Proth RES64: 1B2BC19051C36B5D Time : 164.898 sec.
[2009-01-07 16:26:44 GMT] pgllr: 2403*2^350888+1 is not prime. Residue 1B2BC19051C36B5D Time : 164
____________
My stats
Badge score: 1*1 + 5*1 + 8*3 + 9*11 + 10*1 + 11*1 + 12*3 = 186 |
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 793
ID: 18447
Credit: 355,699,359
RAC: 1,449,978
|
|
That is true.
The first line is from the llr program.
The second is from prpnet itself and reflects how it reports the result to the server.
So it's actually useful to see both lines imho.
____________
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1902
ID: 352
Credit: 3,602,484,350
RAC: 5,414,201
|
|
Thanks for explanation, Vato.
I got a prime and it barely made it to TOP5000 (4984).
http://primes.utm.edu/primes/page.php?id=86188
Do we need to push harder on this...?
____________
My stats
Badge score: 1*1 + 5*1 + 8*3 + 9*11 + 10*1 + 11*1 + 12*3 = 186 |
|
|
|
|
Thanks for explanation, Vato.
I got a prime and it barely made it to TOP5000 (4984).
http://primes.utm.edu/primes/page.php?id=86188
Do we need to push harder on this...?
We have done most off it and found some bugs.
Now we will sieve this range much deeper
Have to wait about a month before we can start searching top5000.
I will soon load smaller n numbers that we have todo.
We need to run port 7171 clean then i will start
low n numbers on that one to we get the file sieved.
Port 6868 SG will replace LLRNet SG
Port 8181 CWG13 will also replace LLRNet.
I have to thank all who helped us test this new client server platform.
It uses LLR 3.7.1 and Phrot and is much faster then LLRNet.
/Lennart |
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 793
ID: 18447
Credit: 355,699,359
RAC: 1,449,978
|
|
And thanks to you for doing your part too Lennart!
____________
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1902
ID: 352
Credit: 3,602,484,350
RAC: 5,414,201
|
We have done most off it and found some bugs.
Now we will sieve this range much deeper
Have to wait about a month before we can start searching top5000.
I will soon load smaller n numbers that we have todo.
OK, I'll keep 3 cores running of this test for a while longer.
Will start PPS2 Sieving once the sieve file is downloaded (it will takes hours so I better go sleep) and let 12 cores doing their job for a start. Later I may add another cores running x64 for 24/7.
____________
My stats
Badge score: 1*1 + 5*1 + 8*3 + 9*11 + 10*1 + 11*1 + 12*3 = 186 |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
Mac testers needed!!! :)
We are seeking Mac users to help test the MacPPC and MacIntel versions of the prpclient.
The file (prpclient-1.0.5-mac.zip) can be downloaded here: http://pgllr.mine.nu/PRPNet/
A few notes:
- both executables are in the same file. You select which one to run on your computer.
- edit the prpclient.ini file and provide the following info:
email=
clientid=
- MacPPC users should comment out the llrexe=./mllr-3.7.1c line as it is only available for MacIntel.
Once everything is running fine, if possible, please test the following:
- so we can test phrot on the MacIntel version, please adjust project share to the following:
server=50:gcw13:pgllr.mine.nu:6868
server=50:pps:pgllr.mine.nu:7171
server=0:sg:pgllr.mine.nu:8181
- For phrot, the static version is defaulted in the prplient.ini file. Please also test the non-static version by modifying the ini file.
If you have any questions, please post here or PM us. Thanks for your help!
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
Up to now, there has been a lot of alpha/beta testing taking place. We are moving closer to "production" and have a set of "pre-production" files available.
Please, at your convenience, download the new archives at http://pgllr.mine.nu/PRPNet/:
prpclient-1.0.5-linux.zip
prpclient-1.0.5-windows.zip
The Windows version should work on both 32 bit and 64 bit OS.
The Linux version contains both 32 bit and 64 bit executables. In addition, it contains static and not static versions of the client and of phrot. Be sure to run a few tests on the GCW13 server so phrot can be tested.
Please put these through the "wringer"!!!
Thank You!
EDIT: Phrot builds are based on the 0.63 version. It's stable but has a residue issue. We are awaiting the 0.64 build. You can still test with the current version.
Here are the builds for the latest 1.0.5 release:
* prpclient-1.0.5-linux (32 bit and 64 bit)
* prpclient-1.0.5-windows (32 bit - will run on 64 bit)
* prpclient-1.0.5-mac (PPC and Intel)
You can download the zip archives here:
http://pgllr.mine.nu/PRPNet/
Please provide feedback. :)
------------------------------------
Here are some BETA instructions:
1. Download the client for your OS and extract. If you are running multiple cores, then each core will need its own folder and full set of files.
2. Open the prpclient.ini and modify the following settings:
- email= Your email address. E-mail will be sent to this address when a number is found to be PRP.
- clientid= This identifies a specific instance of the client running for the specified e-mail.
- server= This points to PrimeGrid's PRPNet server from which the client will get work. It is already prepared with default values. Please make adjustments to your preference.
- maxworkunits= The maximum number of workunits that a client can request at a time. The server also has a limit, so the actual limit will be the less of this value and the value on the server. PrimeGrid defaults the value to 1 but you can change it to your preference.
