mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   No Prime Left Behind (https://www.mersenneforum.org/forumdisplay.php?f=82)
-   -   LLRnet server rally port 300 May 23rd-25th (https://www.mersenneforum.org/showthread.php?t=10323)

gd_barnes 2008-05-21 05:44

LLRnet server rally port 300 May 23rd-25th
 
It's time for another LLRnet rally on port 300! :smile:

This time around, we're going to run one for 48 hours this weekend from Friday May 23rd at 7 PM GMT to Sunday May 25th at 7 PM GMT. Adam and Anon, this would be 2 PM CDT and 3 PM EDT. Do you see a problem starting the rally mid-afternoon U.S. on a weekday?

Here is the server info.:
server = "nplb.rieselprime.org"
port = 300

General info. on setting up a connection to an LLRnet server can be found [URL="http://www.mersenneforum.org/showthread.php?t=9959"][COLOR=#22229c]here[/COLOR][/URL]. The above info. should be put in the llr-clientconfig.txt file.

For this one, can everyone report how many cores they will be bringing? I'd also like to suggest that we run a test the day before the rally. Can everyone run their rally cores for ~1 hour, perhaps from 10-11 PM GMT on Thurs May 22nd. Hopefully everyone would be home from work in the U.S. and people won't yet be in bed in Europe at that time.

I'm going to bring ~18 cores this time around and we'll probably want to run a test to see if Adam's server can handle the load. I know we have a new searcher or two since the last rally so this is a precaution more than anything since we got burned on our 2nd rally.

Everyone bring all of your firepower to the rally as we see how fast we can bring drive 1 up to n=450K! :smile:


Gary

mdettweiler 2008-05-21 05:58

[quote=gd_barnes;133815]It's time for another LLRnet rally on port 300! :smile:

What does everyone think of running one for 48 hours this weekend from Friday May 23rd at 7 PM GMT to Sunday May 25th at 7 PM GMT?

Adam and Anon, this would be 2 PM CDT and 3 PM EDT. Do you see a problem starting the rally mid-afternoon U.S. on a weekday?

If we do one, can everyone report how many cores they will be bringing?I'd also like to suggest that we run a test the day before the rally. Can everyone run their rally cores for ~1 hour, perhaps from 10-11 PM GMT on Thurs May 22nd. Hopefully everyone would be home from work in the U.S. and people won't yet be in bed in Europe at that time.

I'm going to bring ~18 cores this time around and we'll probably want to run a test to see if Adam's server can handle the load. I know we have a new searcher or two since the last rally so this is a precaution more than anything since we got burned on our 2nd rally.


Gary[/quote]
Sounds fine as far as I'm concerned. I know I definitely won't be out of town those days, and I should be home most of the time.

As for any difficulties starting the rally in the mid-afternoon on a weekday: I don't think it will be a big deal. If anyone won't be able to access their machines at the time of the beginning of the rally, I recommend that they get them going on port 300 a little bit ahead of time--not only will this ensure that the machines don't miss any of the rally, it will also help out with our pre-rally server stress test a bit. :wink:

I plan on bringing both cores of my Core 2 Duo (2.2Ghz) to the rally, as usual, and I might have a 1Ghz P3 helping out for part of the rally (though it crunches so slowly that its load on the server is minimal and thus I won't bother with it for the pre-rally server load test). :smile:

Anon :smile:

Mini-Geek 2008-05-21 12:10

The time is fine for me. I'll be at a church event on the 25th so I don't know when I'll be able to pull my 2 cores off, but no harm in running a little longer.
Hopefully I'll put in 48 dual core CPU hours this time...last time my computer rebooted a few times (it occasionally reboots when certain USB device just touch the front USB port, anyone know why? the power and reset buttons are also on the same front panel, and the reset button doesn't work properly, if any of that matters), and I hadn't changed over my Startup folder to start LLRnet, so it didn't run the whole time. I'll change the Startup folder this time, so it'll keep crunching the rally even if it gets rebooted for whatever reason.

