| Author |
Message |
|---|
|
|
|
Here we discuss PRPNet.
:)
/lennart
____________
 |
|
|
 pschoefer Volunteer developer
Volunteer tester
Send message
Joined: 20 Sep 05
Posts: 685
ID: 845
Credit: 2,886,414,412
RAC: 77,022
                              
|
|
PRPnet (1.0.5 for Windows) crashed on my P4M last night after 68 hours. Since it ran stable for 10 days before the Challenge, I disabled the debug log, so I have only the normal log:
[2009-02-04 02:55:32 GMT] pps: 3619*2^362086+1 is not prime. Residue 14140ADA7880EEA5 Time : 602
[2009-02-04 02:55:32 GMT] Total Time: 68:00:13 Total Tests: 455 Total PRPs Found: 0
[2009-02-04 02:55:32 GMT] pps: Returning work to server pgllr.mine.nu at port 7171
[2009-02-04 02:55:36 GMT] pps: INFO: All 1 test results were accepted
[2009-02-04 02:55:36 GMT] pps: Getting work from server pgllr.mine.nu at port 7171
[2009-02-04 03:07:20 GMT] pps: Could not find RES64 residue [lresults.txt]. Is llr broken?
The work_pps* files (last change 2:55 GMT) were still in the directory and it's working again now.
____________
  |
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1949
ID: 352
Credit: 6,014,682,980
RAC: 1,559,051
                                   
|
|
I'm looking forward to see primes from PRPnet listed on Top Prime Finders.
John says it should be relatively easy but not real-time.
Any ETA on that?
Will it work backward (Primes found so far)?
____________
My stats |
|
|
|
|
PRPnet (1.0.5 for Windows) crashed on my P4M last night after 68 hours. Since it ran stable for 10 days before the Challenge, I disabled the debug log, so I have only the normal log:
[2009-02-04 02:55:32 GMT] pps: 3619*2^362086+1 is not prime. Residue 14140ADA7880EEA5 Time : 602
[2009-02-04 02:55:32 GMT] Total Time: 68:00:13 Total Tests: 455 Total PRPs Found: 0
[2009-02-04 02:55:32 GMT] pps: Returning work to server pgllr.mine.nu at port 7171
[2009-02-04 02:55:36 GMT] pps: INFO: All 1 test results were accepted
[2009-02-04 02:55:36 GMT] pps: Getting work from server pgllr.mine.nu at port 7171
[2009-02-04 03:07:20 GMT] pps: Could not find RES64 residue [lresults.txt]. Is llr broken?
The work_pps* files (last change 2:55 GMT) were still in the directory and it's working again now.
This could happend becuse off a bug in LLR.
http://www.mersenne.org/various/math.php
/Lennart |
|
|
|
|
I'm looking forward to see primes from PRPnet listed on Top Prime Finders.
John says it should be relatively easy but not real-time.
Any ETA on that?
Will it work backward (Primes found so far)?
I can't give you ETA on this now But we are working on it.
/Lennart |
|
|
vasmVolunteer tester
Send message
Joined: 6 Dec 08
Posts: 47
ID: 32604
Credit: 990,892
RAC: 0
                
|
|
From the PRPNet sticky
Attention!!!
...
The second is more critical. After stopping the client and restarting, it will not finish the WU it was working on. It will start on the next WU in the queue. Since PPS WU's are so short, this isn't that much of a loss. However, GCW13 and SG WU's are much longer and thus much more computing time is lost. When closing down, do your best to do it right after a WU has been completed.
Observing the Windows 1.0.5 client on my PC that's not exactly how it behaves.
After stopping the client and restarting, it will not start on the next WU in the queue. It will empty the whole queue sending everything back and download a new queue to start working on.
Each time this happens it will also leave behind a zXXXXXX file. I suppose it's safe to delete those.
If the server recognises the returned incomplete and unprocessed tasks and sends them to someone else, then indeed for PPS it's something minor. |
|
|
|
|
From the PRPNet sticky
Attention!!!
...
The second is more critical. After stopping the client and restarting, it will not finish the WU it was working on. It will start on the next WU in the queue. Since PPS WU's are so short, this isn't that much of a loss. However, GCW13 and SG WU's are much longer and thus much more computing time is lost. When closing down, do your best to do it right after a WU has been completed.
Observing the Windows 1.0.5 client on my PC that's not exactly how it behaves.
After stopping the client and restarting, it will not start on the next WU in the queue. It will empty the whole queue sending everything back and download a new queue to start working on.
Each time this happens it will also leave behind a zXXXXXX file. I suppose it's safe to delete those.
If the server recognises the returned incomplete and unprocessed tasks and sends them to someone else, then indeed for PPS it's something minor.
Thats no problem on server side. They will be resended. And no problem to delete the z* files.
/Lennart |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
  
|
From the PRPNet sticky
Attention!!!
...
The second is more critical. After stopping the client and restarting, it will not finish the WU it was working on. It will start on the next WU in the queue. Since PPS WU's are so short, this isn't that much of a loss. However, GCW13 and SG WU's are much longer and thus much more computing time is lost. When closing down, do your best to do it right after a WU has been completed.
Observing the Windows 1.0.5 client on my PC that's not exactly how it behaves.
After stopping the client and restarting, it will not start on the next WU in the queue. It will empty the whole queue sending everything back and download a new queue to start working on.
Each time this happens it will also leave behind a zXXXXXX file. I suppose it's safe to delete those.
If the server recognises the returned incomplete and unprocessed tasks and sends them to someone else, then indeed for PPS it's something minor.
Thank you for the observation. Yes, indeed that's what it does. I'll update.
____________
|
|
|
|
|
|
If maxworkunits=1 is used do you still see prpnet sending in all work on restart? I don't have a client I can play with right now but using it with the small maxworkunits I have seen the work resume from the z file rather then dumping everything.
S. |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
If maxworkunits=1 is used do you still see prpnet sending in all work on restart? I don't have a client I can play with right now but using it with the small maxworkunits I have seen the work resume from the z file rather then dumping everything.
S.
Another confirmed observation. We'll update the instructions with this exception.
____________
|
|
|
|
|
|
What's on port 9000?
____________
|
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1217
ID: 18646
Credit: 859,608,218
RAC: 206,541
                   
|
What's on port 9000?
I am interested in that too
What is the difference between PPS2 an PPS2_2 ??
Meanwhile I switched my config as you mentioned
EDIT:
something goes wrong with port 9000
[2009-02-07 13:24:23 GMT] pgllr.mine.nu:9000 connect to socket failed
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
 |
|
|
|
|
What's on port 9000?
It's Proth Prime search.
Port 7171 k= 4k-5k n= 360k-363k
Port 9000 k= 5k-7k n= 360k-363k
/Lennart |
|
|
|
|
What's on port 9000?
I am interested in that too
What is the difference between PPS2 an PPS2_2 ??
Meanwhile I switched my config as you mentioned
EDIT:
something goes wrong with port 9000
[2009-02-07 13:24:23 GMT] pgllr.mine.nu:9000 connect to socket failed
Can you try now ?
I hope it shall work.
/lennart |
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1217
ID: 18646
Credit: 859,608,218
RAC: 206,541
|
EDIT:
something goes wrong with port 9000
[2009-02-07 13:24:23 GMT] pgllr.mine.nu:9000 connect to socket failed
Can you try now ?
I hope it shall work.
/lennart
Yes, lennart now it work
[2009-02-07 13:42:47 GMT] PPS2_2: Getting work from server pgllr.mine.nu at port 9000
thx
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
|
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 24,054,820
RAC: 0
      
|
|
I just found a prime on port 9000, first prime I tested at this port :p
If it isn't a double check can I submit the prime to the prime database then?
It is reportable (108377 digits) so I am am not going to give you the number yet.
____________
 |
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 841
ID: 18447
Credit: 646,057,353
RAC: 552,567
              
|
|
My understanding is that PRPnet primes can be reported as soon as you spot them.
I'm sure someone will post if this is not the case.
____________
 |
|
|
|
|
I just found a prime on port 9000, first prime I tested at this port :p
If it isn't a double check can I submit the prime to the prime database then?
It is reportable (108377 digits) so I am am not going to give you the number yet.
I am only testing it for GFN first. I send you a mail in 10 min.
Lennart |
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 24,054,820
RAC: 0
|
|
The prime is reported now
http://primes.utm.edu/primes/page.php?id=86571
I dont get any work @ port 7171
____________
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1949
ID: 352
Credit: 6,014,682,980
RAC: 1,559,051
|
|
I also wasn't getting work from PPS so client took some sg work. About an hour ago work from port 7171 was fetched just fine.
In the meanwhile I wonder what that would mean
[2009-02-07 19:22:52 GMT] sg: INFO: Server has no record of this test. The test result for 51351533565*2^666666-1 has been logged
[2009-02-07 19:20:34 GMT] sg: INFO: Server has no record of this test. The test result for 12694873905*2^666668-1 has been logged
[2009-02-07 19:20:54 GMT] sg: INFO: Server has no record of this test. The test result for 12690414495*2^666668-1 has been logged
____________
My stats |
|
|
|
|
I also wasn't getting work from PPS so client took some sg work. About an hour ago work from port 7171 was fetched just fine.
In the meanwhile I wonder what that would mean
[2009-02-07 19:22:52 GMT] sg: INFO: Server has no record of this test. The test result for 51351533565*2^666666-1 has been logged
[2009-02-07 19:20:34 GMT] sg: INFO: Server has no record of this test. The test result for 12694873905*2^666668-1 has been logged
[2009-02-07 19:20:54 GMT] sg: INFO: Server has no record of this test. The test result for 12690414495*2^666668-1 has been logged
When the server got down 2 user got the same number but only one was registered. So your result are in the logfiles but there was no record on that you have got them.
/Lennart |
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1217
ID: 18646
Credit: 859,608,218
RAC: 206,541
|
|
yeah !!! - another prime in the early sunday morning
can you check it "GFN" an send me a PN / eMail ???
EDIT: thx Lennart !
you can find it here
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
We have opened another port, 9191. Again, it's from the Proth Prime Search.
server=35:PPSE:pgllr.mine.nu:7171
server=0:SGS:pgllr.mine.nu:8181
server=0:GCW13:pgllr.mine.nu:6868
server=35:PPSE_2:pgllr.mine.nu:9000
server=30:PPSE_3:pgllr.mine.nu:9191
The Proth port breakout is as follows:
7171 - 4000<k<5000 for 360K<n<363K
9000 - 5000<k<7000 for 360K<n<363K
9191 - 7000<k<10000 for 360K<n<361K
We're doing this to help provide a little buffer between the 5000th prime and where PRPNet's current search is. Once all ports reach 363K where we have a larger buffer, we'll condense down to one port and run 1200<k<10000 for 363K<n<380K.
Please give the above percentages a run and see how you like it. The higher the k, the less chance that it's been previously searched.
Thanks for your help and BEST of LUCK finding primes!!! :)
p.s. SGS and GCW13 still have plenty of work. We're just focusing on PPSE currently until there's a comfortable buffer.
____________
|
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1217
ID: 18646
Credit: 859,608,218
RAC: 206,541
|
|
lennart wrote:
clientid=user name
Enter you PrimeGrid User name !!!
This is important if we shall try do any stats.
when I use my user-name, the client does not start because of the '@'
what shall i use instead ???
PS: changing my user-name is not an alternative ;)
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
|
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 24,054,820
RAC: 0
|
when I use my user-name, the client does not start because of the '@'
what shall i use instead ???