Save the prpclient.ini file
3. If in Windows, just doubleclick the prpclient executable to run. If in Linux, just doubleclick the prpclient.bat file and select "Run in Terminal" if you want to watch. NOTE: Linux users, be sure to edit the prpclient.bat file to reflect the proper executable for your system. Mac users will have to make sure they select the proper executable as well.
4. The client will start, communicate with the server, and download WU's based on your server settings in the prpclient.ini file.
4. The client will automatcially select which "primality testing program" (LLR or Phrot) to use on the downloaded WU's. It will continue until interrupted.
5. To stop client, hit ctrl-C.
Additional Information
Server Information (server=AA:BB:CC:DD)
AA is a percentage of work to get from that server
BB is a unique identifier appended to work file names (work_BB.in, work_BB.out, work_BB.save) created by the client for that server
CC is the server IP
DD is the server port
Server: pgllr.mine.nu
Port: 6868 for the Generalized Cullen/Woodall Prime Search base=13
Port: 7171 for the Proth Prime Search (extended)
Port: 8181 for the Sophie Germain Prime Search
Default Server Settings
server=0:gcw13:pgllr.mine.nu:6868
server=0:pps:pgllr.mine.nu:7171
server=100:sg:pgllr.mine.nu:8181
This tells the client to only get work from the Sophie Germain Prime Search. However, you can select other combinations such as 10, 60, 30. This would provide 10% work from GCW13, 60% work from PPS, and 30% work from SG. You can make any combinations you like. Just make sure all numbers add up to 100. :)
NOTE: Using a percentage of 0 will get work from that server only if the client could not connect to any other servers. Therefore, if you don't want work from a particular server, then comment it out and adjust the percentages over the remaining servers.
Finishing the queue and closing the client
To finish the WU's in cache, upload the results and close the client, change maxworkunits=0. Stop the client and restart for the new setting to take effect.
____________
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1902
ID: 352
Credit: 3,602,484,350
RAC: 5,414,201
|
|
I was able to get latest 1.0.5 running both on Win 32 and Win x64 with no problems.
I'm a bit unclear of what projects (servers) are prefered to run.
And, btw, I'm starting to get "no work" from PPS.
____________
My stats
Badge score: 1*1 + 5*1 + 8*3 + 9*11 + 10*1 + 11*1 + 12*3 = 186 |
|
|
|
|
|
no work for Proth Prime Search ?
[2009-01-10 10:08:42 GMT] c: INFO: No available candidates are left on this server.
[2009-01-10 10:08:43 GMT] c: Getting work from server pgllr.mine.nu at port 7171
[2009-01-10 10:08:48 GMT] c: INFO: No available candidates are left on this server.
[2009-01-10 10:08:48 GMT] c: Getting work from server pgllr.mine.nu at port 7171
[edit]
now it works
____________
|
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 793
ID: 18447
Credit: 355,699,359
RAC: 1,449,978
|
|
latest 1.0.5 linux runs fine - but the permissions of all the programs in the zipfile are missing the execute bit again.
____________
|
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1219
ID: 12001
Credit: 18,565,548
RAC: 0
|
Mac testers needed!!! :)
We are seeking Mac users to help test the MacPPC and MacIntel versions of the prpclient.
The file (prpclient-1.0.5-mac.zip) can be downloaded here: http://pgllr.mine.nu/PRPNet/
A few notes:
- both executables are in the same file. You select which one to run on your computer.
- edit the prpclient.ini file and provide the following info:
email=
clientid=
- MacPPC users should comment out the llrexe=./mllr-3.7.1c line as it is only available for MacIntel.
Once everything is running fine, if possible, please test the following:
- so we can test phrot on the MacIntel version, please adjust project share to the following:
server=50:gcw13:pgllr.mine.nu:6868
server=50:pps:pgllr.mine.nu:7171
server=0:sg:pgllr.mine.nu:8181
- For phrot, the static version is defaulted in the prplient.ini file. Please also test the non-static version by modifying the ini file.
If you have any questions, please post here or PM us. Thanks for your help!
I've been running the PPC version (my own build) against GCW13 without a problem. As soon as some resources free up, I will run the MacIntel build for a short wile. |
|
|
|
|
|
Hello all .
The test range 1200<k<4000 350k<n<351k is done.
Now i have loaded 1201<k<10000k 45k<n50k
This is very short number to test.
so set "maxworkunits=10" or more in prpclient.ini
You have to restart client when change is done.
This is also a test to see if it works with those short numbers and heavy traffic.
Thanks for your help.
/Lennart
EDIT: Port 7171 1200<k<10000 45000<n<50000 base=2
Port 6868 SG will replace LLRNet SG n=666666 -666676 base=2
Port 8181 CWG13 150k<n<1M base =13 |
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 23,566,369
RAC: 73,396
|
|
I set maxworkunit to 15 but I only get 10 ..
I also did some tests with this small numbers calculating them with proth and everythings works fine, but it is 4 times slower..
(port 7171)
____________
|
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 23,566,369
RAC: 73,396
|
|
lennart, troubles:
http://img176.imageshack.us/img176/935/screenshot002so8.jpg
Only pps has 100%, the others 0%
____________
|
|
|
|
|
lennart, troubles:
http://img176.imageshack.us/img176/935/screenshot002so8.jpg
Only pps has 100%, the others 0%
Yes i have seen that before.
Using a percentage
// of 0 will get work from that server only if the client could not
// connect to any other servers. Please read the prpnet_servers.txt
// in this directory for information on some known PRPNet servers.