mdettweiler 2008-05-21 15:16

[quote=Mini-Geek;133841]The time is fine for me. I'll be at a church event on the 25th so I don't know when I'll be able to pull my 2 cores off, but no harm in running a little longer.
Hopefully I'll put in 48 dual core CPU hours this time...last time my computer rebooted a few times (it occasionally reboots when certain USB device just touch the front USB port, anyone know why? the power and reset buttons are also on the same front panel, and the reset button doesn't work properly, if any of that matters), and I hadn't changed over my Startup folder to start LLRnet, so it didn't run the whole time. I'll change the Startup folder this time, so it'll keep crunching the rally even if it gets rebooted for whatever reason.[/quote]
Hmm...that's odd. I've seen one particular computer that would freeze up whenever I plugged in my USB stick (I didn't try it with any other USB sticks), but I've never seen one reboot like that. The particular computer I've seen freeze up from my USB stick was running Ubuntu (no, not my primary machine)--maybe that explains the different "symptoms" of a similar problem?

Mini-Geek 2008-05-21 15:57

[quote=Anonymous;133857]Hmm...that's odd. I've seen one particular computer that would freeze up whenever I plugged in my USB stick (I didn't try it with any other USB sticks), but I've never seen one reboot like that. The particular computer I've seen freeze up from my USB stick was running Ubuntu (no, not my primary machine)--maybe that explains the different "symptoms" of a similar problem?[/quote]
Hm, that sounds like a different problem...like something with the software/OS, not the hardware. I built this machine, so it's possible that I wired something from the front panel wrong and that's the problem, but I don't see what I could've done wrong that would make touching the metal (not fully plugging in, just touching) act like I hit the reset button (and, no, it's not accidentally pressing the power or reset buttons).

mdettweiler 2008-05-21 16:01

[quote=Mini-Geek;133866]Hm, that sounds like a different problem...like something with the software/OS, not the hardware. I built this machine, so it's possible that I wired something from the front panel wrong and that's the problem, but I don't see what I could've done wrong that would make touching the metal (not fully plugging in, just touching) act like I hit the reset button (and, no, it's not accidentally pressing the power or reset buttons).[/quote]
Aha, that might be it--I can imagine how it might think you hit the reset button when you plug in a USB stick, if somehow the USB connector on the motherboard is shorting the two pins of the reset button together. That would also explain why the reset button doesn't work. :smile:

Maybe you could try taking the front panel USB connector and switching it to a different internal-USB connector on the motherboard? (one that's farther away from the reset button connector)

em99010pepe 2008-05-21 17:38

Already added 5 cores. They will be running for at least 4 days in a row.

glennpat 2008-05-21 23:50

I'll bring 16 cores and I can be there for the test run.

gd_barnes 2008-05-22 00:28

[quote=glennpat;133928]I'll bring 16 cores and I can be there for the test run.[/quote]

OK, great Glenn. Bruce, do you know how many cores you'll be bringing to the rally? So far we have 43 cores committed that I've counted.

In case anyone missed it, I'd like for us to run a stress test on Adam's server from 10-11 PM GMT on Thurs. night. (5-6 PM CDT and 6-7 PM EDT in the U.S.) Please focus your rally machines on port 300 at that time.

If we have problems with the stress test, I think we could bring in Ironbits to add a temporary server and a couple of the bigger crunchers only could run on that one for the rally. That or whatever it takes to make things runs smoothly.

I'm crossing my fingers that we can run upwards of 100 cores with no problems for 48 hours! :smile:


Gary

Brucifer 2008-05-22 04:01

I'll commit 30.

gd_barnes 2008-05-22 09:16

OK, we're up to 73 committed cores for the rally now. If you haven't already done so, please put them on port 300 for a stress test later tonight from 10PM-11PM GMT.

I currently have 4 on it and will add the other 14 later today.


All times are UTC. The time now is 07:35.

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