mersenneforum.org Assignment key CE8Dxxxxxxxx744 belongs to another user
 Register FAQ Search Today's Posts Mark Forums Read

 2021-04-19, 23:45 #1 drkirkby   "David Kirkby" Jan 2021 Althorne, Essex, UK 3228 Posts Assignment key CE8Dxxxxxxxx744 belongs to another user I'm not sure what should be public here, and what should not Should assignment keys be kept private? Anyway, until I hear otherwise, I will assume they are, so have no reproduced this in full. Rather foolishly I started to attempt the 100 million digit exponent M332646233 https://www.mersenne.org/report_expo...2646233&full=1 I see in my list of assignments at https://www.mersenne.org/workload/ Code:  PRP=CE8D2xxxxxxxxxxxxxxxxFC744,1,2,332646233,-1,81,0,3,1 However, with mprime running, I thought I would take a look at how things were, so run mprime -m. I see a message that the assignment belongs to another use. Code: [Comm thread Apr 20 00:19] od=0 [Comm thread Apr 20 00:19] ==END== [Comm thread Apr 20 00:19] [Comm thread Apr 20 00:19] Sending expected completion date for M332646233: Aug 6 2021 [Comm thread Apr 20 00:19] URL: http://v5.mersenne.org/v5server/?v=0.95&px=GIMPS&t=ap&g=bbe21397xxxxxxxxxxxx8811ad3&k=CE8D2xxxxxxxxxxFC744&c=0&p=0.0000&d=10800&e=9401651&ss=1547&sh=Fxxxxxxxxxxxxxxxxxxx99E2A [Comm thread Apr 20 00:19] == Info: Trying 162.212.57.131:80... [Comm thread Apr 20 00:19] == Info: Connected to v5.mersenne.org (162.212.57.131) port 80 (#0) [Comm thread Apr 20 00:19] => Send header: GET /v5server/?v=0.95&px=GIMPS&t=ap&g=bbexxxxxxxxxxxxxxxx811ad3&k=CE8DxxxxxxxC744&c=0&p=0.0000&d=10800&e=9401651&ss=1547&sh=F0AEB25A3B115916F95939B699F99E2A HTTP/1.1 [Comm thread Apr 20 00:19] Host: v5.mersenne.org [Comm thread Apr 20 00:19] Accept: */* [Comm thread Apr 20 00:19] [Comm thread Apr 20 00:19] == Info: Mark bundle as not supporting multiuse [Comm thread Apr 20 00:19] <= Recv header: HTTP/1.1 200 OK [Comm thread Apr 20 00:19] <= Recv header: Content-Type: text/html; charset=UTF-8 [Comm thread Apr 20 00:19] <= Recv header: Server: Microsoft-IIS/8.5 [Comm thread Apr 20 00:19] <= Recv header: X-Powered-By: PHP/7.1.10 [Comm thread Apr 20 00:19] <= Recv header: Date: Mon, 19 Apr 2021 23:19:29 GMT [Comm thread Apr 20 00:19] <= Recv header: Content-Length: 116 [Comm thread Apr 20 00:19] <= Recv header: [Comm thread Apr 20 00:19] <= Recv data: pnErrorResult=40 [Comm thread Apr 20 00:19] pnErrorDetail=ap: Assignment key CE8D218xxxxxxxxxxxxxxFC744 belongs to another user [Comm thread Apr 20 00:19] ==END== [Comm thread Apr 20 00:19] The other odd thing is, this shows I have 16 exponents on this machine Code: Below is a report on the work you have queued and any expected completion dates. [Worker thread #1] M332646233, PRP, Fri Aug 6 20:06 2021 M110601373, PRP, Mon Aug 16 19:00 2021 M110390123, PRP, Thu Aug 26 17:27 2021 M110390429, PRP, Sun Sep 5 15:54 2021 M110469433, PRP, Wed Sep 15 14:32 2021 [Worker thread #2] M110600639, PRP, Thu Apr 29 23:26 2021 M110601551, PRP, Sun May 9 22:20 2021 M110274583, PRP, Wed May 19 20:32 2021 M110594479, PRP, Sat May 29 19:26 2021 M108827221, PRP, Tue Jun 8 10:38 2021 M108827293, PRP, Fri Jun 18 01:50 2021 M108827311, PRP, Sun Jun 27 17:02 2021 M108827321, PRP, Wed Jul 7 08:14 2021 M108828857, PRP, Fri Jul 16 23:26 2021 M108828869, PRP, Mon Jul 26 14:38 2021 M108830251, PRP, Thu Aug 5 05:51 2021 The chance that one of the 16 exponents you are testing will yield a Mersenne prime is about 1 in 54761. which includes the big M332646233 on worker 1. Yet https://www.mersenne.org/workload/ shows I have 28 exponents on that machine so where are the other 12 gone? Those completion dates are totally wrong. M110600639 should finish today (April 20th) in about 14 hours, not 9 days later on the 29th April. The big M332646233 should finish in 11 days, not the 10 weeks or so that mprime says.
 2021-04-20, 00:03 #2 slandrum   Jan 2021 California 10001112 Posts If you are not changing your working conditions, then the estimate of completion dates should smooth out as you complete more exponents. It is usually wildly inaccurate until you've completed several exponents. If you change the conditions under which the work is running, the estimates will take a while to settle down again. I would not ever read too much into the expected time of completion. You can probably get a much more accurate back-of-the-hand estimate yourself.
 2021-04-20, 00:23 #3 Prime95 P90 years forever!     Aug 2002 Yeehaw, FL 22×1,873 Posts The server shows that the 100M expo is assigned to you: https://www.mersenne.org/report_expo...ll=1&ecmhist=1 Are all the others showing as assigned to you too? Did some of the exponents accidentally get assigned while your computer was using a different user id?
2021-04-20, 10:42   #4
drkirkby