You can use anything you want, and on every computer you run prpclient you should use another username. You will be credited as long as your e-mailaddres is ok.
____________
|
|
|
|
|
when I use my user-name, the client does not start because of the '@'
what shall i use instead ???
You can use anything you want, and on every computer you run prpclient you should use another username. You will be credited as long as your e-mailaddres is ok.
I would like to have Boinc Username on clientid !!
If you dont use Boinc email there is no way i can se who it is if you not use your Boinc username.
I need it todo stats and give credit.
If you have @ in your name replace that with "at"
/Lennart
/Lennart |
|
|
|
|
|
Sorry, I'm confused.
The original post said that clientid "...differtiates clients using the same e-mail ID"
So I've been using different clientid's based on the host, i.e. Vista64, PentiumM, etc.
So does that mean I will not be attributed any of the WUs I've processed so far?
Thanks,
Peter
____________
35 x 2^3587843+1 is prime! |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
So does that mean I will not be attributed any of the WUs I've processed so far?
Sorry for the confusion. We're attempting to work around the lack of a username= field in PRPNet. As long as your email in PRPNet matches the one you used in PrimeGrid, then you'll receive credit.
Yes, primary distinction is email address. We will use this to award credit and primes. However, if it does not match the one used at PrimeGrid, then we need a backup way of determining who the user is.
Currently we have a PRPNet user with 2 reportable primes whose email and clientid matches nothing in PrimeGrid's database. Additionally, they have not responded to emails and thus have not reported their primes. We would very much like for everyone to receive their due credit. However, after two weeks, the prime will be reported anonymously.
____________
|
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1217
ID: 18646
Credit: 859,608,218
RAC: 206,541
|
when I use my user-name, the client does not start because of the '@'
what shall i use instead ???
I would like to have Boinc Username on clientid !!
If you dont use Boinc email there is no way i can se who it is if you not use your Boinc username.
I need it todo stats and give credit.
If you have @ in your name replace that with "at"
/Lennart
Just replaced it to
clientid=SysadmAtNbg
hope this works for you ..
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
|
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1255
ID: 12001
Credit: 18,565,548
RAC: 0
|
|
The clientid is intended to help a user to differentiate between PRPNet clients as users typically run more than one instance of PRPNet across multiple machines or even on the same machine. This is useful if someone on the server sees unusual activity with a client allowing them to communicate with the person (via the supplied e-mail) that they have a misbehaving box and can thus tell them which box is having a problem.
The next release will have a separate userid configuration for the client. I've just been too busy to finish the release. |
|
|
|
|
|
Thanks for the clarification.
I've been using my boinc/primegrid email address so that's good news.
Thanks again,
Peter.
____________
35 x 2^3587843+1 is prime! |
|
|
|
|
|
I tried to find the answer, and it may be somewhere, but I couldn't find it. Sorry if it is answered elsewhere.
What exactly is the difference between crunching PG workunits on BOINC, and crunching them on PRPNet? Are the WU's on the 2 programs being doubled up on? Or are half of the numbers to be checked being sent to BOINC, and the other half being sent to PRPNet? Or are they doublechecking the same numbers?
Thanks in advance. |
|
|
|
|
I tried to find the answer, and it may be somewhere, but I couldn't find it. Sorry if it is answered elsewhere.
What exactly is the difference between crunching PG workunits on BOINC, and crunching them on PRPNet? Are the WU's on the 2 programs being doubled up on? Or are half of the numbers to be checked being sent to BOINC, and the other half being sent to PRPNet? Or are they doublechecking the same numbers?
Thanks in advance.
On both Boinc and PRPNet we do PPS ( k*2^n+1)
On Boinc we do k=5-1200*2^n+1 and we are at n~530k
On PRPnet we do k=1200-10000*2^n+1 we are at n~361k now.
/Lennart
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
What exactly is the difference between crunching PG workunits on BOINC, and crunching them on PRPNet? Are the WU's on the 2 programs being doubled up on? Or are half of the numbers to be checked being sent to BOINC, and the other half being sent to PRPNet? Or are they doublechecking the same numbers?
In general, PRPNet is used in the Project Staging Area (PSA). Here projects are prepared for entry into BOINC. This includes both sieving and primality testing. PRPNet is used to do the primality testing. Lately, the PSA has grown to include additional projects that may not enter BOINC. BOINC is reserved for PrimeGrid's main efforts while PSA now includes some secondary efforts which are run in PRPNet.
WU's in both areas are mutually exclusive. They are not duplicating efforts.
More specifically, the main LLR efforts (in BOINC) are 321, Cullen, PSP, PPS, TPS, and Woodall. The secondary LLR efforts (in PRPNet) are GCW13, PPSE, and SGS.
BOINC and PRPNet use the same LLR program to primality test. One advantage that PRPNet has over BOINC is that Mac's can also participate (using LLR or phrot). We do not yet have a Mac application for BOINC.
Adding projects to BOINC requires a fair amount of work. Adding projects to PRPNet is quite easy. However, there are no bells and whistles...just a dos/terminal window. With BOINC comes the flexibility to join many "BOINC" projects. PRPNet just allows you to test for prime numbers which is quite nice if that's all you want to do. :) Both BOINC and PRPNet are very simple to install and run.
We plan to continue to add more projects to the PSA and test them through PRPNet. Hopefully this will give PrimeGrid users a variety of prime projects to participate in.
As Lennart points out, Proth Prime Search (PPS) is in BOINC for 4<k<1200. This is PrimeGrid's main effort. Proth Prime Search Extended (PPSE) is in the Project Staging Area using PRPNet for 1200<k<10000. As you can see, both do the same test for k*2^n+1...just for different k ranges.
____________
|
|
|
|
|
|
Great, that completely answers my questions. Thank you to both of you!
Chris |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
I have just uploaded some experimental multicore (dual, quad, & dual-quad) installations of PRPClient. If anyone is interested in testing, feedback would be greatly appreciated. Just follow the instructions in the readme_primegrid.txt located in the archive.
Only Windows is available right now but if this works, I'll create ones for Linux and Mac.
Archives can be accessed here: http://pgllr.mine.nu/PRPNet/
____________
|
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 24,054,820
RAC: 0
|
|
They work fine but the unique name is the same for the 4 cores
____________
|
|
|
BenvaVolunteer tester
Send message
Joined: 5 May 08
Posts: 73
ID: 22332
Credit: 2,715,050
RAC: 0
|
|
I'm trying it and it works fine, thank you.
____________
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1949
ID: 352
Credit: 6,014,682,980
RAC: 1,559,051
|
|
Install PRPClients.bat copy only prpclient-1.0.5.exe, not cllr-3.7.1c.exe and phrot.p3-0.63-mingw32.exe.
Otherwise, it's fine.
____________
My stats |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
Install PRPClients.bat copy only prpclient-1.0.5.exe, not cllr-3.7.1c.exe and phrot.p3-0.63-mingw32.exe.
Otherwise, it's fine.
Each folder should not need it's own copy of those programs...unless feedback determines otherwise.
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
They work fine but the unique name is the same for the 4 cores
I think this is fine for now as the "real" purpose of the clientid is to determine the machine and not necessarily which core...although core detail can be added manually.
So yes, each core will have the same name. However, you can still have different names for each machine.
See this post for the "intention" of clientid: http://www.primegrid.com/forum_thread.php?id=1212&nowrap=true#13468
For now, until the username option is added in the next prpclient release, we are asking users to use their PrimeGrid username in this field. If you still want to uniquely identify the host, you can use this naming converntion: PrimeGrid Username_client name. For example:
john_linux1
____________
|
|
|
|
|
|
That multi-install crashes everytime i complete a test under win32 on a dualcode. |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
That multi-install crashes everytime i complete a test under win32 on a dualcode.
We have been unable to duplicate the crash. Anyone else experiencing this? If so, please provide details.
____________
|
|
|
pschoeferVolunteer developer
Volunteer tester
Send message
Joined: 20 Sep 05
Posts: 685
ID: 845
Credit: 2,886,414,412
RAC: 77,022
|
That multi-install crashes everytime i complete a test under win32 on a dualcode.
We have been unable to duplicate the crash. Anyone else experiencing this? If so, please provide details.
Same here (quad with win xp64). All four instances crashed after completing the test (and unfortunately before reporting the result, so all work was lost).
There is a popup window "Microsoft Visual C++ Debug Library" with the following text:
Debug Error!
Program: ...\prpclient-1.0.5.exe
Module: ...\prpclient-1.0.5.exe
File:
Run-Time Check Failure #2 - Stack around the variable 'command' was corrupted.
From the windows event log:
Application Failure prpclient-1.0.5.exe 0.0.0.0 in ntdll.dll 5.2.3790.3959 at offset 0001002d
This only happens if Phrot was used, no problems after tests using LLR.
Edit: Looks like the client only crashed if you click the retry button in that "Microsoft Visual C++ Debug Library" window. After clicking the ignore button, the client is working again.
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
Thanks for the info. We are currently exploring solutions. We kindly ask anyone running a multicore installation and using phrot to please stop the testing until we're able to resolve the issue. We hope to have a solution in the next few hours. Until then, feel free to continue testing on PPSE or SGS which will use the LLR application.
Thanks!
____________
|
|
|
|
|
|
Well copying LLR and Phrot into the subdirs would be an easy fix for that, right? There is absicaly no downside by having 1 or 2 more MB then. |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
Updated archives have been uploaded. It was discovered that phrot 0.63 was used in those releases when it should have been at least 0.64. Therefore, we took the opportunity to go ahead and update to 0.66.
We're not sure if that solved the problem but so far there have been no errors.
We have also uploaded a version of the install that copies the LLR and Phrot programs into each folder. You can download the updated archives here:
http://pgllr.mine.nu/PRPNet/multicore/
Both versions include the updated phrot 0.66
Copies LLR and Phrot into each folder
prpclient-dualquad-1.0.5-windows.zip
prpclient-dual-1.0.5-windows.zip
prpclient-quad-1.0.5-windows.zip
Calls LLR and Phrot from the programs folder
prpclient-quad2-1.0.5-windows.zip
prpclient-dualquad2-1.0.5-windows.zip
prpclient-dual2-1.0.5-windows.zip
We appreciate the feedback.
EDIT: We have removed the 2nd version as this seemed to be the problem. While LLR worked fine as a shared program, phrot kept tripping up prpclient. Now with local program files in each folder, we have not seen the error again. Please let us know what you experience.
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
Linux multicore installation released for testing
Thanks to Ward De Ridder, we now have a multicore installation for Linux. His version has been tested with ubuntu x64, but should also work with kde and xface on any linux-distribution.
However, I made a few changes for consistency (filenames) and added a start for those who want to run the static versions of prpclient and phrot. Starts for 32 bit as well as support for duals and dual quads were added. phrot was updated from 0.64 to 0.66.
Not all of these changes have been tested. If anyone is interested, please give them a run through. The instructions are in the readme_primegrid.txt file.
http://pgllr.mine.nu/PRPNet/multicore/
Feedback is appreciated!
Again, thanks to Ward De Ridder! :)
____________
|
|
|
|
|
|
PRPNet status
Port 9000 is on n=360750
Port 7171 n=361250
Port 9191 n=360750
/Lennart |
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1949
ID: 352
Credit: 6,014,682,980
RAC: 1,559,051
|
I'm looking forward to see primes from PRPnet listed on Top Prime Finders.
John says it should be relatively easy but not real-time.
This may get some more attraction to PRPNet.