You have to comment other servers out if you don't want any work from them anytime
I did some maintain and had to restart 7171.
It should be up in 5 min.
/Lennart |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
latest 1.0.5 linux runs fine - but the permissions of all the programs in the zipfile are missing the execute bit again.
Thanks! It will be corrected...again. :(
____________
|
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 23,566,369
RAC: 73,396
|
|
not working yet
____________
|
|
|
|
|
not working yet
Soon :)
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
From the prpclient.ini file:
Using a percentage of 0 will get work from that server only if the client could not connect to any other servers.
Therefore, it's working as intended. If you don't want this to happen, then comment out the servers you don't want work from.
____________
|
|
|
|
|
|
6441*2^45003+1 is prime! Time : 1.759 sec.
really short
____________
|
|
|
|
|
6441*2^45003+1 is prime! Time : 1.759 sec.
really short
set "maxworkunits=40" in prpclient.ini
Tell me how it works :)
/Lennart |
|
|
|
|
|
thank you
it`s a little better
[2009-01-10 15:39:42 GMT] c: Getting work from server pgllr.mine.nu at port 7171
[2009-01-10 15:39:51 GMT] c: 2499*2^45007+1 is not prime. Residue E7F508511CE47633 Time : 2
[2009-01-10 15:39:54 GMT] c: 2505*2^45007+1 is not prime. Residue C197FAAB42771480 Time : 2
[2009-01-10 15:39:58 GMT] c: 2525*2^45007+1 is not prime. Residue B46897286E761C0C Time : 2
[2009-01-10 15:40:02 GMT] c: 2625*2^45007+1 is not prime. Residue BD8268FA6CCC21B9 Time : 3
[2009-01-10 15:40:07 GMT] c: 2657*2^45007+1 is not prime. Residue C0427EC54476B4C9 Time : 3
[2009-01-10 15:40:10 GMT] c: 2669*2^45007+1 is not prime. Residue F582B6480DF00617 Time : 2
[2009-01-10 15:40:13 GMT] c: 2679*2^45007+1 is not prime. Residue E1B21E75C3339E95 Time : 2
[2009-01-10 15:40:16 GMT] c: 2681*2^45007+1 is not prime. Residue 0FB8616ECC0F854A Time : 2
[2009-01-10 15:40:20 GMT] c: 2745*2^45007+1 is not prime. Residue 527B114F63AE354B Time : 2
[2009-01-10 15:40:23 GMT] c: 2765*2^45007+1 is not prime. Residue 4AF833C15B0C1E0D Time : 2
[2009-01-10 15:40:27 GMT] c: 2895*2^45007+1 is not prime. Residue 1FF5D36BB92A70A5 Time : 3
[2009-01-10 15:40:31 GMT] c: 2927*2^45007+1 is not prime. Residue 665341F5EEB58077 Time : 2
[2009-01-10 15:40:34 GMT] c: 2991*2^45007+1 is not prime. Residue 082FDB2C6BF6ADF5 Time : 2
[2009-01-10 15:40:37 GMT] c: 3071*2^45007+1 is not prime. Residue 89B7E19EF43BE3F5 Time : 2
[2009-01-10 15:40:40 GMT] c: 3089*2^45007+1 is not prime. Residue ED9BF1BEFFFF9901 Time : 3
[2009-01-10 15:40:43 GMT] c: 3129*2^45007+1 is not prime. Residue F574A0D38BB30808 Time : 2
[2009-01-10 15:40:48 GMT] c: 3137*2^45007+1 is not prime. Residue BBED51E294C54086 Time : 4
[2009-01-10 15:40:51 GMT] c: 3147*2^45007+1 is not prime. Residue 7B5CD341EFDBAE4A Time : 2
[2009-01-10 15:40:52 GMT] c: 3197*2^45007+1 is not prime. Residue ECD104C845056797 Time : 1
[2009-01-10 15:40:54 GMT] c: 3201*2^45007+1 is not prime. Residue C8116D71A1BC728E Time : 1
[2009-01-10 15:40:56 GMT] c: 3221*2^45007+1 is not prime. Residue B78EE71A147DAD7F Time : 1
[2009-01-10 15:40:58 GMT] c: 3399*2^45007+1 is not prime. Residue 0B28D3F76A0EB811 Time : 1
[2009-01-10 15:41:00 GMT] c: 3455*2^45007+1 is not prime. Residue FECF493328BC5EC3 Time : 1
[2009-01-10 15:41:02 GMT] c: 3485*2^45007+1 is not prime. Residue 212F1171A526F806 Time : 1
[2009-01-10 15:41:04 GMT] c: 1201*2^45008+1 is not prime. Residue B7ACD9F0763BBBD6 Time : 1
[2009-01-10 15:41:07 GMT] c: 1213*2^45008+1 is not prime. Residue E260B1FF2606DEE1 Time : 2
[2009-01-10 15:41:10 GMT] c: 1273*2^45008+1 is not prime. Residue 66FE1CD6A607C86F Time : 3
[2009-01-10 15:41:13 GMT] c: 1315*2^45008+1 is not prime. Residue 695329DA2EC15C87 Time : 2
[2009-01-10 15:41:15 GMT] c: 1317*2^45008+1 is not prime. Residue 76010D1EB3B0D38E Time : 2
[2009-01-10 15:41:19 GMT] c: 1395*2^45008+1 is not prime. Residue 5F37068CAA565D84 Time : 3