"David Kirkby"
Jan 2021
Althorne, Essex, UK

2·3·5·7 Posts

Quote:
 Originally Posted by Prime95 The server shows that the 100M expo is assigned to you: https://www.mersenne.org/report_expo...ll=1&ecmhist=1 Are all the others showing as assigned to you too? Did some of the exponents accidentally get assigned while your computer was using a different user id?
I've never used another user ID. Some of these exponents were started on another computer, but have been on this one for at least a week after I manually edited worktodo.txt. During that time a few exponents have been completed, and so the computer connected to the server multiple times.

https://www.mersenne.org/report_expo...0904847&full=1
finished 19th April

https://www.mersenne.org/report_expo...0469433&full=1

Finished on 16th April

https://www.mersenne.org/report_expo...0601373&full=1
finished on 16th April

https://www.mersenne.org/report_expo...0317973&full=1
finished on 14th April

So for over a week, twenty odd exponents have been on the machine, then suddenly it fell to 16. Now I see it's back to the expected 28! I've not edited anything myself during that time. The only thing that's happened during that time is I stuck a laptop on the account, but that has a different name, and is not doing PRP tests.

Code:
[Worker thread #1]
M110825213, PRP, Wed Apr 21 09:20 2021
M110385461, PRP, Thu Apr 29 13:31 2021
M110806637, PRP, Sat May  8 12:35 2021
M110643451, PRP, Sun May 16 17:39 2021
M110600107, PRP, Tue May 25 01:47 2021
M110825587, PRP, Thu Jun  3 00:53 2021
M110825509, PRP, Fri Jun 11 23:59 2021
M110825977, PRP, Sun Jun 20 16:37 2021
M110825251, PRP, Tue Jun 29 08:24 2021
M110823613, PRP, Thu Jul  8 07:01 2021
M110825359, PRP, Sat Jul 17 06:07 2021
M110642027, PRP, Mon Jul 26 04:25 2021
M110826173, PRP, Tue Aug  3 21:24 2021
M110825311, PRP, Thu Aug 12 20:21 2021
M110385683, PRP, Sat Aug 21 04:48 2021
M110825807, PRP, Mon Aug 30 03:54 2021
M108668867, PRP, Tue Sep  7 19:21 2021
M110825149, PRP, Thu Sep 16 12:19 2021
M332646233, PRP, Thu Jun  3 02:44 2021
M110826539, PRP, Sat Jun 12 01:51 2021
M110826091, PRP, Mon Jun 21 00:36 2021
M110806819, PRP, Tue Jun 29 23:13 2021
M110825893, PRP, Thu Jul  8 22:20 2021
M108663719, PRP, Sat Jul 17 13:45 2021
M110806807, PRP, Mon Jul 26 12:33 2021
M110825789, PRP, Wed Aug  4 11:39 2021
M110825257, PRP, Fri Aug 13 10:20 2021
M108819209, PRP, Sun Aug 22 01:21 2021
The chance that one of the 28 exponents you are testing will yield a
Mersenne prime is about 1 in 30907.
Worker 1 has almost finished another exponent, so that should be done in less than 5 hours
Code:
[Worker #1 Apr 20 11:39] Iteration: 100780000 / 110825213 [90.93%], ms/iter:  1.647, ETA: 04:35:48
It will be interesting to see if the number of exponents drops to 27, stays at 28, or makes some random leap up/down.

Dave

Last fiddled with by drkirkby on 2021-04-20 at 10:45

 2021-04-21, 13:57 #5 LaurV Romulan Interpreter     Jun 2011 Thailand 22×17×139 Posts My bet: it will grow After you finish the 332M exponent, which takes much longer than the 1xxM exponents, you will get more assignments for the second worker. This is normal. The number of assignments means nothing, your computer will reserve as much work as it is able to do in the time you set to reserve work for. If you set the time to 7 days, you may get only one exponent per worker, if the work for that exponents take longer than a week, or you can get 700 assignments per worker if the assignments are short (like PRP-CF) and you can do 100 per day. At the end of the 7 days you still get about the same credit (i.e. worth of 7 days of work) and have the same chance to find a prime (more or less... ) Last fiddled with by LaurV on 2021-04-22 at 15:21

 Similar Threads Thread Thread Starter Forum Replies Last Post Dr Sardonicus Soap Box 90 2021-02-02 16:27 hgerhardt PrimeNet 2 2018-05-07 16:32 Unregistered Information & Answers 1 2012-10-18 23:31 EarlGlynn Information & Answers 6 2008-12-25 14:21 KevinLee Hardware 6 2003-12-12 17:06

All times are UTC. The time now is 17:14.

Mon May 17 17:14:41 UTC 2021 up 39 days, 11:55, 0 users, load averages: 2.21, 2.74, 2.79