It would be also fair to count primes found on PRPNet amog others...
____________
My stats |
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1217
ID: 18646
Credit: 859,608,218
RAC: 206,541
|
|
result of an internal Seti.Germany discussion (pschoefer, /me):
It would be nice, if there was an option in PRPClient to limit the number of workunits for each port. At failover (prpclient.ini-Definition=0%) only a minimum of workunits should be downloaded.
Benefit:
- prevent of an DoS (maxworkunit=1 and failover-port is 10000)
- prevent of download huge numbers from a port with long running workunits
other opinions ???
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
|
|
|
|
|
result of an internal Seti.Germany discussion (pschoefer, /me):
It would be nice, if there was an option in PRPClient to limit the number of workunits for each port. At failover (prpclient.ini-Definition=0%) only a minimum of workunits should be downloaded.
Benefit:
- prevent of an DoS (maxworkunit=1 and failover-port is 10000)
- prevent of download huge numbers from a port with long running workunits
other opinions ???
I could do that in the servers. But i don't think 10000 is a good server to set maxwu=1
That should be to many request/sec because off the short wu's.
port 5000: 12121 Prime Search k=121+/- k=27+/- Maxwu=5
port 6868: Generalized Woodall & Cullen; b=13 (n*13^n-/+1) Maxwu=5
port 7171: Proth Prime Search Extended on1200<k<10000 360k<n<380k(k*2^n+1) Maxwu=20
port 8181: Sophie Germain Prime Search (SGS) (k*2^666666-1) Maxwu=5
port 9000: Proth Prime Search Extended on 1200<k<10000 360k<n<380k Maxwu=20
port 9191: Proth Prime Search Extended on 1200<k<10000 360k<n<380k Maxwu=1
port 10000: Proth Prime Search Extended on 1200<k<10000 50k<n<360k Maxwu=60
What do you think ??
/Lennart |
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1217
ID: 18646
Credit: 859,608,218
RAC: 206,541
|
Benefit:
- prevent of an DoS (maxworkunit=1 and failover-port is 10000)
But i don't think 10000 is a good server to set maxwu=1
That should be to many request/sec because off the short wu's.
I fully agree! DoS means Denial of Service or like you said: to many request ...
Your server-side settings match for me - what are other users thinking ???
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1949
ID: 352
Credit: 6,014,682,980
RAC: 1,559,051
|
|
20 task makes a connection interval of ~1 hour. 60 makes it 3 hours - not big difference.
Connetion interval once a day would make difference.
Found a prime 3 days ago on PRPNet without getting e-mail.
Tryed to put it on db but got:
Prime "8255*2^361263+1" is too small (108755 digits). Need 108796 digits or more for a general prime.
I would like to see those primes among Top Prime Finders on PG at least...
We need to speed-up on PRPNet.
____________
My stats |
|
|
|
|
|
Sorry I may be a tad stupid this morning, but because my name has a space in, how would I put that into clientid?
Many Thanks! |
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 24,054,820
RAC: 0
|
|
clientid=StephenBrown
make shure you're e-mail is the same you use for boinc and then you'll certaily be the one who will be credited for the primes you find.
____________
|
|
|
|
|
|
Brilliant, thanks very much! |
|
|
|
|
Sorry I may be a tad stupid this morning, but because my name has a space in, how would I put that into clientid?
Many Thanks!
Stephen_Brown or StephenBrown
:)
/Lennart |
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1949
ID: 352
Credit: 6,014,682,980
RAC: 1,559,051
|
|
Again with no notification e-mail from PG but interesting when put on Prime Database...
At the rate at which primes have been added to this list in the past, the 108855 digit prime "7799*2^361593+1" may only be on the list for about 0 weeks.
____________
My stats |
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 24,054,820
RAC: 0
|
Again with no notification e-mail from PG but interesting when put on Prime Database...
At the rate at which primes have been added to this list in the past, the 108855 digit prime "7799*2^361593+1" may only be on the list for about 0 weeks.
The main thing is that the prime enters the list, then the prime will be in your 5000-prime database forever.
____________
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1949
ID: 352
Credit: 6,014,682,980
RAC: 1,559,051
|
|
I wasn't paying enough attention to Lennart's post about what available on each port, sorry.
Port 10000 for 50k<n<360k. Now ~5 sec per task for k~72k, connection with 60 tasks in batch every 5 minutes.
____________
My stats |
|
|
|
|
|
Port 10000 is out of work.
____________
|
|
|
|
|
|
It would be nice if all move to port 9191 today.
We will then soon have it dry and can focus on port 7171 & 9000.
Port 7171 & 9000 start at n=380k k=1200-10000
I think we will dry port 9191 tonight if we get all on it.
Please :)
This would be i nice config today :)
server=100:PPSE_3:pgllr.mine.nu:9191
server=0:PPSE:pgllr.mine.nu:7171
server=0:PPSE_2:pgllr.mine.nu:9000
/Lennart |
|
|
|
|
It would be nice if all move to port 9191 today.
We will then soon have it dry and can focus on port 7171 & 9000.
Port 7171 & 9000 start at n=380k k=1200-10000
Now we are done with port 9191.
I close port 9191 for now.
I will be loading port 7171 & 9000 with small ranges from
k=1200-10000 n>380k.
Thank for all help to dry this port.  |
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1255
ID: 12001
Credit: 18,565,548
RAC: 0
|
|
I've updated both PRPNet and phrot today. You can read about the changes at
http://www.mersenneforum.org/showthread.php?t=11220
and
http://www.mersenneforum.org/showthread.php?t=11264
|
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 24,054,820
RAC: 0
|
|
I tested the new version of prpclient and it works as it should.
I have tested it with xp x32 yesterday and today I compiled a version for ubuntu x64, I am still doing some tests with it but I think everything is ready to be used on primegrid.
____________
|
|
|
|
|
|
Sorry for the downtime on Port10000.
Its up again. Loaded with n=81k-82k
/Lennart
|
|
|
|
|
|
The Mac apps works...any chance it will be "BOINC"d?
____________
Reno, NV
 |
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 841
ID: 18447
Credit: 646,057,353
RAC: 552,567
|
|
Are we migrating to prpnet-1.0.6 at some point, or are we sticking with 1.0.5 for the moment?
____________
|
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1255
ID: 12001
Credit: 18,565,548
RAC: 0
|
The Mac apps works...any chance it will be "BOINC"d?
Which app are you talking about? |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
The Mac apps works...any chance it will be "BOINC"d?
Yes, the Mac PRPClient works. It uses phrot for MacPPC and the Mac build LLR for MacIntel.
The MacIntel LLR build can be downloaded here:
http://jpenne.free.fr/index2.html
Also, the MacIntel build for sr2sieve is available. It can be downloaded here:
http://www.geocities.com/g_w_reynolds/sr2sieve/
PrimeGrid BOINC uses LLR to primality test and sr2sieve for most of its sieves. The problem has been getting the Mac builds of these applications to work with BOINC. Once this is accomplished, PrimeGrid will have LLR and Sieve projects available for MacIntel computers.
Additionally, if we can get phrot implemented in BOINC, then the Mac PPC will be able to primality test in BOINC as well.
Until then the only MacIntel and MacPPC apps we have are in PRPNet.
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
Are we migrating to prpnet-1.0.6 at some point, or are we sticking with 1.0.5 for the moment?
Yes, we are awaiting builds for PRPNet.
____________
|
|
|
|
|
|
I just downloaded the Mac client for PRPnet and tweeked the ini file. I then tried to execute the client app and got the following error:
dyld: ./prpclient-1.0.5-macppc Undefined symbols:
./prpclient-1.0.5-macppc undefined reference to ___stdoutp expected to be defined in /usr/lib/libSystem.B.dylib
Trace/BPT trap
I have 10.3.9 running on this machine.
Any suggestions are appreciated.
____________
|
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 24,054,820
RAC: 0
|
Are we migrating to prpnet-1.0.6 at some point, or are we sticking with 1.0.5 for the moment?
I made (compiled) a version for ubuntu x64 (it will probably work also for ubuntu x86 and for other linux distro's but untested), this is not an official version but it works.
You'll probably have to install build-essential.
it can be downloaded here: http://upload.warddr.be/
PS. If anyone knows a good free php upload - download-script (with more or less the same functions as the server used by lennart to upload the primegridfiles, but then for linux on an shared hosting account) I am always interested to hear about it.
____________
|
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1255
ID: 12001
Credit: 18,565,548
RAC: 0
|
I just downloaded the Mac client for PRPnet and tweeked the ini file. I then tried to execute the client app and got the following error:
dyld: ./prpclient-1.0.5-macppc Undefined symbols:
./prpclient-1.0.5-macppc undefined reference to ___stdoutp expected to be defined in /usr/lib/libSystem.B.dylib
Trace/BPT trap
I have 10.3.9 running on this machine.
Any suggestions are appreciated.
I built it with gcc 4.0.1 on OS X 10.5. If you have XCode on that Mac, building the client is a no-brainer. I'm not certain how to build it to support prior versions of OS x. |
|
|
|
|
|
How many candidates do we have left to test from the range (port 10000) 1200<k<10000 81.7k<n<360k? 5M candidates....?!
____________
|
|
|
|
|
Sorry for the downtime on Port10000.
Its up again. Loaded with n=81k-82k
/Lennart
Why don't you load up to 90k?
____________
|
|
|
|
|
Sorry for the downtime on Port10000.
Its up again. Loaded with n=81k-82k
/Lennart
Why don't you load up to 90k?
I take smaller range becuse 10k is 1,8M candidates.
I shall count them when i do a new file. The sievefile is 1,3G in abcd format.
/Lennart |
|
|
|
|
I shall count them when i do a new file. The sievefile is 1,3G in abcd format.
/Lennart
Something in the range of 340M candidates, I think.
____________
There are only 10 kinds of people - those who understand binary and those who don't
 |
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 24,054,820
RAC: 0
|
|
port 10000 is out of work
____________
|
|
|
|
|
port 10000 is out of work
Awesome were doing good then, what are the next few that are just about done so I can change mine to help finish more, 9191 was fun to finish I did that one for a while for the feb small challenge for primes and didn't get any primes LOL go figure right :) Just my luck
Edit: Thats how it goes though, in spurts of a bunch in a week then none for a while lol right
____________
John M. Johnson "Novex" |
|
|
pschoeferVolunteer developer
Volunteer tester
Send message
Joined: 20 Sep 05
Posts: 685
ID: 845
Credit: 2,886,414,412
RAC: 77,022
|
|
I've trouble with PRPNet on my Quad with WinXP x64. I started four instances, but one of them crashed after a short time. Phrot.exe was still running 4 times after the crash, but prpclient.exe only 3 times. My remote connection broke, and after reconnecting I saw the following error message:
Application Failure prpclient-1.0.5.exe 0.0.0.0 in 0.0.0.0 at offset 0x0000000
There's no entry in prpclient.log or debug.log.
I tried updating to 1.0.6, but that version does not even start processing:
[2009-03-12 22:07:25 GMT] PRPNet Client application v1.0.6 started
[2009-03-12 22:07:25 GMT] User email address is ***@***.de
[2009-03-12 22:07:25 GMT] GCW13: Could not open file [results.out] for reading. Assuming that process was terminated
[2009-03-12 22:07:25 GMT] GCW13: Quitting due to detected interruption
[2009-03-12 22:07:25 GMT] Total Time: 0:00:00 Total Tests: 0 Total PRPs Found: 0
[2009-03-12 22:07:25 GMT] Client shutdown complete
The old 1.0.4 shows the same problem as 1.0.5.