[2009-01-10 15:41:22 GMT] c: 1507*2^45008+1 is not prime. Residue 48A7E5BB17ECCFD4 Time : 2
[2009-01-10 15:41:26 GMT] c: 1537*2^45008+1 is not prime. Residue D247337835C05CF2 Time : 3
[2009-01-10 15:41:29 GMT] c: 1593*2^45008+1 is not prime. Residue 6798CC653FC1C6F3 Time : 2
[2009-01-10 15:41:34 GMT] c: 1611*2^45008+1 is not prime. Residue 93FA05BEDF6B46C6 Time : 3
[2009-01-10 15:41:38 GMT] c: 1705*2^45008+1 is not prime. Residue D80D91A44B47A29B Time : 3
[2009-01-10 15:41:42 GMT] c: 1735*2^45008+1 is not prime. Residue 70E108572B2A7E1D Time : 2
[2009-01-10 15:41:46 GMT] c: 1743*2^45008+1 is not prime. Residue 3BFBF2BDAEF8AB41 Time : 2
[2009-01-10 15:41:49 GMT] c: 6987*2^45006+1 is not prime. Residue 35A8A5DA1B69E25D Time : 2
[2009-01-10 15:41:52 GMT] c: 7015*2^45006+1 is not prime. Residue ACE1C8C3C289B98E Time : 3
[2009-01-10 15:41:55 GMT] c: 7045*2^45006+1 is not prime. Residue 9E367BA226CAB979 Time : 2
[2009-01-10 15:41:55 GMT] Total Time: 0:02:21 Total Tests: 40 Total PRPs Found: 0
[2009-01-10 15:41:57 GMT] c: Returning work to server pgllr.mine.nu at port 7171
[2009-01-10 15:42:00 GMT] c: INFO: All 40 test results were accepted
____________
|
|
|
|
|
|
I have sth to ask.
It doesn't accept anything which I report few hrs later?
I mean when I pause the prp and start it again, it said that the server reject my result. |
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1902
ID: 352
Credit: 3,602,484,350
RAC: 5,414,201
|
EDIT: Port 7171 1200<k<10000 45000<n<50000 base=2 This makes 44 milions tasks to fly through?
This log doesn't seem OK.
Returned batch of 40 task, downloaded new 40. Next tasks is a prime. But log says that all 40 tasks were returned (?).
Is this just a cosmetic bug or when a prime is found rest of tasks in cache are returned?
[2009-01-10 16:12:12 GMT] Total Time: 0:09:38 Total Tests: 240 Total PRPs Found: 0
[2009-01-10 16:12:12 GMT] pps: Returning work to server pgllr.mine.nu at port 7171
[2009-01-10 16:12:16 GMT] pps: INFO: All 40 test results were accepted
[2009-01-10 16:12:17 GMT] pps: Getting work from server pgllr.mine.nu at port 7171
[2009-01-10 16:12:20 GMT] pps: INFO: Server has a limit of 40 work units.
[2009-01-10 16:12:23 GMT] pps: 6315*2^45037+1 is Prime!
[2009-01-10 16:12:23 GMT] Total Time: 0:09:49 Total Tests: 241 Total PRPs Found: 1
[2009-01-10 16:12:23 GMT] pps: Returning work to server pgllr.mine.nu at port 7171
[2009-01-10 16:12:29 GMT] pps: INFO: All 40 test results were accepted
[2009-01-10 16:12:29 GMT] pps: Getting work from server pgllr.mine.nu at port 7171
[2009-01-10 16:12:30 GMT] pps: INFO: Server has a limit of 40 work units.
[2009-01-10 16:12:33 GMT] pps: 1729*2^45038+1 is not prime. Residue 5C11FA8071C9AAC6 Time : 2
____________
My stats
Badge score: 1*1 + 5*1 + 8*3 + 9*11 + 10*1 + 11*1 + 12*3 = 186 |
|
|
|
|
EDIT: Port 7171 1200<k<10000 45000<n<50000 base=2 This makes 44 milions tasks to fly through?
This log doesn't seem OK.
Returned batch of 40 task, downloaded new 40. Next tasks is a prime. But log says that all 40 tasks were returned (?).
Is this just a cosmetic bug or when a prime is found rest of tasks in cache are returned?
[2009-01-10 16:12:12 GMT] Total Time: 0:09:38 Total Tests: 240 Total PRPs Found: 0
[2009-01-10 16:12:12 GMT] pps: Returning work to server pgllr.mine.nu at port 7171
[2009-01-10 16:12:16 GMT] pps: INFO: All 40 test results were accepted
[2009-01-10 16:12:17 GMT] pps: Getting work from server pgllr.mine.nu at port 7171
[2009-01-10 16:12:20 GMT] pps: INFO: Server has a limit of 40 work units.
[2009-01-10 16:12:23 GMT] pps: 6315*2^45037+1 is Prime!
[2009-01-10 16:12:23 GMT] Total Time: 0:09:49 Total Tests: 241 Total PRPs Found: 1
[2009-01-10 16:12:23 GMT] pps: Returning work to server pgllr.mine.nu at port 7171
[2009-01-10 16:12:29 GMT] pps: INFO: All 40 test results were accepted
[2009-01-10 16:12:29 GMT] pps: Getting work from server pgllr.mine.nu at port 7171
[2009-01-10 16:12:30 GMT] pps: INFO: Server has a limit of 40 work units.
