OFFICIAL "SERVER PROBLEMS" THREAD - Page 65

ET_ · 2014-09-12, 15:24

Quote:

Originally Posted by James Heinrich

Probably, but I'm afraid I don't understand the question. Can you give me an example of what you mean?

Last time I tried to manually submit GMP-ECM results, I got an error message, as if there was not enough information in the results file to correctly recod the range... After all the updates I'm a bit lost, and ask again if something has been done, or the output of the program gves too few info for a correct answer.

Luigi

James Heinrich · 2014-09-12, 15:50

Quote:

Originally Posted by ET_

Last time I tried to manually submit GMP-ECM results, I got an error message, as if there was not enough information in the results file to correctly recod the range... After all the updates I'm a bit lost, and ask again if something has been done, or the output of the program gves too few info for a correct answer.

If you can send me a sample of the output I might be able to answer your question better.

ET_ · 2014-09-12, 16:18

Quote:

Originally Posted by James Heinrich

If you can send me a sample of the output I might be able to answer your question better.

Here is the sample

Thank you

Luigi

preben s · 2014-09-12, 16:56

About 13:30 UTC today (roughly) I tried to get an assignment of 2000 numbers for TF. The transfer stalled before anything was displayed.

Next I tried to get 1000 assignments. Success.

And again another 1000 assignments. Succes again.

Checked the number of lines in worktodo.txt. There were about 1100 lines (it was empty from the beginning).

Now I checked the number of assignments on the server. It showed 9100 TF assignments.
(there were 0 TF entries, when I started)

I have no idea , how this happened.

Madpoo · 2014-09-13, 00:16

Quote:

Originally Posted by preben s

About 13:30 UTC today (roughly) I tried to get an assignment of 2000 numbers for TF. The transfer stalled before anything was displayed.

Next I tried to get 1000 assignments. Success.

And again another 1000 assignments. Succes again.

Checked the number of lines in worktodo.txt. There were about 1100 lines (it was empty from the beginning).

Now I checked the number of assignments on the server. It showed 9100 TF assignments.
(there were 0 TF entries, when I started)

I have no idea , how this happened.

Yeah, you've got a LOT of assigned work. Good luck with it all.

Seriously though, you have 9106 exponents assigned to you and all but the first 6 of those (9100) were assigned at the time you mentioned.

If I had to guess, I'd say there's a difference between the PHP timeout and maybe what IIS has. I wonder if IIS is timing out the server response because it takes too long to generate that output for you, but PHP is actually still running and assigns you the work anyway even though IIS has closed the connection by that point.

Are you sure you didn't ask for 8000 exponents at first? I see that 8000 of those (exactly) were all generated in a row, non-stop. Or maybe you asked for 2000 of them but it sent that request 4 times in a row or something.

Anyway, then there's a roughly 3 minute pause and it gets 1000 more (exactly). Then there's a break of nearly an hour and it gets another 1000.

Does that sound about right with the timing?

It was 15 seconds to assign that first group of 1000 exponents, 2 seconds to assign the last 1000 exponents.

That first batch of 8000 took almost 2 minutes. So I am wondering about the IIS/PHP timeouts for something that long running. I thought I saw PHP was set to a 5 minute timeout but I'm not sure.

Request smaller chunks is my advice.

Madpoo · 2014-09-13, 06:04

There will be a brief outage of the www.mersenne.org site for some maintenance, commencing at approximately 11:10 PM Pacific time (6:10 AM UTC). During this time, web facing features and client communications (the API) will be unavailable for about 5-10 minutes.

It's just normal monthly maintenance. HP and Microsoft updates/fixes, nothing too spectacular. I'll edit this message with an update when it's back online.

EDIT #1: Still going on... It got started 5 minutes late, so we're still about 5 minutes away from being back online.

EDIT #2: Back online, enjoy.

tha · 2014-09-13, 08:32

The most recent cleared page now displays lines twice if a composite factor has been found. (or three times if it consists of three factors)

It did not use to do that.

Code:

alpertron            Manual testing   917789    F-PM1    Sep 12 2014 10:50PM   0.0   0.0162  2311938349319194381898303231634487                                                                                                                                                                      
alpertron            Manual testing   917789    F-PM1    Sep 12 2014 10:50PM   0.0   0.0162  2311938349319194381898303231634487

error · 2014-09-13, 11:59

Quote:

Originally Posted by error

Originally Posted by error
Manual results form does not recognize ECM factor found.

Originally Posted by James Heinrich
Fixed.

Thanks, but ...
still complains about the checksum and the factor is not accepted.