At the moment I have only 3 prpclients running, but I'd really like to add the fourth core for our mini-challenge. ;)
____________
|
|
|
|
|
At the moment I have only 3 prpclients running, but I'd really like to add the fourth core for our mini-challenge. ;)
Clean that folder Remove all work files and z........ files
Start with only
prpclient
phrot
llr
prpclient.ini
set debuglevel=1
/Lennart
|
|
|
pschoeferVolunteer developer
Volunteer tester
Send message
Joined: 20 Sep 05
Posts: 685
ID: 845
Credit: 2,886,414,412
RAC: 77,022
|
Clean that folder Remove all work files and z........ files
...
set debuglevel=1
Done that.
One client crashed again, here's the complete debug.log:
[2009-03-12 23:22:07 GMT] PRPNet Client application v1.0.5 started
[2009-03-12 23:22:07 GMT] User email address is xxxxxxxxx@zzzzzzzz.yy
[2009-03-12 23:22:14 GMT] in FindNextServerForWork: total time for client=0 seconds
[2009-03-12 23:22:22 GMT] suffix: GCW13, no work done yet, target pct work done=100
[2009-03-12 23:25:46 GMT] socket 916 >>>> FROM xxxxxxxx@zzzzzzz.yy pschoefer
[2009-03-12 23:26:02 GMT] GCW13: Getting work from server pgllr.mine.nu at port 6868
[2009-03-12 23:26:03 GMT] socket 916 >>>> GETWORK 1.0.5 1
[2009-03-12 23:26:03 GMT] socket 916 <<<< ServerVersion: 1.0.5
[2009-03-12 23:26:03 GMT] socket 916 <<<< WorkUnit: 348198*13^348198+1 1236900356 348198 13 348198 1
[2009-03-12 23:26:03 GMT] socket 916 <<<< End of Message
[2009-03-12 23:26:03 GMT] socket 916 >>>> GETGREETING
[2009-03-12 23:26:04 GMT] socket 916 <<<< This is the new PRPNet Server version 1.0.5
[2009-03-12 23:26:04 GMT] socket 916 <<<< PrimeGrid PRP Test Server GCW13
[2009-03-12 23:26:04 GMT] socket 916 <<<< OK.
[2009-03-12 23:26:04 GMT] socket 916 >>>> QUIT
[2009-03-12 23:26:04 GMT] closing socket 916
According to Windows' event log, it crashed the same second that it was started (same error message as before). When Phrot was started, the prpclient.exe disappeared from task manager. |
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1255
ID: 12001
Credit: 18,565,548
RAC: 0
|
One client crashed again, here's the complete debug.log:
[2009-03-12 23:22:07 GMT] PRPNet Client application v1.0.5 started
[2009-03-12 23:22:07 GMT] User email address is xxxxxxxxx@zzzzzzzz.yy
[2009-03-12 23:22:14 GMT] in FindNextServerForWork: total time for client=0 seconds
[2009-03-12 23:22:22 GMT] suffix: GCW13, no work done yet, target pct work done=100
[2009-03-12 23:25:46 GMT] socket 916 >>>> FROM xxxxxxxx@zzzzzzz.yy pschoefer
[2009-03-12 23:26:02 GMT] GCW13: Getting work from server pgllr.mine.nu at port 6868
[2009-03-12 23:26:03 GMT] socket 916 >>>> GETWORK 1.0.5 1
[2009-03-12 23:26:03 GMT] socket 916 <<<< ServerVersion: 1.0.5
[2009-03-12 23:26:03 GMT] socket 916 <<<< WorkUnit: 348198*13^348198+1 1236900356 348198 13 348198 1
[2009-03-12 23:26:03 GMT] socket 916 <<<< End of Message
[2009-03-12 23:26:03 GMT] socket 916 >>>> GETGREETING
[2009-03-12 23:26:04 GMT] socket 916 <<<< This is the new PRPNet Server version 1.0.5
[2009-03-12 23:26:04 GMT] socket 916 <<<< PrimeGrid PRP Test Server GCW13
[2009-03-12 23:26:04 GMT] socket 916 <<<< OK.
[2009-03-12 23:26:04 GMT] socket 916 >>>> QUIT
[2009-03-12 23:26:04 GMT] closing socket 916
According to Windows' event log, it crashed the same second that it was started (same error message as before). When Phrot was started, the prpclient.exe disappeared from task manager.
The problem is not with the socket communication. There appears to be some issue with the client, but I have idea what it could be outside of a possible memory leak. I presume that you are running each instance in a separate folder and the one that is failing is the same one. It makes no sense why some would work and others would not considering it is the same exe. All that happens is that the PRPNet client passes a command to the system that in turn runs phrot (or llr) to do the PRP test. It doesn't do anything until the system returns, which is after phrot or llr completes it PRP test. If it returns early, then it presumes the phrot or llr was terminated. I don't know why it would return before phrot/llr completes. That has more to do with how Windows works than PRPNet.
Unfortunately I do not have a Quad or Win64, so I have no means of debugging this. Has anyone been running PRPNet on this project on a Quad using Win32?
One thing I could try is to give you a build of the client that can add some additional trace information. I'm curious if system() itself fails or if the PRPNet client runs into a memory leak after system() returns. I suspect the former as it appears there is a branch to 0 happening somewhere. The client can't do that unless clobbering its own footprint, which is unlikely or this would be happening to other people and to other copies of the same client on your PC.
If you put that client on PPS instead of GCW, does it still fail? |
|
|
|
|
|
I'm running WinXPHome32bit at the moment. All 4 Cores are doing GCW13, each in a separate folder. No Problems so far.
____________
There are only 10 kinds of people - those who understand binary and those who don't
|
|
|
pschoeferVolunteer developer
Volunteer tester
Send message
Joined: 20 Sep 05
Posts: 685
ID: 845
Credit: 2,886,414,412
RAC: 77,022
|
|
Yes, I was running each instance in a seperate folder and it was always the same instance that crashed. But in the afternoon I also had another instance crashing with the same error while only 3 instances were running.
I just tried running PPS instead of GCW13, but it crashed, too. After closing the other three instances, I was able to run the client that crashed before, to finish the WUs. |
|
|
|
|
|
Hello all,
I had a problem with clients stopping on me and I sent this as a private message to lennart.
I downloaded/installed prpclient-quad-1.0.5-windows on my Vista 32-bit
machine on March 10. Everything worked fine at first, then one time I
came back to the computer and prpclient-3 & prpclient-4 were closed. I couldn't get either of them to restart and still can't. This is the last few lines from the
prpclient.log from prpclient-3 starting from just before it quit working:
[2009-03-11 18:26:51 GMT] PPSE_3: 5973*282449+1 is not prime. Residue A36D10295024F51B Time : 10
[2009-03-11 18:26:51 GMT] Total Time: 6:40:25 Total Tests: 190 Total PRPs Found: 0
[2009-03-11 18:26:52 GMT] PPSE_3: Returning work to server pgllr.mine.nu at port 10000
[2009-03-11 18:26:53 GMT] PPSE_3: INFO: All 1 test results were accepted
[2009-03-11 18:26:53 GMT] PPSE_2: Getting work from server pgllr.mine.nu at port 9000
[2009-03-11 18:26:58 GMT] PPSE_2: Could not find RES64 residue [lresults.txt]. Is llr broken?
[2009-03-11 19:14:28 GMT] PRPNet Client application v1.0.5 started
[2009-03-11 19:14:28 GMT] User email address is jtmarsh@shawneelink.com
[2009-03-11 19:14:29 GMT] PPSE_2: Could not find RES64 residue [lresults.txt]. Is llr broken?
[2009-03-11 19:27:44 GMT] PRPNet Client application v1.0.5 started
[2009-03-11 19:27:44 GMT] User email address is jtmarsh@shawneelink.com
[2009-03-11 19:27:44 GMT] PPSE_2: Could not find RES64 residue [lresults.txt]. Is llr broken?
[2009-03-12 05:11:42 GMT] PRPNet Client application v1.0.5 started
[2009-03-12 05:11:42 GMT] User email address is jtmarsh@shawneelink.com
[2009-03-12 05:11:42 GMT] PPSE_2: Could not find RES64 residue [lresults.txt]. Is llr broken?
[2009-03-12 13:48:00 GMT] PRPNet Client application v1.0.5 started
[2009-03-12 13:48:00 GMT] User email address is jtmarsh@shawneelink.com
[2009-03-12 13:48:00 GMT] PPSE_2: Could not find RES64 residue [lresults.txt]. Is llr broken?
The log for prpclient-4 has the same final entries in it.
I found in message 13338 where this problem was posted but pschoefer stated
his started back up. Then in message 13341, lennart replied:
This could happen because of a bug in LLR.
http://www.mersenne.org/various/math.php
Since mine have not restarted, I was wondering what to do to fix this
problem?
Lennart got back to me with the following.
I am investigating this problem.
I have seen it here to, but I didn't really know why !
Stop all 4 clients
Remove all work**** files.
Remove all z???????? files,
Restart it and see if it works.
That worked, but I am wondering if this happens to anyone else, and what the actual cause/solution might be if anyone knows.
____________
 |
|
|
|
|
|
Hello everybody.
I just noticed that my computer found a prime on prpclient-1 on 3/11 @ 18:33 UTC.
Here is a copy/paste from the llr.prime file in that folder.
4025*2^82451+1 is prime! Time : 10.263 sec.
I have seen in messages in this forum something about testing for GFN which I just found out to be Generalized Fermat Number. Is this something I do or something that gets done automatically in the background?
Did/does this prime get reported automatically? Or do I need to do something. How do I tell if I’m the primary finder or secondary checker? How do I find out how many digits are in this number?
As you can tell, I’m completely lost here. I appreciate any help you can give me.
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
Is this something I do or something that gets done automatically in the background?
Lennart does that manually. However, in the next PRPNet release, this is done automatically.
Did/does this prime get reported automatically? Or do I need to do something.
This prime is from port 10000 and is too small to enter into the Top 5000 primes list. You can see the size of the 5000th prime here.
While your prime is not reportable, PrimeGrid is keeping record of it. It will eventually be listed in PG's primes database.
How do I tell if I’m the primary finder or secondary checker?
In PRPNet, we are not doublechecking yet...therefore, you are the primary finder.
How do I find out how many digits are in this number?
d = log(k) + n*log(2)
In your case, d = log(4025) + 82451*log(2) = 24823.83 rounding up gives 24,824 digits long.
____________
|
|
|
|
|
|
Is there a way to minimize the client so it just shows next to the BOINC icon, so that the actual window does not need to be open? I do not know if this question makes sense. I can open the BOINC Manager window and BOINC will pop up from where it is on the Taskbar, now is there a way to get the client to just show up there, and I can open it periodically to make sure it is working all right?
Ni
____________
|
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1255
ID: 12001
Credit: 18,565,548
RAC: 0
|
Is there a way to minimize the client so it just shows next to the BOINC icon, so that the actual window does not need to be open? I do not know if this question makes sense. I can open the BOINC Manager window and BOINC will pop up from where it is on the Taskbar, now is there a way to get the client to just show up there, and I can open it periodically to make sure it is working all right?
Ni
Possibly, but it is very low on my priority list. Actually it isn't even on my priority list. This is a DOS application, not a GUI application. It is that way because it is easier to port. If someone with Windows GUI experience wants to GUIfy the app for Windows, I will work with that person to make it happen. |
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1217
ID: 18646
Credit: 859,608,218
RAC: 206,541
|
|
Anybody get work from SGS ???