[2009-01-10 16:12:33 GMT] pps: 1729*2^45038+1 is not prime. Residue 5C11FA8071C9AAC6 Time : 2
It is not a bug. It is programed to stop and send back all when it finds a prime. We know about it and it shall be changed in next version.
/Lennart |
|
|
|
|
I have sth to ask.
It doesn't accept anything which I report few hrs later?
I mean when I pause the prp and start it again, it said that the server reject my result.
Could it be that you had old wu from last batch. n=350k-351k
and now got the short one n=45k-50k
/Lennart |
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 793
ID: 18447
Credit: 355,699,359
RAC: 1,449,978
|
|
Looks mostly OK - but found a odd one.
[EDIT - ignore - had been reported above]
When it found a prime which was the 5th test in a batch of 10, it immediatelty reported back, and did not run the other 5 tests that had been downloaded. I would have expected it to either complete the whole batch of 10 and then report, or report immediately and then complete the rest of the batch prior to downloading any more work.
...
[2009-01-10 16:27:39 GMT] pps: Getting work from server pgllr.mine.nu at port 7171
[2009-01-10 16:27:45 GMT] pps: 2775*2^45059+1 is not prime. Residue 70E066199A4EDBC7 Time : 4
[2009-01-10 16:27:49 GMT] pps: 2829*2^45059+1 is not prime. Residue 01F85E235D30C875 Time : 4
[2009-01-10 16:27:54 GMT] pps: 2915*2^45059+1 is not prime. Residue 7FD39679AF9E51D0 Time : 4
[2009-01-10 16:27:58 GMT] pps: 2937*2^45059+1 is not prime. Residue C9567544B62D64AE Time : 4
[2009-01-10 16:28:02 GMT] pps: 2967*2^45059+1 is not prime. Residue 433815D3EEB7A4AA Time : 3
[2009-01-10 16:28:06 GMT] pps: 2969*2^45059+1 is not prime. Residue 73C61A009E1BD19C Time : 3
[2009-01-10 16:28:09 GMT] pps: 3095*2^45059+1 is not prime. Residue 4811DFA787D97F32 Time : 3
[2009-01-10 16:28:13 GMT] pps: 3149*2^45059+1 is not prime. Residue 185AE2530FB5A9EE Time : 3
[2009-01-10 16:28:17 GMT] pps: 3161*2^45059+1 is not prime. Residue F69E0C7520D25403 Time : 3
[2009-01-10 16:28:21 GMT] pps: 3245*2^45059+1 is not prime. Residue 90E8EA1CBD86C8D5 Time : 3
[2009-01-10 16:28:21 GMT] Total Time: 0:06:37 Total Tests: 90 Total PRPs Found: 0
[2009-01-10 16:28:21 GMT] pps: Returning work to server pgllr.mine.nu at port 7171
[2009-01-10 16:28:23 GMT] pps: INFO: All 10 test results were accepted
[2009-01-10 16:28:23 GMT] pps: Getting work from server pgllr.mine.nu at port 7171
[2009-01-10 16:28:31 GMT] pps: 1257*2^45060+1 is not prime. Residue 400D8B40DB2B06F3 Time : 3
[2009-01-10 16:28:35 GMT] pps: 1261*2^45060+1 is not prime. Residue E6663E7FAD3DC481 Time : 3
[2009-01-10 16:28:39 GMT] pps: 1303*2^45060+1 is not prime. Residue D5ACD97064DB6D9C Time : 3
[2009-01-10 16:28:42 GMT] pps: 1335*2^45060+1 is not prime. Residue ADE74EA05FE85041 Time : 3
[2009-01-10 16:28:46 GMT] pps: 1401*2^45060+1 is Prime!
[2009-01-10 16:28:46 GMT] Total Time: 0:07:02 Total Tests: 95 Total PRPs Found: 1
[2009-01-10 16:28:46 GMT] pps: Returning work to server pgllr.mine.nu at port 7171
[2009-01-10 16:28:55 GMT] pps: INFO: All 10 test results were accepted
[2009-01-10 16:28:55 GMT] pps: Getting work from server pgllr.mine.nu at port 7171
[2009-01-10 16:29:02 GMT] pps: 6015*2^45002+1 is not prime. Residue 1485158F0F770A1D Time : 3
[2009-01-10 16:29:05 GMT] pps: 6103*2^45002+1 is not prime. Residue 2580EFCE742D8DB1 Time : 3
[2009-01-10 16:29:09 GMT] pps: 6117*2^45002+1 is not prime. Residue E747CA29C2A75D76 Time : 3
[2009-01-10 16:29:13 GMT] pps: 6255*2^45002+1 is not prime. Residue 8D612A31B5830CEB Time : 3
[2009-01-10 16:29:17 GMT] pps: 6387*2^45002+1 is not prime. Residue 4928B04F1ECEA92A Time : 4
[2009-01-10 16:29:21 GMT] pps: 5971*2^45060+1 is not prime. Residue 03B21B2D5AB9F8FE Time : 4
[2009-01-10 16:29:25 GMT] pps: 6081*2^45060+1 is not prime. Residue A98143667D390CE7 Time : 3
[2009-01-10 16:29:29 GMT] pps: 6123*2^45060+1 is not prime. Residue 9CFCE8F69335081C Time : 3
[2009-01-10 16:29:32 GMT] pps: 6127*2^45060+1 is not prime. Residue A9CB98FB003FB929 Time : 3
[2009-01-10 16:29:36 GMT] pps: 6141*2^45060+1 is not prime. Residue 75EE49D50F8B9D20 Time : 3
[2009-01-10 16:29:36 GMT] Total Time: 0:07:52 Total Tests: 105 Total PRPs Found: 1
[2009-01-10 16:29:37 GMT] pps: Returning work to server pgllr.mine.nu at port 7171
[2009-01-10 16:29:42 GMT] pps: INFO: All 10 test results were accepted
[2009-01-10 16:29:45 GMT] pps: Getting work from server pgllr.mine.nu at port 7171
[2009-01-10 16:29:52 GMT] pps: 8275*2^45060+1 is not prime. Residue 174E2F2E7D6D467E Time : 3
...
