Maybe you can get that mysql stuff working with the llrnet server to while you're at it :)
I was never able to get it working :cry:

[quote=IronBits;147608]Maybe you can get that mysql stuff working with the llrnet server to while you're at it :)
I was never able to get it working :cry:[/quote]
I'm afraid the MySQL stuff is even farther over my head than yours. The only experience I've ever had with MySQL is when I once needed to install it to run a phpBB-based forum off an old Linux box I had--and even that I pretty much just followed instructions I got off the Internet to create the phpBB database, and let phpBB handle the rest. :smile:

Meanwhile, I've got a status page set up for the G4000 server:

[url]http://nplb-gb1.no-ip.org/llrnet/[/url]

It reports the lowest and highest k/n pairs in knpairs.txt, and the # of k/n pairs remaining. Results are copied off daily, and also a file containing all the results since the last daily copy-off is updated every time the main status page is updated. (I haven't gotten the "Today's Results Processing Log" link to work yet, though--mainly because I have yet to edit the Perl script to actually generate much of any meaningful console output. :smile:)

Now the next step is to get a script to handle the CSV conversion. :smile:

Looking good so far! :thumbs:

Max,

The lowest k/n pair outstanding is now n>540K on this drive. Therefore, please process port 400 results to me for the following:

n=532.8K-537K; all results should be there; please verify all of them vs. the original sieved file.

n=537K-540K; Adam's port 300 had most of the pairs from n=537001-538498 already processed. Please only verify the range of n=538499-540000 vs. the original sieved file.

I've saved off Adam's results in the n=537001-538498 range that were processed and will merge them in with what you send me here, sort everything by n-value/k-value and then verify that the entire range of n=532801-540000 contains the correct # of results vs. the sieved file.

Note: Since Adam's range above would already be in the stats totals on his server before he shut it down, we would not need to create CSV results for them for loading to David's server. Taking that a step further, there should be NO results from Adam's server that need CSV results created for them. It's only what is in other servers that need that done plus what is in David's server (if that is not being done automatically by his process).

It'll be nice to get this messy range out of the way. That's why I wanted to get port 400 past n=540K before moving to port 5000.


Gary

[quote=gd_barnes;147646]Max,

The lowest k/n pair outstanding is now n>540K on this drive. Therefore, please process port 400 results to me for the following:

n=532.8K-537K; all results should be there; please verify all of them vs. the original sieved file.

n=537K-540K; Adam's port 300 had most of the pairs from n=537001-538498 already processed. Please only verify the range of n=538499-540000 vs. the original sieved file.

I've saved off Adam's results in the n=537001-538498 range that were processed and will merge them in with what you send me here, sort everything by n-value/k-value and then verify that the entire range of n=532801-540000 contains the correct # of results vs. the sieved file.

Note: Since Adam's range above would already be in the stats totals on his server before he shut it down, we would not need to create CSV results for them for loading to David's server. Taking that a step further, there should be NO results from Adam's server that need CSV results created for them. It's only what is in other servers that need that done plus what is in David's server (if that is not being done automatically by his process).

It'll be nice to get this messy range out of the way. That's why I wanted to get port 400 past n=540K before moving to port 5000.


Gary[/quote]
Okay, I'll do that tomorrow. (Remind me if I it's still not done on Tuesday. :smile:)

542.2-.4 is complete, 1 prime.
Estimate for 542.4-.6 (finishing this range) is 11/13.

532.8K-537K complete by LLRnet IB400, results emailed to Gary. :smile:

Gary, I'm going to do 537K-540K next. As you said, I'll send you the lresults for everything in that range from IB400, though I'll only verify 538499-540000 with the original sieve file since the stuff before that will have gaps that will be filled in with AES300's results on your end. :smile: Note: I won't make a post here saying that that range is done, though, since the range isn't truly complete until you've put together and verified all the results for the entire range. I'll leave it up to you to officially note it as complete.

Oh, one thing I just thought of about your last message: you mentioned comparing the # of results in the results file to the # of k/n pairs in the original sieve file. Just curious, are you also verifying that the k/n data from the sieve file and the results match, line by line, or are you just comparing the # of lines?

[quote=mdettweiler;147716]532.8K-537K complete by LLRnet IB400, results emailed to Gary. :smile:

Gary, I'm going to do 537K-540K next. As you said, I'll send you the lresults for everything in that range from IB400, though I'll only verify 538499-540000 with the original sieve file since the stuff before that will have gaps that will be filled in with AES300's results on your end. :smile: Note: I won't make a post here saying that that range is done, though, since the range isn't truly complete until you've put together and verified all the results for the entire range. I'll leave it up to you to officially note it as complete.

Oh, one thing I just thought of about your last message: you mentioned comparing the # of results in the results file to the # of k/n pairs in the original sieve file. Just curious, are you also verifying that the k/n data from the sieve file and the results match, line by line, or are you just comparing the # of lines?[/quote]


By saying that, I was making the assumption that you do the match on them before sending to me.

But regardless of whether you send them to me or I process them, I actually verify them line by line. I have an Excel spreadsheet with formulas that parse the k and n values out of both the sieve file and the results file. I then sort the results by n/k, hit calc, and it tells me if there are any lines that don't match. So far, whenever the line count has been the same, there have been no mismatches but you never know.


Gary

[quote=gd_barnes;147764]By saying that, I was making the assumption that you do the match on them before sending to me.

But regardless of whether you send them to me or I process them, I actually verify them line by line. I have an Excel spreadsheet with formulas that parse the k and n values out of both the sieve file and the results file. I then sort the results by n/k, hit calc, and it tells me if there are any lines that don't match. So far, whenever the line count has been the same, there have been no mismatches but you never know.


Gary[/quote]
Okay, thanks--just wanted to confirm that. :smile:

[QUOTE=gd_barnes;147764]But regardless of whether you send them to me or I process them, I actually verify them line by line. I have an Excel spreadsheet with formulas that parse the k and n values out of both the sieve file and the results file. I then sort the results by n/k, hit calc, and it tells me if there are any lines that don't match. So far, whenever the line count has been the same, there have been no mismatches but you never know.
[/QUOTE]

that's what i do with my scripts:
- collect all pairs from the LLRnet result file of a special range
- sort this by n and k
- compare this with the original sieve
- write results, primes, LLRnet-entries in a separate file

all is done in a few seconds!

I have now merged the former port 300 results with the more recent port 400 results for the range of n=537K-540K. All are there so the range is now considered complete.

This means that the full range of n=520K-540K is complete. I have verified the primes in post 1 of this thread, the 1st drive page, the k=300-2000 page, and the top 5000 site. No problems were found. Nice work everyone! :smile:

If the Primesearch site will ever allow entry of primes or reservation/completion of ranges again, I'll report ranges and primes on this range as well as a couple of others there. It's been over a month since it allowed any kind of update or reservation.


Gary