ECM factors are not accepted as manual submission. Asks for a checksum which is not there.
"Factor found" is not accompanied by a checksum (presumably because factors can not be faked).

James Heinrich · 2014-09-13, 12:29

Quote:

Originally Posted by error

ECM factors are not accepted as manual submission. Asks for a checksum which is not there.
"Factor found" is not accompanied by a checksum (presumably because factors can not be faked).

Sorry about that, please try again and let me know if it's still not working.

error · 2014-09-13, 16:40

Quote:

Originally Posted by James Heinrich

Sorry about that, please try again and let me know if it's still not working.

Works now. Many thanks!

preben s · 2014-09-13, 17:05

Quote:

Originally Posted by Madpoo

Yeah, you've got a LOT of assigned work. Good luck with it all.

Seriously though, you have 9106 exponents assigned to you and all but the first 6 of those (9100) were assigned at the time you mentioned.

If I had to guess, I'd say there's a difference between the PHP timeout and maybe what IIS has. I wonder if IIS is timing out the server response because it takes too long to generate that output for you, but PHP is actually still running and assigns you the work anyway even though IIS has closed the connection by that point.

Are you sure you didn't ask for 8000 exponents at first? I see that 8000 of those (exactly) were all generated in a row, non-stop. Or maybe you asked for 2000 of them but it sent that request 4 times in a row or something.

Anyway, then there's a roughly 3 minute pause and it gets 1000 more (exactly). Then there's a break of nearly an hour and it gets another 1000.

Does that sound about right with the timing?

It was 15 seconds to assign that first group of 1000 exponents, 2 seconds to assign the last 1000 exponents.

That first batch of 8000 took almost 2 minutes. So I am wondering about the IIS/PHP timeouts for something that long running. I thought I saw PHP was set to a 5 minute timeout but I'm not sure.

Request smaller chunks is my advice.

8000 has not been entered any time, so it must be 2000 repeated 4 times.
Big brother here is installing new transparent proxies for more in-depth scrutinizing of the Internet. Maybe some hickups from them.

2014-09-12, 16:56	#708
preben s "Preben Soeberg" Nov 2013 Thailand 1000₂ Posts	Weird assignment About 13:30 UTC today (roughly) I tried to get an assignment of 2000 numbers for TF. The transfer stalled before anything was displayed. Next I tried to get 1000 assignments. Success. And again another 1000 assignments. Succes again. Checked the number of lines in worktodo.txt. There were about 1100 lines (it was empty from the beginning). Now I checked the number of assignments on the server. It showed 9100 TF assignments. (there were 0 TF entries, when I started) I have no idea , how this happened. Last fiddled with by preben s on 2014-09-12 at 16:57

2014-09-13, 06:04	#710
Madpoo Serpentine Vermin Jar Jul 2014 6361₈ Posts	Maintenance downtime - scheduled There will be a brief outage of the www.mersenne.org site for some maintenance, commencing at approximately 11:10 PM Pacific time (6:10 AM UTC). During this time, web facing features and client communications (the API) will be unavailable for about 5-10 minutes. It's just normal monthly maintenance. HP and Microsoft updates/fixes, nothing too spectacular. I'll edit this message with an update when it's back online. EDIT #1: Still going on... It got started 5 minutes late, so we're still about 5 minutes away from being back online. EDIT #2: Back online, enjoy. Last fiddled with by Madpoo on 2014-09-13 at 06:30 Reason: Updated timeline

2014-09-13, 08:32	#711
tha Dec 2002 815₁₀ Posts	The most recent cleared page now displays lines twice if a composite factor has been found. (or three times if it consists of three factors) It did not use to do that. Code: alpertron Manual testing 917789 F-PM1 Sep 12 2014 10:50PM 0.0 0.0162 2311938349319194381898303231634487 alpertron Manual testing 917789 F-PM1 Sep 12 2014 10:50PM 0.0 0.0162 2311938349319194381898303231634487 Last fiddled with by tha on 2014-09-13 at 08:32 Reason: typo

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Official "Faits erronés dans de belles-lettres" thread	ewmayer	Lounge	39	2015-05-19 01:08
Official "all-Greek-to-me Fiction Literature and Cinema" Thread	ewmayer	Science & Technology	41	2014-04-16 11:54
Official "Lasciate ogne speranza" whinge-thread	cheesehead	Soap Box	56	2013-06-29 01:42
Official "Ernst is a deceiving bully and George is a meanie" thread	cheesehead	Soap Box	61	2013-06-11 04:30
Official "String copy Statement Considered Harmful" thread	Dubslow	Programming	19	2012-05-31 17:49