____________
|
|
|
|
|
|
"Found" a couple of little un's at last:
pps: 8023*2^45116+1 is Prime!
pps: 5375*2^45223+1 is Prime!
Running 1.0.5 in Vista64.
About 1.9s per WU on Q9550 2.8GHz
Running 1.0.5 in XPPro32.
About 8.1s per WU on Pentium M 1.6GHz
Getting:
[2009-01-10 19:45:30 GMT] pgllr.mine.nu:7171 connect to socket failed
[2009-01-10 19:45:30 GMT] Could not connect to any servers and no work is pending. Pausing 5 minutes
Do you want me to continue or are you happy with the windows client now?
____________
35 x 2^3587843+1 is prime! |
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1902
ID: 352
Credit: 3,602,484,350
RAC: 5,414,201
|
|
[2009-01-10 20:02:55 GMT] pgllr.mine.nu:7171 connect to socket failed
There may be some work in progress.
In the meanwhile, client downloaded some SG tests to keep cores busy.
Now back in bussiness with PPS tests...
____________
My stats
Badge score: 1*1 + 5*1 + 8*3 + 9*11 + 10*1 + 11*1 + 12*3 = 186 |
|
|
|
|
[2009-01-10 20:02:55 GMT] pgllr.mine.nu:7171 connect to socket failed
There may be some work in progress.
In the meanwhile, client downloaded some SG tests to keep cores busy.
Now back in bussiness with PPS tests...
Had to restart the computer shall work now.
/Lennart
|
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 23,566,369
RAC: 73,396
|
|
I found a prime :D
First prime I find with my asus eee
8397*2^45330+1
____________
|
|
|
|
|
|
One here!
9911*2^45619+1 is Prime! |
|
|
|
|
|
The client is abandoning all work when it gets restarted. If you have completed work it will send that in but drops all partially finished or non started work.
I let the client run and finish a couple tests then closed the window with ctrl-c. Waited a few seconds and restarted prpclient-1.0.5.exe. It started up and reported the completed work but reported the rest abandoned. It then did not restart the test that was 1/2 completed when I closed the client. The LLR client is leaving behind the Z files though.
On these short tests it is not too bad but on the ones that take 36+ hours to complete there will be a lot of partially finished tests that get thrown out.
S. |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
The client is abandoning all work when it gets restarted. If you have completed work it will send that in but drops all partially finished or non started work.
I let the client run and finish a couple tests then closed the window with ctrl-c. Waited a few seconds and restarted prpclient-1.0.5.exe. It started up and reported the completed work but reported the rest abandoned. It then did not restart the test that was 1/2 completed when I closed the client. The LLR client is leaving behind the Z files though.
On these short tests it is not too bad but on the ones that take 36+ hours to complete there will be a lot of partially finished tests that get thrown out.
S.
This behavior has been confirmed and noted. Thanks!
____________
|
|
|
|
|
|
no work available
[2009-01-11 20:01:45 GMT] pgllr.mine.nu:7171 connect to socket failed
[2009-01-11 20:01:45 GMT] Could not connect to any servers and no work is pending. Pausing 5 minutes
[2009-01-11 20:06:18 GMT] pgllr.mine.nu:7171 connect to socket failed
[2009-01-11 20:06:18 GMT] Could not connect to any servers and no work is pending. Pausing 5 minutes
[2009-01-11 20:10:55 GMT] pgllr.mine.nu:7171 connect to socket failed
[2009-01-11 20:10:55 GMT] Could not connect to any servers and no work is pending. Pausing 5 minutes
[2009-01-11 20:15:31 GMT] pgllr.mine.nu:7171 connect to socket failed
[2009-01-11 20:15:31 GMT] Could not connect to any servers and no work is pending. Pausing 5 minutes
[2009-01-11 20:20:58 GMT] pgllr.mine.nu:7171 connect to socket failed
[2009-01-11 20:20:58 GMT] Could not connect to any servers and no work is pending. Pausing 5 minutes
____________
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1902
ID: 352
Credit: 3,602,484,350
RAC: 5,414,201
|
|
Found another prime
[2009-01-12 06:21:51 GMT] pps: 2683*2^351092+1 is Prime!
I guess it's now enough to enter Top5000.
Previous one already slipped from Top5000
http://primes.utm.edu/primes/page.php?id=86188
Other then that, workflow is fine right now...
____________
My stats
Badge score: 1*1 + 5*1 + 8*3 + 9*11 + 10*1 + 11*1 + 12*3 = 186 |
|
|
|
|
Found another prime
[2009-01-12 06:21:51 GMT] pps: 2683*2^351092+1 is Prime!