[2009-03-20 19:30:45 GMT] SGS: Getting work from server pgllr.mine.nu at port 8181
[2009-03-20 19:30:50 GMT] SGS: INFO: No available candidates are left on this server.
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
|
|
|
|
|
Anybody get work from SGS ???
[2009-03-20 19:30:45 GMT] SGS: Getting work from server pgllr.mine.nu at port 8181
[2009-03-20 19:30:50 GMT] SGS: INFO: No available candidates are left on this server.
It is empty !
I am just now cleaning up and load new wu.
Shall be up in 20min.
/Lennart
EDIT:IT is up again. |
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1217
ID: 18646
Credit: 859,608,218
RAC: 206,541
|
|
sorry for late answer (had no access to my logfile, cause of been away):
got some SGS meanwhile
[2009-03-21 00:33:11 GMT] SGS: Getting work from server pgllr.mine.nu at port 8181
thanks for filling up to lennart
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
|
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1217
ID: 18646
Credit: 859,608,218
RAC: 206,541
|
|
an other question:
Is it a problem, if a PRPNetClient after more then one week does not return a result ?? what is the deadline of the ports ??
I am asking, because my laptop might be a few days offline. Shall I start to crunch with PRPNetClient or not ...
An other posibility is to use only the short port 10000 and let the Client run dry!
what do you think of ...
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
|
|
|
|
|
|
Let's make a rally on port 10000? I know it's a hard work to run millions of small candidates but someone has to do it. Don't forget that this is a very important task.
Carlos
____________
|
|
|
|
|
|
Hi everyone :)
I have setup a PRPNet server with sr5 work
Project forum Sierpinski/Riesel Base 5
I need a testrun on that port at least 36 hr starting NOW :)
Use this.
server=100:sr5:samband.mine.nu:7001
or.
server=100:sr5:pgllr.mine.nu:7001
You will end up on the same server.
The work is k*5^n-1
There n= 300k- 350k
Base=5
Please come and help me test this 36 hr from now.
Dont go for more then 1 wu cache before you see the time.
/Lennart |
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 24,054,820
RAC: 0
|
Hi everyone :)
I have setup a PRPNet server with sr5 work
Project forum Sierpinski/Riesel Base 5
I need a testrun on that port at least 36 hr starting NOW :)
Use this.
server=100:sr5:samband.mine.nu:7001
or.
server=100:sr5:pgllr.mine.nu:7001
You will end up on the same server.
The work is k*5^n-1
There n= 300k- 350k
Base=5
Please come and help me test this 36 hr from now.
Dont go for more then 1 wu cache before you see the time.
/Lennart
Can you give manual credit for this? (I am test-running 4 wu's (quad-core) as a test and they are quite long...)
____________
|
|
|
|
|
Hi everyone :)
/Lennart
Can you give manual credit for this? (I am test-running 4 wu's (quad-core) as a test and they are quite long...)
Yes i can, but only those 36 hr.
Lennart |
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1949
ID: 352
Credit: 6,014,682,980
RAC: 1,559,051
|
|
Fired up 4 instances on E5420, a test takes almost an hour.
Will leave it running another ~24 hours...if I don't forget to reverb back....
____________
My stats |
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1217
ID: 18646
Credit: 859,608,218
RAC: 206,541
|
|
I am with you:
[15.04.2009 16:34:02][sysadm@ubuntu-quad:~/prpclient_test]
$ ./prpclient-1.0.5-x86_64-linux
[2009-04-15 14:34:12 GMT] PRPNet Client application v1.0.5 started
[2009-04-15 14:34:12 GMT] User email address is ****
[2009-04-15 14:34:12 GMT] sr5: Getting work from server pgllr.mine.nu at port 7001
This is the new PRPNet Server version 1.0.5
SR5 PRP Test Server
Phil Carmody's Phrot (0.64)
Input 170446*5^317141-1 : Actually testing 532643750*390625^39642-1 (witness=3 39644/81920 limbs)
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
|
|
|
|
|
|
is this something for the 64 bit machines?
should i bother running it on my 32 bit AMD XP 1700? |
|
|
|
|
is this something for the 64 bit machines?
should i bother running it on my 32 bit AMD XP 1700?
It works great on both 32Bit and 64Bit OS.
Not faster on 64bit :)
/Lennart |
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1217
ID: 18646
Credit: 859,608,218
RAC: 206,541
|
|
After ~27h and 22 tests I need the core for another project.
Sorry for not crunching 36h but I hope my 27h help you ...
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
|
|
|
|
|
|
I thank you all who helped out with sr5 work.
We have done ~1000 numbers and thats good. :)
/Lennart |
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1949
ID: 352
Credit: 6,014,682,980
RAC: 1,559,051
|
|
I've done some 200 tests on sr5 work, reverting back to ports 7171/10000
____________
My stats |
|
|
|
|
|
Where do my stats for PRPNet work show up?
____________
FUCK THIS PROJECT AND THEIR COMMIE BULLSHIT POLITICS |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
Where do my stats for PRPNet work show up?
Stats for PRPNet currently are not available. However, a new release of PRPNet will be coming out soon and we'll have access to stats then.
____________
|
|
|
|
|
Where do my stats for PRPNet work show up?
Stats for PRPNet currently are not available. However, a new release of PRPNet will be coming out soon and we'll have access to stats then.
Sounds good. As near as I can tell, 1000 tests or so on my Q6600 w/ PRPNet and no primes yet, but stats are always a good thing to look at. :)
____________
FUCK THIS PROJECT AND THEIR COMMIE BULLSHIT POLITICS |
|
|
|
|
|
I found my first PRPNet prime. :)
http://primes.utm.edu/primes/page.php?id=87807
____________
FUCK THIS PROJECT AND THEIR COMMIE BULLSHIT POLITICS |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
I found my first PRPNet prime. :)
http://primes.utm.edu/primes/page.php?id=87807
Congratulations!!! Not bad for only a 1000 or so tests. :)
____________
|
|
|
|
|
|
[2009-04-21 20:52:42 GMT] pgllr.mine.nu:10000 connect to socket failed
____________
FUCK THIS PROJECT AND THEIR COMMIE BULLSHIT POLITICS |
|
|
|
|
|
geeknik is a little Lucky Luke I think. 500 Test for a prime at PPS and now here a prime after again only a few tests.
:)
____________
 |
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 24,054,820
RAC: 0
|
|
I once found a prime in the first workunit I crunshed with prpnet on a new port, if I wasn't lucky!
____________
|
|
|
|
|
|
I found a couple more primes w/ PRPNet, but I haven't received an email yet. They look to be around 150K digits in length. :)
____________
FUCK THIS PROJECT AND THEIR COMMIE BULLSHIT POLITICS |
|
|
|
|
|
PRPNet port 6868 is down !!!!!!!!!!!!!!!!!!!!!!!!!!
We are investigating a problem with phrot.
GCW13 will be down to we have solved the issue.
/Lennart
|
|
|
|
|
|
Why do the stats @ http://pgllr.mine.nu:7171/user_stats.html show the first part of my email address instead of my nickname?
____________
FUCK THIS PROJECT AND THEIR COMMIE BULLSHIT POLITICS |
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1255
ID: 12001
Credit: 18,565,548
RAC: 0
|
Why do the stats @ http://pgllr.mine.nu:7171/user_stats.html show the first part of my email address instead of my nickname?
Did you put your e-mail address as the userid instead of your nickname? |
|
|
|
|
Did you put your e-mail address as the userid instead of your nickname?
email=brian.carpenter@xxx.com
clientid=geeknik
This is how it is on all of my PRPNet clients. :)
____________
FUCK THIS PROJECT AND THEIR COMMIE BULLSHIT POLITICS |
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1255
ID: 12001
Credit: 18,565,548
RAC: 0
|
|
The problem here is that if you have a 1.x client connecting to a 2.x server that the 2.x server takes the first portion of the e-mail address and uses that as your userid. The 2.x client has a separate userid setting in the ini file. clientid is only used to distinguish between individual CPUs. |
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 24,054,820
RAC: 0
|
|
hello,
I found a bug in prpclient 2.0.3,
This picture is taken more than 1 hour after the test, prpnet didn't continue with the next test but stopped.
Just found out the worst part, the prime isn't even reported to primegrid :O
This are the last lines of prpclient.log:
[2009-04-26 18:42:13 GMT] pps9k: Getting work from server pgllr.mine.nu at port 9000
[2009-04-26 18:42:24 GMT] pps10k: Getting work from server pgllr.mine.nu at port 10000
[2009-04-26 18:42:36 GMT] pps10k: 6985*2^86750+1 is not prime. Residue BB6987CB10227B83
[2009-04-26 18:42:45 GMT] pps10k: 8647*2^86752+1 is not prime. Residue 7A3582E2CDC3B6A5
[2009-04-26 18:42:53 GMT] pps10k: 6805*2^86752+1 is not prime. Residue 9A82A49566229B66
[2009-04-26 18:43:02 GMT] pps10k: 8903*2^86749+1 is not prime. Residue 111F83C334696A66
[2009-04-26 18:43:11 GMT] pps10k: 7213*2^86750+1 is not prime. Residue 7EA91093DDDF361E
[2009-04-26 18:43:19 GMT] pps10k: 5175*2^86750+1 is Prime!
[2009-04-26 18:43:19 GMT] pps10k: PFGW will test for GFN divisors
[2009-04-26 20:26:40 GMT] Accepted force quit. Waiting to close sockets before exiting
[2009-04-26 20:28:10 GMT] pps10k: Attempting to send previously completed work
[2009-04-26 20:28:10 GMT] pps10k: Returning work to server pgllr.mine.nu at port 10000
____________
|
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1255
ID: 12001
Credit: 18,565,548
RAC: 0
|
|
I can't reproduce this issue right now. Which process was taking CPU? Was there anything in the .out file from PFGW? Which version of PFGW? |
|
|
|
|
|
Is the version 2 client more of a beta or is it ready for mainstream use? What has been corrected/added to this build, as opposed to 1.0.5?
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
Is the version 2 client more of a beta or is it ready for mainstream use? What has been corrected/added to this build, as opposed to 1.0.5?
See Mark's Mersenne post here:
http://www.mersenneforum.org/showpost.php?p=171133&postcount=115
There is an unresolved bug for Proth primes being tested for xGFN divisibility. We need more feedback about when this bug occurs. However, if you wish to run error free, just comment out the pfgw line in the prpclient.ini file:
pfgwexe=
to
//pfgwexe=
The main benefits are extra functionality for prime testing and the addition of basic stats.
We are working on a PrimeGrid release and will get it out as soon as possible.
____________
|
|
|
vasmVolunteer tester
Send message
Joined: 6 Dec 08
Posts: 47
ID: 32604
Credit: 990,892
RAC: 0
|
|
The linux 2.03 prpclient zip from http://pgllr.mine.nu/PRPNet/ has a folder inside named v.2.02 and contains what appears to be the windows client (prpclient_win.exe)
Can someone have a look.. |
|
|
|
|
The linux 2.03 prpclient zip from http://pgllr.mine.nu/PRPNet/ has a folder inside named v.2.02 and contains what appears to be the windows client (prpclient_win.exe)
Can someone have a look..
We are just now updating to PRPNet 2.0.4
I will remove 2.0.3
/Lennart
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1949
ID: 352
Credit: 6,014,682,980
RAC: 1,559,051
|
|
Updated to 2.0.4
Seems like stats are not available right now...
