Chess Brain: Experimental internet accessible peernode monitor

[Confused]

*apologises for typo* 😉

<bump>


Garry

 

[sciencewhiz]

well, you would have only seen the typo if a client hadn't checked in in days ;-)

I only found it when I ran it with perl -w

 

[Confused]

*apologises for typo anyway* 😉😛

 

[sciencewhiz]

Version 0.4 is done. This allows you to define a second peernode monitor to send data to. This allows me to have my own internal stats (linked below), but also show up on Confused stats.

It also shows the aborted percent for each host.

script: http://sciencewhiz.is-a-geek.com/peernode_file.pl

output: http://sciencewhiz.is-a-geek.com/peernodestats.html

 

[Confused]

Version 0.41 is now available...fixes a stupid bug I didn't think about that made it repeat some values on the last checkin for a client (ie showing "02 h 06 m 14 s06 m 14 s" instead of "02 h 06 m 14 s", which would only have got worse the longer a client didn't connect 😱)

script: http://teamanandtech.d2g.com/cb/peernode_file.pl

output: http://teamanandtech.d2g.com/cb/


Garry

 

[sciencewhiz]

Version 0.5 is done, but hardly been tested because of the lack of work. This adds a users section.

script: http://sciencewhiz.is-a-geek.com/peernode_file5.pl
output: http://sciencewhiz.is-a-geek.com/peernodemonitor.html

edit: coolkid is running it: http://coolkidsserver.no-ip.com:12380/

hopefully there will be some more work soon

edit2: ignore the rate/host column. the rate column is really rate/host.

 

[Coolkid]

Thanks for setting up the latest version on my server, sciencewhiz 🙂

Anyone that wants to use my monitor, change the netmon= line in your cbspn.conf to netmon=coolkidsserver.no-ip.com and restart the client 🙂

 

[Confused]

I seem to be doing all the .01 updates so far! 😉

Version 0.51 fixes the "rate" issue for user stats, and also adds "To use this monitor..." blurb to the "time Program Started Waiting for Peernodes to connect..." in the web page when program started and waiting for client to connect 🙂

script: http://teamanandtech.d2g.com/cb/peernode_file_051.pl
output: http://teamanandtech.d2g.com/cb/

🙂


Garry

 

[sciencewhiz]

Originally posted by: Confused
I seem to be doing all the .01 updates so far! 😉

I did the first two, now you've done the second two 😛

edit: Confused, after the clients have started doing work again, can you give me an idea of how much cpu time this uses?

 

[Confused]

Originally posted by: sciencewhiz

Originally posted by: Confused
I seem to be doing all the .01 updates so far! 😉

I did the first two, now you've done the second two 😛

edit: Confused, after the clients have started doing work again, can you give me an idea of how much cpu time this uses?

Oh yeah, so it has been! 🙂

Sure can let you know how much CPU time it uses 🙂


Garry

 

[sciencewhiz]

Aparently, if the second monitor isn't reachable, the peernode monitor stops working. I'll have to look at that later.

I may not have time to do any work on it until sometime next week, though.

 

[Confused]

It'll have been running for 2 days as of 4hrs time, so i'll take the CPU time it's used and divide by 2 to get a daily CPU usage (on my XP1800+ server)


Garry

 

[sciencewhiz]

2 hours seems like a lot. I guess I definetly will try to add a feature that only outputs the webpage at certain (user defined) intervals. However, I'm going to be gone for the weekend.

 

[Confused]

Well I would have had a time for you...if i hadn't knocked the box when I sat down and it reset itself 🙁

It should *crosses fingers* have had a day's worth of actual time tonight, so can give you a value for CPU time if you want 🙂


I think you mis-understood my last message, but I shall tell you exactly how much CPU time it has used tonight 🙂


Garry

 

[Confused]

OK, CPU usage results are in 🙂

It's been running 1 day 1 hour 14 mins, and has used a whopping:
.
..
...
....
.....
......
.....
....
...
..
.
<wait for it>
.
..
...
....
.....
......
.....
....
...
..
.
<keep going>
.
..
...
....
.....
......
.....
....
...
..
.
<don't you hate emails that go on like this>
.
..
...
....
.....
......
.....
....
...
..
.
<just a little more>
.
..
...
....
.....
......
.....
....
...
..
.
23 seconds


🙂 Which is pretty much naff all 🙂


Garry

 

[sciencewhiz]

oh, hehe. You're right, I did misunderstand

 

[Confused]

Originally posted by: sciencewhiz
oh, hehe. You're right, I did misunderstand

Oh well! 🙂


Only thing I need to do is figure out why my fvcking server decides to restart every 2-3 days :| I think it's something to do with 2003 Server, as there's always some service that's failed to start when it turns on!

Looks like i'll be backing up and reinstalling when I get home from work

Then once i've got some stability, i'll think about doing some more updates to the monitor 🙂


Garry

 

[sciencewhiz]

I tried getting perl's sort function to work, but I don't think I understand it completely, especially since everything is stored in hash of arrays. I'm about to give up on that.

The next thing would be to get it to save the data so it can be restored when the peernode monitor is restarted. Coolkid suggested a mysql database, which would be nice, but I don't know mysql. Does anyone think it would be worth it enough for me to learn mysql (I don't mind, but it would take a while and delay other features) or just dump everything to a plaintext file?