I guess it's now enough to enter Top5000.
Previous one already slipped from Top5000
http://primes.utm.edu/primes/page.php?id=86188
Other then that, workflow is fine right now...
Sorry To small :)
/Lennart |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
Phrot 0.64 has been released. All the archives have been updated to use this new version. You can download at:
http://pgllr.mine.nu/PRPNet/
If all you want to do is to replace your phrot file, you can download the bin here and extract the build that is appropriate for your OS.
http://pgllr.mine.nu/PRPNet/phrot/
Make sure you edit your prpclient.ini file to point to the new version. It's the line that starts with:
phrotexe=
Please post here if you encounter any issues. Phrot is used to test non base 2 numbers so it will only affect those testing on the GCW13 server. Also, MacPPC users only use phrot so it will affect them as well.
Phrot 0.64 - Januray 8, 2009
1) Fixed a bug where the incorrect residue was calculated for Proth when the candidate is composite.
2) Fixed the printing of the residue for MinGW.
____________
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1902
ID: 352
Credit: 3,602,484,350
RAC: 5,414,201
|
|
Running Phrot 0.64 just fine but on pps under Win.
Lennart, any updates on upcoming ranges? Just curious...
I've couple cores doing PPS 24/7.
Not expecting much of primes, perhaps some manual credit :-)
____________
My stats
Badge score: 1*1 + 5*1 + 8*3 + 9*11 + 10*1 + 11*1 + 12*3 = 186 |
|
|
|
|
Running Phrot 0.64 just fine but on pps under Win.
Lennart, any updates on upcoming ranges? Just curious...
I've couple cores doing PPS 24/7.
Not expecting much of primes, perhaps some manual credit :-)
You will soon get more info. :)
/Lennart
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1902
ID: 352
Credit: 3,602,484,350
RAC: 5,414,201
|
|
I let latest app run on 3-4 cores 24/7 for several days, completed several thousands of tests with new version and no primes at all.
Bad luck?
Any updates on progress?
____________
My stats
Badge score: 1*1 + 5*1 + 8*3 + 9*11 + 10*1 + 11*1 + 12*3 = 186 |
|
|
|
|
I let latest app run on 3-4 cores 24/7 for several days, completed several thousands of tests with new version and no primes at all.
Bad luck?
Any updates on progress?
[2009-01-11 07:55:08 GMT] 2221*2^351012+1: Prime returned by pschoefer
[2009-01-12 06:21:52 GMT] 2683*2^351092+1: Prime returned by Honza
[2009-01-13 11:32:03 GMT] 3985*2^351168+1: Prime returned by Honza
[2009-01-16 03:37:51 GMT] 3627*2^351347+1: Prime returned by Sysadm@Nbg
[2009-01-19 13:29:35 GMT] 3591*2^351552+1: Prime returned by pschoefer
This is what we found on the last batch. I think 2 was into Top 5000 but is now below 5000.
We will run this batch up to n= 352000 then we will insert ranges better sieved and higher n so we can submitt it to top5000 list :)
/Lennart |
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1902
ID: 352
Credit: 3,602,484,350
RAC: 5,414,201
|
|
Looks like we will be up to n=352000 in cca 2 days...
____________
My stats
Badge score: 1*1 + 5*1 + 8*3 + 9*11 + 10*1 + 11*1 + 12*3 = 186 |
|
|
|
|
Looks like we will be up to n=352000 in cca 2 days...
I shall put in some core to make it finnish sooner
/Lennart |
|
|
|
|
|
Current Projects
server = "pgllr.mine.nu"
port 7171: PPS Test on 2000<k<4000 360k<n<363K (k*2^n+1
port 8181: Sophie Germain Prime Search (SGS) (k*2^666666-1)
port 6868: Generalized Woodall & Cullen; b=13 (n*b^n-/+1)
Port 7171 and 8181 is reloaded.
/Lennart |
|
|
|
|
Current Projects
server = "pgllr.mine.nu"
port 7171: PPS Test on 2000<k<4000 360k<n<363K (k*2^n+1
port 8181: Sophie Germain Prime Search (SGS) (k*2^666666-1)
port 6868: Generalized Woodall & Cullen; b=13 (n*b^n-/+1)
Port 7171 and 8181 is reloaded.
/Lennart
Approximately how many Sophie Germain's and Woodall/Cullen base 13's come in each week? Just curious what the turn out is for those two projects, if there very popular or not as much yet?
____________
John M. Johnson "Novex" |
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1902
ID: 352
Credit: 3,602,484,350
RAC: 5,414,201
|
|
If a prime is found, it should be about 4700 in Top5000.
____________
My stats
Badge score: 1*1 + 5*1 + 8*3 + 9*11 + 10*1 + 11*1 + 12*3 = 186 |
|
|
|
|
Current Projects
server = "pgllr.mine.nu"
port 7171: PPS Test on 2000<k<4000 360k<n<363K (k*2^n+1
port 8181: Sophie Germain Prime Search (SGS) (k*2^666666-1)
port 6868: Generalized Woodall & Cullen; b=13 (n*b^n-/+1)
Port 7171 and 8181 is reloaded.
/Lennart
Approximately how many Sophie Germain's and Woodall/Cullen base 13's come in each week? Just curious what the turn out is for those two projects, if there very popular or not as much yet?