____________
My stats |
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1217
ID: 18646
Credit: 859,608,218
RAC: 206,541
|
|
Updated to 2.0.4 on Ubuntu-Linux 64-bit
found first prime - pfgw runs without crashing the client
good work so far!
EDIT: cmdline-shot
Starting Proth prime test of 2841*2^88864+1
Using FFT length 6K, a = 5
2841*2^88864+1 is prime! Time : 9.966 sec.
[2009-04-28 16:08:27 GMT] PPSE11K: 2841*2^88864+1 is Prime!
[2009-04-28 16:08:27 GMT] PPSE11K: PFGW will test for GFN divisors
PFGW Version 1.2.0 for Pentium and compatibles [FFT v23.8]
Output logging to file work_PPSE11K.out
No factoring at all, not even trivial division
2841*2^88864+1 is 3-PRP! (39.8544s+0.0001s)
[2009-04-28 16:12:26 GMT] Total Time: 0:25:08 Total Tests: 106 Total PRPs Found: 1
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
PRPNet 2.0.4 has been released
Please upgrade to the new PRPNet 2.0.4. Some of the new features include:
- basic stats
- port specific max WU setting. This will allow you to set individual max WU's per project instead of the old global one.
server=<suffix>:<pct>:<workunits>:<server IP>:<port>
- (x)(G)FN testing of Proth primes using PFGW
- support for primorial/factorial prime searches
Latest builds
* prpclient-2.0.4-linux (32 bit - will run on 64 bit)
* prpclient-2.0.4-windows (32 bit - will run on 64 bit)
Download the client for your OS and extract. You can download the archives here: http://pgllr.mine.nu/PRPNet/
If you are running multiple cores, then look here for multicore installion files: http://pgllr.mine.nu/PRPNet/multicore/
NOTE: These packages do not include phrot. A new 0.70 version has been released but we have yet to test it. As soon as testing is done, we'll include it.
The first basic stats can be found here:
port 7171: Server : User
port 9000: Server : User
port 10000: Server : User
port 11000: Server : User
BUG REPORT: After starting the windows client from a double click, when you go to close with a ctrl-c, you DO NOT get the "closing" options. However, if you start the windows client from a command line, when you go to close with a ctrl-c, you DO get the "closing" options.
____________
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1949
ID: 352
Credit: 6,014,682,980
RAC: 1,559,051
|
|
Port 10000 is not working, port 11000 and 717 are fine.
____________
My stats |
|
|
|
|
Port 10000 is not working, port 11000 and 717 are fine.
Should be fixed now.
/Lennart |
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1949
ID: 352
Credit: 6,014,682,980
RAC: 1,559,051
|
|
Thanks.
Can we have user statistics for each port for a bit longer? It doesn't make much sense to reset them every now and then...
____________
My stats |
|
|
|
|
|
The links in message #15181 don't work
____________
There are only 10 kinds of people - those who understand binary and those who don't
|
|
|
|
|
|
Upgrading to PRPNet 2.04. Can I just overwrite everything? Or do I need to delete everything first, or what? Thanks.
____________
FUCK THIS PROJECT AND THEIR COMMIE BULLSHIT POLITICS |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
Puzzle-Peter wrote: The links in message #15181 don't work
The PRPNet server was without power for a while. All links should work now.
geeknik wrote: Upgrading to PRPNet 2.04. Can I just overwrite everything? Or do I need to delete everything first, or what? Thanks.
It's best for a clean install.
____________
|
|
|
pschoeferVolunteer developer
Volunteer tester
Send message
Joined: 20 Sep 05
Posts: 685
ID: 845
Credit: 2,886,414,412
RAC: 77,022
|
|
Looks like there are still big problems with the stats... user stats for port 9000 were just partially resetted. harvey563 shows up three times (harvey563, rvey563 and ey563) and I think some of my >500 tests before the reset are now listed under the name efer. :|
____________
|
|
|
|
|
Looks like there are still big problems with the stats... user stats for port 9000 were just partially resetted. harvey563 shows up three times (harvey563, rvey563 and ey563) and I think some of my >500 tests before the reset are now listed under the name efer. :|
There is some issue with stats. I have to investigate this.
Its no problem all is in result files and logfiles.
/Lennart |
|
|
vasmVolunteer tester
Send message
Joined: 6 Dec 08
Posts: 47
ID: 32604
Credit: 990,892
RAC: 0
|
|
I'm having some stop/resume problems despite starting client from command line (prpclient 2.0.4/WinXP SP3 32bit).
Running Primorial (asked/received 3 tests, pressed Ctrl+c during the third test)
D:\Software\PrimeGrid\PRPNet\core1>prpclient.exe
Client will run in idle class.
[2009-05-04 12:11:24 GMT] PRPNet Client application v2.0.4 started
[2009-05-04 12:11:24 GMT] User name vassilism at email address is __
[2009-05-04 12:11:25 GMT] PRIM: Getting work from server pgllr.mine.nu at port 9191
This is the new PRPNet Server version 2.0.4 9191
PFGW Version 1.2.0 for Windows [FFT v23.8]
Output logging to file work_PRIM.out
No factoring at all, not even trivial division
90359#-1 is composite: [2386D7418237DA6C] (88.7005s+0.0262s)
[2009-05-04 12:12:58 GMT] PRIM: 90359#-1 is not prime. Residue 2386D7418237DA6C
PFGW Version 1.2.0 for Windows [FFT v23.8]
Output logging to file work_PRIM.out
No factoring at all, not even trivial division
90371#-1 is composite: [3B5413D243A15D03] (87.9071s+0.0263s)
[2009-05-04 12:14:26 GMT] PRIM: 90371#-1 is not prime. Residue 3B5413D243A15D03
PFGW Version 1.2.0 for Windows [FFT v23.8]
Output logging to file work_PRIM.out
No factoring at all, not even trivial division
PRP: 90403#-1 67500/129956 mro=0.013672 sum=4023.26/29.00
^C Handler Routine called, shutting down program
What should the client do with untested workunits?
1 = Abort all (including current tests)
2 = Report completed tests to server
3 = Nothing
Choose option: [2009-05-04 12:15:14 GMT] PRIM: No data in file [work_PRIM.out]. Is PFGW broken? Client waited at this point but pressing 2 didn't do anything as there were no reporting messages and those 2 completed tests do not appear in my stats
When restarting, the client asked straight for new work and left behind XXXXXXX.pfr file.
Running PPSE10K (asked/received 5 tests, pressed Ctrl+c during the second test)
D:\Software\PrimeGrid\PRPNet\core1>prpclient.exe
Client will run in idle class.
[2009-05-04 12:36:33 GMT] PRPNet Client application v2.0.4 started
[2009-05-04 12:36:33 GMT] User name vassilism at email address is __
[2009-05-04 12:36:33 GMT] PPSE10K: Getting work from server pgllr.mine.nu at port 10000
This is the new PRPNet Server version 2.0.3
Starting Proth prime test of 9683*2^95069+1
Using FFT length 8K, a = 3
9683*2^95069+1 is not prime. Proth RES64: 28F2A88F4C4B3900 Time : 11.522 sec.
[2009-05-04 12:36:52 GMT] PPSE10K: 9683*2^95069+1 is not prime. Residue 28F2A88F4C4B3900
Starting Proth prime test of 9761*2^95069+1
Using FFT length 8K, a = 3
9761*2^95069+1, bit: 20000 / 95082 [21.03%]. Time per bit: 0.155 ms.
Caught signalWhat should the client do with untested workunits?
. 1 = Abort all (including current tests)
T 2 = Report completed tests to server
er 3 = Nothing
miChoose option: nating.
[2009-05-04 12:36:56 GMT] PPSE10K: Could not open file [lresults.txt] for reading. Assuming user stopped
with ^C Client exited without waiting for option.
This time restarting the client returned the 1 completed result and then asked for new work and left behind zXXXXXXX file. |
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1255
ID: 12001
Credit: 18,565,548
RAC: 0
|
|
There is a ^C with the Windows build. I hope to have it fixed in the next release. |
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1255
ID: 12001
Credit: 18,565,548
RAC: 0
|
There is a ^C with the Windows build. I hope to have it fixed in the next release.
I meant a ^C issue. I think that what I will need to do is ask the user what they want to do with unfinished (started by not completed) work from a previous run when they start the client. I'll have to think about it. |
|
|
|
|
|
Hey guys, I'm having a problem with the affinity. I've got an i7 and I'm running the quad 2.0.4 windows version. If I set the affinity manually in the task manager for each instance of LLR then I get over a 30% boost in speed.
However, if I set the affinity for the LLRs in each of the prpclient.ini files, it just doesn't work. I can see from the llr.ini files that the affinity setting is present there, but it seems LLR just ignores it.
Here's the llr.ini file for prpclient-1:
Work=0
PgenInputFile=work_12121.in
PgenOutputFile=work_12121.out
PgenLine=1
WorkDone=0
Affinity=1
And here's the appropriate section from the prpclient.ini folder:
// This sets the CPU affinity for LLR on multi-CPU machines.
affinity=1
What can I do to make these settings kick in? As far as I can tell they're just ignored. |
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1255
ID: 12001
Credit: 18,565,548
RAC: 0
|
Hey guys, I'm having a problem with the affinity. I've got an i7 and I'm running the quad 2.0.4 windows version. If I set the affinity manually in the task manager for each instance of LLR then I get over a 30% boost in speed.
However, if I set the affinity for the LLRs in each of the prpclient.ini files, it just doesn't work. I can see from the llr.ini files that the affinity setting is present there, but it seems LLR just ignores it.
Here's the llr.ini file for prpclient-1:
Work=0
PgenInputFile=work_12121.in
PgenOutputFile=work_12121.out
PgenLine=1
WorkDone=0
Affinity=1
And here's the appropriate section from the prpclient.ini folder:
// This sets the CPU affinity for LLR on multi-CPU machines.
affinity=1
What can I do to make these settings kick in? As far as I can tell they're just ignored.
AFAICT, the setting is correct, so I don't know why LLR ignores it. That would be a question for Jean Penne. |
|
|
vasmVolunteer tester
Send message
Joined: 6 Dec 08
Posts: 47
ID: 32604
Credit: 990,892
RAC: 0
|
|
After the windows ^C issue described above, I tried setting up a client on a linux machine that runs 24/7 so stopping/resuming wouldn't be an issue.
However I bumped into some other problems.
OS is Ubuntu 9.04 64bit.
Primorial on port 9191 downloads a test, hangs for a couple of minutes and then closes.
pfgw_err.log
-----------------------------------------------------------------------
Error occuring in PFGW at Wed May 6 00:03:30 2009
Expr = 154723#-1
Failed at bit 1 of 222734
Msg = Detected in gw_test_mismatched_sums() in gwprp.cpp (PRP Testing)
Iteration: 1/222734 ERROR: SUM(INPUTS) != SUM(OUTPUTS),
9 != 9
(Diff=0 max allowed=8465)
(Test aborted, try again using the -a1 switch)
LLR is working ok.
I see there are about 10 people returning results on that port. Anyone running primorial on linux? If others are running ok, then there must be an issue on my side. Any advice would be appreciated.
PS1. Is there a technical reason for the script in 1-single-start-prpclient.sh that launches in xterm? Any reason not to use ./prpclient from the standard gnome terminal?
