LLRnet servers for NPLB

[gd_barnes]

Originally Posted by IronBits

I'm looking into it, and asking Bok to take a look to see why the permissions/groups keep changing on that server.
It's odd, and I'm not sure what it could be but I'll keep looking for a reason.
If necessary, I'll move all the Servers back to a Windows Server and be done with it.
I remember moving them to the Linux server because the sockets on the Windows servers puke when hammered hard and the llrnet server will stop responding.

I just got the following PM from Lennart. Perhaps the "prunePeriod" is the source of the problem. I believe the below is what he uses for his servers.

Prunetime
-- maximum alowed time for a user to complete a job (in seconds)
jobMaxTime = 4 * 24 * 3600 -- 4 days
-- various files used by the server
knPairsFile = "knpairs.txt"
jobListFile = "joblist.txt"
resultsFile = "results.txt"
-- if you want the server to regularly prune your knpairs and joblist
-- files, then uncomment the following line and set the period of
-- pruning in second.
prunePeriod = 1 * 12 * 3600 -- 1/2 days
----------------------------------------------
It's here in the last line you remove finnished pair.
If you have that set to same as jobMaxTime you can have proccessed pairs in knpairs file 4 days
But i run prune every 12hr then no processed pairs in knpairs longer then 12 hr.s


/Lennart

I hope this helps.


Gary

[IronBits]

I'm using
prunePeriod = 3600 -- hourly
So I see maybe that's a problem
prunePeriod = 1 * 12 * 3600 -- 1/2 days
Looks like it's missing some params :(
I'll fix em all up with what he is doing

[gd_barnes]

Originally Posted by IronBits

I'm using
prunePeriod = 3600 -- hourly
So I see maybe that's a problem
prunePeriod = 1 * 12 * 3600 -- 1/2 days
Looks like it's missing some params :(
I'll fix em all up with what he is doing

Hum, I would think your 3600 would work just fine but as you said, it looks like it's missing some parms.

I wouldn't suggest doing exactly what he's doing. His JobMaxTime is 4 days. We certainly want to leave ours at 3 days.

Also, I think hourly is good for pruning, i.e. 1 * 1 * 3600. But you do what you think is best there.


Gary

[IronBits]

stopped all the servers
I made all the changes to the llr-serverconfig.txt for 1/2 days
restarted all the servers
reads Gary's message /bangs head
changes all the llr-serverconfig.txt to be 1 * 1 * 3600

next time I restart servers, they will be as you wanted.

How many servers am I running over here you ask?

crus4 http://crus.ironbits.net (offline)
crus6 http://crus.ironbits.net

llrnet8000 http://llrnet.no-ip.info or http://llrnet.ironbits.net
llrnet8500 http://llrnet.no-ip.info or http://llrnet.ironbits.net (offline)
llrnet10000 http://llrnet.no-ip.info or http://llrnet.ironbits.net

nplb400 http://nplb.ironbits.net
nplb5000 http://nplb.ironbits.net

llrnet1 (client)
llrnet2 (client)

[gd_barnes]

Carlos,

I'm running one test on each of the misc. servers tonight to verify their n-range and that David's minimum k/n pair remaining is now working properly.

In running yours, I am in sleep mode. Did you bring it down? If so, would you have any problem if we removed the k/n pairs out of it and fed them into port 300, which will be running low within 1-2 days? We'll then load yours up for the next rally, which I'm shooting for the 3rd or last weekend in July.


Gary

[gd_barnes]

Originally Posted by IronBits

stopped all the servers
I made all the changes to the llr-serverconfig.txt for 1/2 days
restarted all the servers
reads Gary's message /bangs head
changes all the llr-serverconfig.txt to be 1 * 1 * 3600

next time I restart servers, they will be as you wanted.

How many servers am I running over here you ask?

crus4 http://crus.ironbits.net (offline)
crus6 http://crus.ironbits.net

llrnet8000 http://llrnet.no-ip.info or http://llrnet.ironbits.net
llrnet8500 http://llrnet.no-ip.info or http://llrnet.ironbits.net (offline)
llrnet10000 http://llrnet.no-ip.info or http://llrnet.ironbits.net

nplb400 http://nplb.ironbits.net
nplb5000 http://nplb.ironbits.net

llrnet1 (client)
llrnet2 (client)

Very nice! Crus4 can be dismantled as far as I'm concerned but in the very near future, I'm thinking we will add a new server there for base 3 that will get LOTS of activity. As soon as some people come back to NPLB here, I'll put a core or two on Crus6. I have 90% of my resources on port 300 here right now.

Are you still OK with hosting the personal servers for people? They come in very handy!


Gary

[gd_barnes]

David,

There was a prime on port 8500 that did not get listed on your web page nor in the "All primes for port 8500 to date" file link on the page. There should be 3 primes shown for port 8500 but there are only 2.

I'm not mentioning the prime here because Bruce hasn't reported it yet. It is in the June 30th results file.


Gary

[em99010pepe]

Originally Posted by gd_barnes

Carlos,

I'm running one test on each of the misc. servers tonight to verify their n-range and that David's minimum k/n pair remaining is now working properly.

In running yours, I am in sleep mode. Did you bring it down? If so, would you have any problem if we removed the k/n pairs out of it and fed them into port 300, which will be running low within 1-2 days? We'll then load yours up for the next rally, which I'm shooting for the 3rd or last weekend in July.


Gary

No problem. Here are all files from my server:

joblist.txt
knpairs.txt
results.txt

[em99010pepe]

IB,

Tell me one thing. When you process the results the server is on or shutdowns automatically? If it shutdowns the countdown to pruning restarts from scratch when you turn on the server, maybe that's the problem.

Carlos

[IronBits]

I just move results.txt out of there while the server is running.

Yes, I see 8500 had three primes.
No problem running personal servers for you or the base 3.

[gd_barnes]

Originally Posted by IronBits

I just move results.txt out of there while the server is running.

Yes, I see 8500 had three primes.
No problem running personal servers for you or the base 3.

So what caused the problem with the prime not showing up on your page? Has the problem been fixed?

Also, why is port 400 still showing the same lowest k/n pair remaining as before? Do you need to shut down the server(s) and restart it/them? I think you said the problem would be fixed the next time you restart the server(s).


Gary