There is about 150 Sophie Germain's /day
250 Woodall/Cullen base 13's /day
And about 5000 PPS2 /day |
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1198
ID: 18646
Credit: 566,086,232
RAC: 939,121
|
[2009-01-16 03:37:51 GMT] 3627*2^351347+1: Prime returned by Sysadm@Nbg
Rank (*): 5036
Entrance Rank (*): 4985
to small for longer time in TOP5000 :(
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
|
|
|
|
|
|
Say is there anyway to run these server client dos type builds like prpnet or more what I'm looking for is the psieve.exe one without it showing up a dos box so it's not noticeable?? That would be awesome if there was a way, anyone know of any?
____________
John M. Johnson "Novex" |
|
|
|
|
|
I have it running as a service on several machines. The catch is that you can not shut it down quickly. You can kill the prpnet app but it will not kill the child app until it finishes the current test - cllr will run until it finishes its work.
Microsoft has a utility so you can run any executable as a service. You are looking for srvany.exe and instsrv.exe. |
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1902
ID: 352
Credit: 3,602,484,350
RAC: 5,414,201
|
port 7171: PPS Test on 2000<k<4000 360k<n<363K (k*2^n+1 We have finished this range yesterday (n<=363k). It looks like n is going again from 360k but this time with k from 3000 to 4000.
Is this correct observation?
____________
My stats
Badge score: 1*1 + 5*1 + 8*3 + 9*11 + 10*1 + 11*1 + 12*3 = 186 |
|
|
|
|
port 7171: PPS Test on 2000<k<4000 360k<n<363K (k*2^n+1 We have finished this range yesterday (n<=363k). It looks like n is going again from 360k but this time with k from 3000 to 4000.
Is this correct observation?
Yes that is correct.
/Lennart |
|
|
|
|
|
Okay, I just tried out the mac app (just intel so far). First issue, "prpclient-1.0.5-macintel" was not executable, so I had to fix that. The package should be updated so that both the intel and PPC apps are executable. Second issue, looks like there are some permission issues and missing files:
% chmod +x prpclient-1.0.5-macintel
% ls -la
total 9904
drwxr-xr-x@ 15 zombie67 staff 510 Jan 30 12:27 .
drwx------ 8 zombie67 staff 272 Jan 30 12:10 ..
-rw-r--r-- 1 zombie67 staff 81 Jan 30 12:25 llr.ini
-rw-r--r--@ 1 zombie67 staff 4068944 Jan 9 22:58 mllr-3.7.1c
-rwxr-xr-x@ 1 zombie67 staff 199616 Jan 11 22:29 phrot.p3-0.64-linux
-rwxr-xr-x@ 1 zombie67 staff 635656 Jan 11 22:30 phrot.p3-0.64-linux.static
-rwxr-xr-x@ 1 zombie67 staff 54104 Jan 9 23:05 prpclient-1.0.5-macintel
-rw-r--r--@ 1 zombie67 staff 54268 Jan 9 23:04 prpclient-1.0.5-macppc
-rw-r--r--@ 1 zombie67 staff 2516 Jan 30 12:12 prpclient.ini
-rw-r--r-- 1 zombie67 staff 968 Jan 30 12:25 prpclient.log
-rw-r--r--@ 1 zombie67 staff 10023 Jan 9 18:38 readme_llr.txt
-rw-r--r--@ 1 zombie67 staff 5168 Jan 9 18:48 readme_phrot.txt
-rw-r--r--@ 1 zombie67 staff 1254 Jan 9 18:48 readme_prpclient.txt
-rw-r--r-- 1 zombie67 staff 36 Jan 30 12:25 work_sg.in
-rw-r--r-- 1 zombie67 staff 123 Jan 30 12:25 work_sg.save
% ./prpclient-1.0.5-macintel
[2009-01-30 20:21:19 GMT] PRPNet Client application v1.0.5 started
[2009-01-30 20:21:19 GMT] User email address is zombie67@xxxxxx.com
[2009-01-30 20:21:19 GMT] sg: Getting work from server pgllr.mine.nu at port 8181
This is the new PRPNet Server version 1.0.5
PrimeGrid PRP Test Server SG
sh: ./mllr-3.7.1c: Permission denied
[2009-01-30 20:21:22 GMT] sg: Could not open file [lresults.txt] for reading. Assuming that process was terminated
[2009-01-30 20:21:22 GMT] sg: Quitting due to detected interruption
[2009-01-30 20:21:22 GMT] Total Time: 0:00:03 Total Tests: 0 Total PRPs Found: 0
[2009-01-30 20:21:22 GMT] Client shutdown complete
Edit: Just so everyone knows, in addition to making the mac app executable, you also have to change the permissions on the mllr-3.7.1c file. Changing it to 777 works. Perhaps something less would work too, I didn't test every step in between. In any case, I have it running successfully on an intel Mac (leopard). however, it does not work on my PPC mac (tiger).
____________
Reno, NV
 |
|
|
|
|
Edit: Just so everyone knows, in addition to making the mac app executable, you also have to change the permissions on the mllr-3.7.1c file. Changing it to 777 works. Perhaps something less would work too, I didn't test every step in between. In any case, I have it running successfully on an intel Mac (leopard). however, it does not work on my PPC mac (tiger).
Update: Also runs successfully on Intel Mac (Tiger) with the two changes mentioned above.
____________
Reno, NV
|
|
|
