24 hour rally

[em99010pepe]

NPLB100 current status :
nil
Proposing pair 571/283047 to Free-DC_Beyond
connection closed (socket 444)
Proposing pair 703/283047 to Free-DC_Beyond
connection request from ee651e4c:2984 (socket 436)
Proposing pair 709/283047 to Free-DC_Beyond
Proposing pair 751/283047 to Free-DC_Beyond
Proposing pair 795/283047 to Free-DC_Beyond
Proposing pair 805/283047 to Free-DC_Beyond
connection closed (socket 524)
connection request from 19c2c462:3721 (socket 520)
Proposing pair 831/283047 to Free-DC_Beyond
Proposing pair 933/283047 to Free-DC_Beyond
Proposing pair 961/283047 to Free-DC_Beyond
Proposing pair 963/283047 to Free-DC_Beyond
Proposing pair 617/283048 to Free-DC_Beyond
Proposing pair 623/283048 to Free-DC_Beyond
Proposing pair 657/283048 to Free-DC_Beyond
connection closed (socket 436)
connection request from 5498c850:2508 (socket 432)
StartServiceCtrlDispatcher returns 0

Look at last line. I get that then the server goes down!

[mdettweiler]

Originally Posted by em99010pepe

NPLB100 current status :
nil
Proposing pair 571/283047 to Free-DC_Beyond
connection closed (socket 444)
Proposing pair 703/283047 to Free-DC_Beyond
connection request from ee651e4c:2984 (socket 436)
Proposing pair 709/283047 to Free-DC_Beyond
Proposing pair 751/283047 to Free-DC_Beyond
Proposing pair 795/283047 to Free-DC_Beyond
Proposing pair 805/283047 to Free-DC_Beyond
connection closed (socket 524)
connection request from 19c2c462:3721 (socket 520)
Proposing pair 831/283047 to Free-DC_Beyond
Proposing pair 933/283047 to Free-DC_Beyond
Proposing pair 961/283047 to Free-DC_Beyond
Proposing pair 963/283047 to Free-DC_Beyond
Proposing pair 617/283048 to Free-DC_Beyond
Proposing pair 623/283048 to Free-DC_Beyond
Proposing pair 657/283048 to Free-DC_Beyond
connection closed (socket 436)
connection request from 5498c850:2508 (socket 432)
StartServiceCtrlDispatcher returns 0

Look at last line. I get that then the server goes down!

Are you by any chance running the LLRnet server as a Windows service? From that last line, that's what I'd be inclined to guess.

If so, how about you try just running it normal (i.e. double-clicking llrserver.exe and letting it open a DOS window)?

[em99010pepe]

Originally Posted by Anonymous

Are you by any chance running the LLRnet server as a Windows service? From that last line, that's what I'd be inclined to guess.

If so, how about you try just running it normal (i.e. double-clicking llrserver.exe and letting it open a DOS window)?

I'm not running it as a service, weird...

[gd_barnes]

Port 300 just went down. On one machine, 5 mins. ago, I started it up and got 2 k/n pairs for a test, my machine tested them, but it was unable to sent the results.

On the other machine, it now says that port 300 is sleeping before I was able to get anything to test.

Edit: Checking my son's machine now which had been running port 300 for almost 30 mins. prior to this...

[mdettweiler]

Originally Posted by em99010pepe

I'm not running it as a service, weird...

Hmm.

You mentioned a memory leak in the port 100 LLRnet server--do you think that could be a result of running too many LLRnet servers on your system (since you said before that they were all using tons of RAM)?

You could always try moving the port 100 server to your quadcore, just while the rally's going...that way you can be sure that it will work fine. I know you don't want to run anything but crunching on your quadcore, but maybe you could make an exception just this once?

[mdettweiler]

Originally Posted by gd_barnes

Port 300 just went down. On one machine, 5 mins. ago, I started it up and got 2 k/n pairs for a test, my machine tested them, but it was unable to sent the results.

On the other machine, it now says that port 300 is sleeping before I was able to get anything to test.

Oooh. Not good.

Let's hope we can get it up and running stably, or else we might have to call off the rally until next week.

[mdettweiler]

According to a quick little check with netcat, both 300 and 100 seem to be up --for now.

[tnerual]

what you can do is shutting down the riesel crus server to gain some memory ... i only have one computer on it, but i don't have any access to it before wednesday ... not a big deal ...

maybe we can make a fund raising to get you a 1GB memory stick ...

[mdettweiler]

Originally Posted by tnerual

what you can do is shutting down the riesel crus server to gain some memory ... i only have one computer on it, but i don't have any access to it before wednesday ... not a big deal ...

maybe we can make a fund raising to get you a 1GB memory stick ...

Or, maybe someone else could volunteer to run one of the LLRnet servers. I would gladly run a few servers, except that I don't have any machines that are consistently on 24/7.

[gd_barnes]

Originally Posted by tnerual

what you can do is shutting down the riesel crus server to gain some memory ... i only have one computer on it, but i don't have any access to it before wednesday ... not a big deal ...

maybe we can make a fund raising to get you a 1GB memory stick ...

I have one machine on Riesel also. Do you have any on Sierp? If not, let's shut Sierp down.

OK, here's the deal: I just checked my son's machine. It ran 3 tests, apparently stopped for about 15 minutes, and is now running again.

So it appears that port 300 went down for 15 minutes and came back up. My other 2 machines now just successfully tested 1 k/n pair per core. I'm satisfied that my machines are set up correctly.

This was the first observed outage by me in 2 weeks on port 300 so I'm comfortable with it. We'll hope for the best come rally-time. I'll be running 2 cores of my son's machine on it until rally-time and I'll be able to tell if there was a large gap in between any tests.

Edit: Before the rally starts, I'm going to queue up 100 k/n pairs on every machine. I would suggest everyone else do the same. If we lose the server for 1-2 hours, that's no reason to call off the rally. Everyone's machine should be able to process their queue until it comes back up.

Opinions, thoughts, objections?


Gary

[mdettweiler]

Originally Posted by gd_barnes

I have one machine on Riesel also. Do you have any on Sierp? If not, let's shut Sierp down.

OK, here's the deal: I just checked my son's machine. It ran 3 tests, apparently stopped for about 15 minutes, and is now running again.

So it appears that port 300 went down for 15 minutes and came back up. My other 2 machines now just successfully tested 1 k/n pair per core. I'm satisfied that my machines are set up correctly.

This was the first observed outage by me in 2 weeks on port 300 so I'm comfortable with it. We'll hope for the best come rally-time. I'll be running 2 cores of my son's machine on it until rally-time and I'll be able to tell if there was a large gap in between any tests.

Edit: Before the rally starts, I'm going to queue up 100 k/n pairs on every machine. I would suggest everyone else do the same. If we lose the server for 1-2 hours, that's no reason to call off the rally. Everyone's machine should be able to process their queue until it comes back up.

Opinions, thoughts, objections?


Gary

Okay, that sounds like a good idea. That way we can still do the rally on port 100!

BTW, if you want to queue more than 100 k/n pairs at a time, all you have to do is make a small modification to the llrnet.lua file. I've attached a replacement llrnet.lua (just remove .pdf from the end of the name and replace the one in your LLRnet folder with it) that will let you queue up to 1000 k/n pairs at a time.