mersenneforum.org  

Go Back   mersenneforum.org > Prime Search Projects > No Prime Left Behind

Closed Thread
 
Thread Tools
Old 2008-05-24, 09:39   #67
gd_barnes
 
gd_barnes's Avatar
 
May 2007
Kansas; USA

72×11×19 Posts
Default

Yes, the clients are down. All my cores are sleeping even after ending the tasks in task manager and rebooting. I'm running other processes on all of them right now under the lowest priority. I'm leaving port 300 sleeping on all cores and changed it to medium priority.

I'm going to bed shortly so hopefully all my cores will start port 300 again automatically when it comes back up and the other processes, at lower priority, will not run or at least will interfere only minimally.

I got a PM response from Ironbits earlier today that he can set up a new server any time but that he isn't on very often. I'm sending him a PM shortly to set up a new server on port 400 and will send him n=455K-460K to load into it.

Here will be the server info. for the new server:
server = "llrnet.ironbits.net"
port = 400


Either Ironbits or I will post here shortly after it is set up.

Regardless of where port 300 is at when I'm next on in about 8 hours, if Ironbits has his server set up by that time, I'm going to shift all 22 cores over to it. I would recommend that anyone else who is able to do the same until we know that port 300 is stable again. Not everyone needs to do so but just enough to makes things stable again. Let's try to keep it at < 100 cores on new server port 400. I'm thinking that it put it in the 40-60 core range on port 300.

Sorry guys. I tried to avoid this this time with the stress test ahead of time but we have a lot more cores running the server than were running it during the test.

All future rallies will be run on 2 servers for the same drive. If it takes longer to sort out the results files, then that's what we'll have to do. I do not want us to be burned a 3rd time! 2 times is enough.


Gary
gd_barnes is offline  
Old 2008-05-24, 10:00   #68
Lennart
 
Lennart's Avatar
 
"Lennart"
Jun 2007

25·5·7 Posts
Default

Quote:
Originally Posted by gd_barnes View Post

Regardless of where port 300 is at when I'm next on in about 8 hours, if Ironbits has his server set up by that time, I'm going to shift all 22 cores over to it. I would recommend that anyone else who is able to do the same until we know that port 300 is stable again. Not everyone needs to do so but just enough to makes things stable again. Let's try to keep it at < 100 cores on new server port 400. I'm thinking that it put it in the 40-60 core range on port 300.

Sorry guys. I tried to avoid this this time with the stress test ahead of time but we have a lot more cores running the server than were running it during the test.



Gary
I think i stay here or you send me some pairs so i can run on one of my own servers.
I have 68 core on and it's better i run on my own server.
I wait and see if Adam get's it running in a hr.

/Lennart
Lennart is offline  
Old 2008-05-24, 10:03   #69
em99010pepe
 
em99010pepe's Avatar
 
Sep 2004

2·5·283 Posts
Default

Quote:
Originally Posted by Lennart View Post
I think i stay here or you send me some pairs so i can run on one of my own servers.
I have 68 core on and it's better i run on my own server.
I wait and see if Adam get's it running in a hr.

/Lennart
You can go here and reserve some pairs. 455k-456k is available. You also have team drive 3 with more files.

Last fiddled with by em99010pepe on 2008-05-24 at 10:03
em99010pepe is offline  
Old 2008-05-24, 10:05   #70
em99010pepe
 
em99010pepe's Avatar
 
Sep 2004

2×5×283 Posts
Default

The server is up and running.
em99010pepe is offline  
Old 2008-05-24, 10:07   #71
gd_barnes
 
gd_barnes's Avatar
 
May 2007
Kansas; USA

72×11×19 Posts
Default

Quote:
Originally Posted by Lennart View Post
I think i stay here or you send me some pairs so i can run on one of my own servers.
I have 68 core on and it's better i run on my own server.
I wait and see if Adam get's it running in a hr.

/Lennart

It's 6 AM in his time zone. Unless he's a very early bird, I doubt it.

I don't know what Anon and Karsten will think of this and how we'll get the results processed, but I'll send you some k/n pairs shortly.

Let's see, at 3.5 mins per test, that's 17 tests/hour/core or 1166 tests per hour times 30 hours = ~35000 k/n pairs. I'll send you somewhere in the vicinty of 35000 k/n pairs.

We'll sort out how you can report the results later and how we want to determine the results processed vs. the end of the rally.

Thanks for being flexible.


Gary
gd_barnes is offline  
Old 2008-05-24, 11:43   #72
Mini-Geek
Account Deleted
 
Mini-Geek's Avatar
 
"Tim Sorbera"
Aug 2006
San Antonio, TX USA

10AB16 Posts
Default

My cores were since about midnight CDT with the same symptoms as other Windows clients reported, but a restart kicked them into working. (They were idle over 6 hours ) We should definitely run future rallies on two servers, maybe even three if we expect a huge load.