PS2. A quick attempt to run the 2.0.4 client on a Ubuntu 8.10 32bit machine (that was running 1.0.5 without problems) failed even more impressively. The window attempts to launch and instantly closes. I don't have time now, but I may post more details tomorrow after I play with it a bit more.
|
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1255
ID: 12001
Credit: 18,565,548
RAC: 0
|
After the windows ^C issue described above, I tried setting up a client on a linux machine that runs 24/7 so stopping/resuming wouldn't be an issue.
However I bumped into some other problems.
OS is Ubuntu 9.04 64bit.
Primorial on port 9191 downloads a test, hangs for a couple of minutes and then closes.
pfgw_err.log
-----------------------------------------------------------------------
Error occuring in PFGW at Wed May 6 00:03:30 2009
Expr = 154723#-1
Failed at bit 1 of 222734
Msg = Detected in gw_test_mismatched_sums() in gwprp.cpp (PRP Testing)
Iteration: 1/222734 ERROR: SUM(INPUTS) != SUM(OUTPUTS),
9 != 9
(Diff=0 max allowed=8465)
(Test aborted, try again using the -a1 switch)
LLR is working ok.
I see there are about 10 people returning results on that port. Anyone running primorial on linux? If others are running ok, then there must be an issue on my side. Any advice would be appreciated.
PS1. Is there a technical reason for the script in 1-single-start-prpclient.sh that launches in xterm? Any reason not to use ./prpclient from the standard gnome terminal?
PS2. A quick attempt to run the 2.0.4 client on a Ubuntu 8.10 32bit machine (that was running 1.0.5 without problems) failed even more impressively. The window attempts to launch and instantly closes. I don't have time now, but I may post more details tomorrow after I play with it a bit more.
I am aware of the issue and will fix it in the next release. It is an issue with PFGW that the client needs to catch and address.
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
PS1. Is there a technical reason for the script in 1-single-start-prpclient.sh that launches in xterm? Any reason not to use ./prpclient from the standard gnome terminal?
No technical reason...just wanted to be consistent across the builds.
EDIT: Feel free to use whatever start-up you like. If there's something else you'd like to see, please let us know. :)
PS2. A quick attempt to run the 2.0.4 client on a Ubuntu 8.10 32bit machine (that was running 1.0.5 without problems) failed even more impressively. The window attempts to launch and instantly closes. I don't have time now, but I may post more details tomorrow after I play with it a bit more.
Double check all the permissions.
____________
|
|
|
|
|
Hey guys, I'm having a problem with the affinity. I've got an i7 and I'm running the quad 2.0.4 windows version. If I set the affinity manually in the task manager for each instance of LLR then I get over a 30% boost in speed.
However, if I set the affinity for the LLRs in each of the prpclient.ini files, it just doesn't work. I can see from the llr.ini files that the affinity setting is present there, but it seems LLR just ignores it.
Here's the llr.ini file for prpclient-1:
Work=0
PgenInputFile=work_12121.in
PgenOutputFile=work_12121.out
PgenLine=1
WorkDone=0
Affinity=1
And here's the appropriate section from the prpclient.ini folder:
// This sets the CPU affinity for LLR on multi-CPU machines.
affinity=1
What can I do to make these settings kick in? As far as I can tell they're just ignored.
AFAICT, the setting is correct, so I don't know why LLR ignores it. That would be a question for Jean Penne.
Seems you're right, I put the affinity setting into an llr.ini file for the regular GUI version of LLR, and when run it's pinned to the correct CPU.
I did the same thing for cLLR but got no joy. I even tried adding -oaffinity=x to the command line. |
|
|
vasmVolunteer tester
Send message
Joined: 6 Dec 08
Posts: 47
ID: 32604
Credit: 990,892
RAC: 0
|
|
I see Port 6868 is alive again sending GCW13 work at n~158k.
Is this double-checking because of the phrot problem or was it planned anyway?
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
I see Port 6868 is alive again sending GCW13 work at n~158k.
Is this double-checking because of the phrot problem or was it planned anyway?
Yes, this is double checking. A full double check is being completed until we are comfortable that the residue bug in phrot did not affect GCW13.
All previous versions of phrot will not be accepted. You MUST update to the latest phrot 0.70 here:
phrot-0.70-bin.zip
Unzip and rename the build you need to phrot-0.70. Then remember to update your master_prpclient.ini and udpate.
phrotexe=phrot-0.70.exe (Windows)
phrotexe=./phrot-0.70 (Linux)
Please let us know if you have any questions.
NOTE: PRPNet 2.1.0 alpha will be out as soon as we can get a Linux build...so be sure to check back. We will need testers for this.
____________
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1949
ID: 352
Credit: 6,014,682,980
RAC: 1,559,051
|
|
User stats on port 9191 got resetted and messed-up...
____________
My stats |
|
|
|
|
|
I have been running the 2.0.4 client for windows for a while now, doing PPSE 10000 and 11000, and twice now I have had the window that the client is running in close. When I check the log I see this as the last entry:
[2009-05-07 20:31:00 GMT] PPSE11K: Could not find RES64 residue [lresults.txt]. Is llr broken?
Other then these two times, it runs like a champ. Nothing out of the ordinary is running when this shuts down. Any ideas why this is happening? Thanks.
____________
|
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1255
ID: 12001
Credit: 18,565,548
RAC: 0
|
I have been running the 2.0.4 client for windows for a while now, doing PPSE 10000 and 11000, and twice now I have had the window that the client is running in close. When I check the log I see this as the last entry:
[2009-05-07 20:31:00 GMT] PPSE11K: Could not find RES64 residue [lresults.txt]. Is llr broken?
Other then these two times, it runs like a champ. Nothing out of the ordinary is running when this shuts down. Any ideas why this is happening? Thanks.
This occurs if the PRPNet client cannot find the residue of the test. It presumes that the user cancelled. When this happens, what does the terminal display? What are the contents of the work*.out files? |
|
|
|
|
|
All servers down again? |
|
|
|
|
This occurs if the PRPNet client cannot find the residue of the test. It presumes that the user cancelled. When this happens, what does the terminal display? What are the contents of the work*.out files?
It happened again last night, I opened the prpclient with a double click, and the terminal window opened, running just fine, so i let it be. I checked it this morning and the terminal window was gone. Looked at the log and the last line once again said:
[2009-05-07 21:54:43 GMT] PPSE11K: Could not find RES64 residue [lresults.txt]. Is llr broken?
I looked at the project files, (there are only ones for PPSE11K) and here are the contents:
work_PPSE11K.in
ABC $a*$b^$c+1
7833 2 89796
work_PPSE11K.out
ABC $a*$b^$c+1
work_PPSE11K.save
ServerName=pgllr.mine.nu
PortID=11000
ServerVersion=2.1.0
WorkUnit=7833*2^89796+1 1241728199 na notdone na
WorkUnit=1581*2^89800+1 1241728199 na notdone na
WorkUnit=3263*2^89797+1 1241728199 na notdone na
WorkUnit=7953*2^89797+1 1241728199 na notdone na
WorkUnit=8019*2^89798+1 1241728199 na notdone na
WorkUnit=9045*2^89798+1 1241728199 na notdone na
WorkUnit=7507*2^89796+1 1241728199 na notdone na
WorkUnit=7767*2^89796+1 1241728199 na notdone na
WorkUnit=6765*2^89798+1 1241728199 na notdone na
WorkUnit=1377*2^89800+1 1241728199 na notdone na
WorkUnit=9333*2^89798+1 1241728199 na notdone na
WorkUnit=8827*2^89798+1 1241728199 na notdone na
WorkUnit=6659*2^89797+1 1241728199 na notdone na
WorkUnit=9965*2^89797+1 1241728199 na notdone na
WorkUnit=7509*2^89797+1 1241728199 na notdone na
WorkUnit=9031*2^89796+1 1241728199 na notdone na
WorkUnit=5319*2^89798+1 1241728199 na notdone na
WorkUnit=1625*2^89799+1 1241728199 na notdone na
WorkUnit=9365*2^89797+1 1241728199 na notdone na
WorkUnit=9543*2^89797+1 1241728199 na notdone na
WorkUnit=4409*2^89797+1 1241728199 na notdone na
WorkUnit=7195*2^89798+1 1241728199 na notdone na
WorkUnit=7979*2^89797+1 1241728199 na notdone na
WorkUnit=7743*2^89797+1 1241728199 na notdone na
WorkUnit=3705*2^89798+1 1241728199 na notdone na
WorkUnit=7743*2^89796+1 1241728199 na notdone na
WorkUnit=4173*2^89798+1 1241728199 na notdone na
WorkUnit=4269*2^89797+1 1241728199 na notdone na
WorkUnit=3071*2^89797+1 1241728199 na notdone na
WorkUnit=8517*2^89796+1 1241728199 na notdone na
lresults.txt
Bit: 52224/89808, ERROR: ROUND OFF (0.4375) > 0.40
Continuing from last save file.
Disregard last error. Result is reproducible and thus not a hardware problem.
For added safety, redoing iteration using a slower, more reliable method.
Continuing from last save file.
7833*2^89796+1 is not prime. Proth RES64: 4BD94D322400736C Time : 6.479 sec.
Let me know if there is any more info that I can provide.
____________
|
|
|
|
|
All servers down again?
There is some stupid DNS error I don't know why.
I shall see if i can solve it.
It should work now
Lennart |
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1255
ID: 12001
Credit: 18,565,548
RAC: 0
|
This occurs if the PRPNet client cannot find the residue of the test. It presumes that the user cancelled. When this happens, what does the terminal display? What are the contents of the work*.out files?
It happened again last night, I opened the prpclient with a double click, and the terminal window opened, running just fine, so i let it be. I checked it this morning and the terminal window was gone. Looked at the log and the last line once again said:
[2009-05-07 21:54:43 GMT] PPSE11K: Could not find RES64 residue [lresults.txt]. Is llr broken?
I looked at the project files, (there are only ones for PPSE11K) and here are the contents:
work_PPSE11K.in
ABC $a*$b^$c+1
7833 2 89796
work_PPSE11K.out
ABC $a*$b^$c+1
lresults.txt
Bit: 52224/89808, ERROR: ROUND OFF (0.4375) > 0.40
Continuing from last save file.
Disregard last error. Result is reproducible and thus not a hardware problem.
For added safety, redoing iteration using a slower, more reliable method.
Continuing from last save file.
7833*2^89796+1 is not prime. Proth RES64: 4BD94D322400736C Time : 6.479 sec.
Let me know if there is any more info that I can provide.
That's very helpful. I'll fix it in the next beta.
|
|
|
vasmVolunteer tester
Send message
Joined: 6 Dec 08
Posts: 47
ID: 32604
Credit: 990,892
RAC: 0
|
|
Testing 2.1.1 with Primorial
Windows XP 32bit
Client gets stuck testing the same workunit over and over without reporting anything.
Client will run in idle class.
[2009-05-08 22:37:47 GMT] PRPNet Client application v2.1.1 started
[2009-05-08 22:37:47 GMT] User name vasm at email address is _
[2009-05-08 22:37:47 GMT] PRS: Getting work from server pgllr.mine.nu at port 9191
This is the new PRPNet Server version 2.0.4 9191
PFGW Version 1.2.0 for Windows [FFT v23.8]
Output logging to file work_PRS.out
No factoring at all, not even trivial division
212099#+1 is composite: [2C9C89C7B62D35D1] (470.5700s+0.1335s)
PFGW Version 1.2.0 for Windows [FFT v23.8]
Output logging to file work_PRS.out
No factoring at all, not even trivial division
212099#+1 is composite: [2C9C89C7B62D35D1] (682.3016s+0.1339s)
PFGW Version 1.2.0 for Windows [FFT v23.8]
Output logging to file work_PRS.out
No factoring at all, not even trivial division
212099#+1 is composite: [2C9C89C7B62D35D1] (849.4100s+0.1341s)
PFGW Version 1.2.0 for Windows [FFT v23.8]
Output logging to file work_PRS.out
No factoring at all, not even trivial division
PRP: 212099#+1 5000/305462 mro=5.722e-006 sum=344.23/0.00
etc
It is also totally unresponsive to ^C.