Have enough cores been moved to IB400 and private servers that we shouldn't crash again?

This is why I usually do manual LLR. There's much fewer points of failure (my computer, my power) compared to LLRnet (my computer, my power, my internet, server's internet, server's computer, server's power), 2 vs 6, and the server's computer is especially vulnerable during rallies.
Mini-Geek is offline  
Old 2008-05-24, 11:45   #73
Lennart
 
Lennart's Avatar
 
"Lennart"
Jun 2007

25·5·7 Posts
Default

Quote:
Originally Posted by gd_barnes View Post
It's 6 AM in his time zone. Unless he's a very early bird, I doubt it.

I don't know what Anon and Karsten will think of this and how we'll get the results processed, but I'll send you some k/n pairs shortly.

Let's see, at 3.5 mins per test, that's 17 tests/hour/core or 1166 tests per hour times 30 hours = ~35000 k/n pairs. I'll send you somewhere in the vicinty of 35000 k/n pairs.


Gary
I have them and my server is open if you need.
"samband.mine.nu"
Port 6

I run 28 core on AES but the rest on my server, And if we get more stop i can switch all to my own server.-

/Lennart
Lennart is offline  
Old 2008-05-24, 15:40   #74
Brucifer
 
Brucifer's Avatar
 
Dec 2005

313 Posts
Default

Well it's8:40am PST and I can't send anything, and will have to go through the reboot again. Unfortunately I've also got to leave and can't get to it for several hours. I think I'm done with the rally as this is turning into too much idle cpu time. :( I think the gent that mentioned the manual thing has the correct answer as the llrnet server seems to be connection limited.
Brucifer is offline  
Old 2008-05-24, 15:50   #75
em99010pepe
 
em99010pepe's Avatar
 
Sep 2004

2·5·283 Posts
Default

Quote:
Originally Posted by Brucifer View Post
Well it's8:40am PST and I can't send anything, and will have to go through the reboot again. Unfortunately I've also got to leave and can't get to it for several hours. I think I'm done with the rally as this is turning into too much idle cpu time. :( I think the gent that mentioned the manual thing has the correct answer as the llrnet server seems to be connection limited.
Well, I moved all my cores to:

server = "llrnet.ironbits.net"
port = 5000

Carlos
em99010pepe is offline  
Old 2008-05-24, 16:13   #76
mdettweiler
A Sunny Moo
 
mdettweiler's Avatar
 
Aug 2007
USA (GMT-5)

792 Posts
Default

WHOA. Why do all the bad things happen when I'm asleep?

Thanks Gary and IronBits for getting the backup server set up! Also, thanks Lennart for setting up an additional server in case we need it.

One quick thing about running IronBits' server, though: all the k/n pairs crunched on his server will NOT be counted in the live stats on http://nplb.rieselprime.org. So if stats are a big priority for you, I'd recommend that you stay on Adam's server. If you don't care about the stats that much, you can go on IronBits' server; we might be able to cobble together some way of getting it wired into Adam's stats, but don't count on it.

Anon
mdettweiler is offline  
Old 2008-05-24, 16:15   #77
mdettweiler
A Sunny Moo
 
mdettweiler's Avatar
 
Aug 2007
USA (GMT-5)

792 Posts
Default

Quote:
Originally Posted by Lennart View Post
I have them and my server is open if you need.
"samband.mine.nu"
Port 6

I run 28 core on AES but the rest on my server, And if we get more stop i can switch all to my own server.-

/Lennart
@Gary: should we consider this an "official" server and thus change the name on Lennart's reservation to "LLRnet (L6)"?
mdettweiler is offline  
Closed Thread

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Rally Jan. 23rd-25th gd_barnes No Prime Left Behind 89 2009-01-25 22:59
LLRnet server rally 400<k<1001 August 8-10 mdettweiler No Prime Left Behind 66 2008-08-11 03:00
LLRnet server rally 400<k<1001 June 20-22 mdettweiler No Prime Left Behind 67 2008-06-23 15:32
LLRnet server rally port 300 May 3rd-4th gd_barnes No Prime Left Behind 45 2008-05-05 19:56
LLRnet server rally March 8th-9th gd_barnes No Prime Left Behind 135 2008-03-14 19:52

All times are UTC. The time now is 05:04.

Sat Nov 28 05:04:54 UTC 2020 up 79 days, 2:15, 3 users, load averages: 1.09, 1.20, 1.17

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.