Output logging to file work_PRS.out
No factoring at all, not even trivial division
PRP: 212099#+1 10000/305462 mro=5.722e-006 sum=344.23/0.00
^C Handler Routine called, shutting down program
PFGW Version 1.2.0 for Windows [FFT v23.8]
Output logging to file work_PRS.out
No factoring at all, not even trivial division
Resuming at bit 11634
PRP: 212099#+1 12500/305462 mro=5.722e-006 sum=344.23/0.00
^C Handler Routine called, shutting down program
PFGW Version 1.2.0 for Windows [FFT v23.8]
Output logging to file work_PRS.out
No factoring at all, not even trivial division
Resuming at bit 12884
PRP: 212099#+1 12885/305462 mro=3.8147e-006 sum=344.23/0.00
^C Handler Routine called, shutting down program
PFGW Version 1.2.0 for Windows [FFT v23.8]
Output logging to file work_PRS.out
No factoring at all, not even trivial division
Resuming at bit 14295
PRP: 212099#+1 15000/305462 mro=5.722e-006 sum=344.23/0.00
etc
Ubuntu 9.04 64bit
Client starts and after 2-3 minutes produces again the (aborted, try using -a1 switch) error. Looks like it tries to restart and then hangs.
[2009-05-08 22:51:28 GMT] PRPNet Client application v2.1.1 started
[2009-05-08 22:51:28 GMT] User name vasm at email address is _
[2009-05-08 22:51:28 GMT] PRS: Getting work from server pgllr.mine.nu at port 9191
This is the new PRPNet Server version 2.0.4 9191
PFGW Version 1.2.0 for Pentium and compatibles [FFT v23.8]
Output logging to file work_PRS.out
No factoring at all, not even trivial division
*** WARNING *** : An error was detected in the reciprocal routine.
If this occurred during normal use, contact bugs@primeform.net,
including the number you were testing to produce this error,
and disregard the results of tests on this number.
If this error occurs in optimization mode, you may ignore it.
PRP: 212437#-1 1/305869 mro=0.5 sum=12590.08/0.00
Iteration: 1/305869 ERROR: SUM(INPUTS) != SUM(OUTPUTS),
9 != 9
(Diff=0 max allowed=12590)
(Test aborted, try again using the -a1 switch)
Signal SIGILL caught
PFGW Version 1.2.0 for Pentium and compatibles [FFT v23.8]
Output logging to file work_PRS.out
No factoring at all, not even trivial division
CPU is at 100% but no progress reporting and after 30min (that test should have finished by then) I tried ^C.
Not responding.
^C
Ctrl-C detected, shutting down the program.
PFGW Version 1.2.0 for Pentium and compatibles [FFT v23.8]
Output logging to file work_PRS.out
No factoring at all, not even trivial division
^C
Ctrl-C detected, shutting down the program.
PFGW Version 1.2.0 for Pentium and compatibles [FFT v23.8]
Output logging to file work_PRS.out
No factoring at all, not even trivial division
^C
Ctrl-C detected, shutting down the program.
PFGW Version 1.2.0 for Pentium and compatibles [FFT v23.8]
Output logging to file work_PRS.out
No factoring at all, not even trivial division
^C
etc
I closed the terminal window but pfgw stayed resident in the background using 100% CPU and kill pid could not get rid of it.
|
|
|
pschoeferVolunteer developer
Volunteer tester
Send message
Joined: 20 Sep 05
Posts: 685
ID: 845
Credit: 2,886,414,412
RAC: 77,022
|
Testing 2.1.1 with Primorial
Windows XP 32bit
Client gets stuck testing the same workunit over and over without reporting anything.
It is also totally unresponsive to ^C.
Same behavior on Win XP x64. Unfortunately, I had port 9191 as fallback for 7171 and 9000 and these ports were down at 22h UTC, so two cores of my quad were doing the same test for almost 9 hours. :(
Had to kill pfgw.exe in the taskmanager.
____________
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1949
ID: 352
Credit: 6,014,682,980
RAC: 1,559,051
|
|
It appears option 2 does nothing; option 1 does report and continue.
It appears that the PRPNet client was aborted without completing
the assigned workunits. What do you want to do?
1 = Report completed and abort the rest
2 = Complete assigned workunits
Choose option: 2
It appears that the PRPNet client was aborted without completing
the assigned workunits. What do you want to do?
1 = Report completed and abort the rest
2 = Complete assigned workunits
Choose option: 1
[2009-05-09 07:30:27 GMT] PPSE11K: Attempting to send previously completed work
[2009-05-09 07:30:28 GMT] PPSE11K: Returning work to server pgllr.mine.nu at port 11000
[2009-05-09 07:30:29 GMT] PPSE11K: INFO: There were no identifiable test results
[2009-05-09 07:30:40 GMT] PRPNet Client application v2.1.1 started
____________
My stats |
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1255
ID: 12001
Credit: 18,565,548
RAC: 0
|
|
I'll look into these issues today if I have time. I thought that I had resolved them, but apparently not. |
|
|
|
|
It's best for a clean install.
Do I need to upgrade? What's the benefits? Will the server quit accepting my results if I don't upgrade? Only reason I ask is because everything is running smoothly right now. :)
____________
FUCK THIS PROJECT AND THEIR COMMIE BULLSHIT POLITICS |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
It's best for a clean install.
Do I need to upgrade? What's the benefits? Will the server quit accepting my results if I don't upgrade? Only reason I ask is because everything is running smoothly right now. :)
No need to update unless you want to test. However, as you can see, problems exist with v2.1.1. When a stable version is finally ready, then yes, we recommend that you upgrade as the new version has additional functionality. Details about improvements will be provided then.
____________
|
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1217
ID: 18646
Credit: 859,608,218
RAC: 206,541
|
|
I watch a strange behavior after founding a prime:
Starting Proth prime test of 9685*2^89860+1
Using FFT length 8K, a = 3
9685*2^89860+1 is prime! Time : 181.296 sec.
[2009-05-10 16:03:00 GMT] PPSE11K: 9685*2^89860+1 is Prime!
[2009-05-10 16:03:00 GMT] PPSE11K: PFGW will test for GFN divisors
PFGW Version 1.2.0 for Windows [FFT v23.8]
Output logging to file work_PPSE11K.out
No factoring at all, not even trivial division
9685*2^89860+1 is 3-PRP! (259.8663s+0.0002s)
[2009-05-10 16:18:20 GMT] PPSE11K: Quitting due to detected interruption
[2009-05-10 16:18:20 GMT] Total Time: 2:14:18 Total Tests: 128 Total PRPs Found: 0
[2009-05-10 16:18:20 GMT] PPSE11K: Returning work to server pgllr.mine.nu at port 11000
[2009-05-10 16:18:23 GMT] PPSE11K: INFO: Test for candidate 5531*2^89861+1 accepted
[2009-05-10 16:18:24 GMT] PPSE11K: INFO: Test for candidate 6027*2^89862+1 accepted
[2009-05-10 16:18:24 GMT] PPSE11K: INFO: Test for candidate 4923*2^89862+1 accepted
[2009-05-10 16:18:24 GMT] PPSE11K: INFO: Test for candidate 5943*2^89862+1 accepted
[2009-05-10 16:18:24 GMT] PPSE11K: INFO: Test for candidate 7349*2^89861+1 accepted
[2009-05-10 16:18:24 GMT] PPSE11K: INFO: Test for candidate 7619*2^89861+1 accepted
[2009-05-10 16:18:24 GMT] PPSE11K: INFO: Test for candidate 7945*2^89860+1 accepted
[2009-05-10 16:18:24 GMT] PPSE11K: INFO: Test for candidate 9685*2^89860+1 accepted
[2009-05-10 16:18:27 GMT] PPSE11K: INFO: All 8 test results were accepted
Datei nicht gefunden
Datei nicht gefunden
Datei nicht gefunden
Datei nicht gefunden
[2009-05-10 16:18:28 GMT] Client shutdown complete
OS is WinXP 32-bit running on an Celeron M @ 1.5 GHz
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
|
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1255
ID: 12001
Credit: 18,565,548
RAC: 0
|
|
I presume this is the 2.1.1 client. If so, then this is an issue that I haven't fixed yet. I hope to get to it tomorrow. I haven't had time to work on it. |
|
|
|
|
|
i have a similar problem after founding a prime
[2009-05-10 14:15:37 GMT] PRPNet Client application v2.1.1 started
[2009-05-10 14:15:37 GMT] User name [SG-SPEG]BerndBrot at email address is ...
[2009-05-10 14:15:37 GMT] PPSE10K: Getting work from server pgllr.mine.nu at port 10000
[2009-05-10 14:15:52 GMT] PPSE10K: 4295*2^95863+1 is not prime. Residue 940B14665E2ACC6B
[2009-05-10 14:16:04 GMT] PPSE10K: 5443*2^95864+1 is not prime. Residue C46797DDCB18792D
[2009-05-10 14:16:15 GMT] PPSE10K: 9407*2^95863+1 is not prime. Residue 392406D9D8A806F3
[2009-05-10 14:16:27 GMT] PPSE10K: 4705*2^95864+1 is not prime. Residue 7A3BF450BE7AB64F
...
[2009-05-10 15:21:20 GMT] PPSE10K: INFO: Test for candidate 7031*2^95871+1 accepted
[2009-05-10 15:21:20 GMT] PPSE10K: INFO: Test for candidate 5977*2^95870+1 accepted
[2009-05-10 15:21:20 GMT] PPSE10K: INFO: Test for candidate 8289*2^95869+1 accepted
[2009-05-10 15:21:20 GMT] PPSE10K: INFO: Test for candidate 1611*2^95873+1 accepted
[2009-05-10 15:21:20 GMT] PPSE10K: INFO: Test for candidate 7553*2^95869+1 accepted
[2009-05-10 15:21:20 GMT] PPSE10K: INFO: All 100 test results were accepted
[2009-05-10 15:21:20 GMT] PPSE11K: Getting work from server pgllr.mine.nu at port 11000
[2009-05-10 15:21:32 GMT] PPSE11K: 8201*2^89859+1 is Prime!
[2009-05-10 15:21:32 GMT] PPSE11K: PFGW will test for GFN divisors
[2009-05-10 15:24:10 GMT] PPSE11K: Quitting due to detected interruption
[2009-05-10 15:24:10 GMT] Total Time: 1:08:33 Total Tests: 400 Total PRPs Found: 0
[2009-05-10 15:24:10 GMT] PPSE11K: Returning work to server pgllr.mine.nu at port 11000
[2009-05-10 15:24:12 GMT] PPSE11K: INFO: Test for candidate 8201*2^89859+1 accepted
[2009-05-10 15:24:15 GMT] PPSE11K: INFO: All 1 test results were accepted
[2009-05-10 15:24:16 GMT] Client shutdown complete
the client run port 10k and 11k with 50%
____________
 |
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1217
ID: 18646
Credit: 859,608,218
RAC: 206,541
|
