Server Move and now ET causing problems

[SNN]

Okay well I moved from VPS to a Dedicated from theplanet (reselled by hostgator) anyways, Now that I'm on the server, I notice etded.x86 coming up to (ps acx time) 233:42 for one server and 2:32 for another. What could be causing this? I'm sure its something software related, but what?

edit: here's loads as of current 10:17 PM Central Time (TX):

# load
Load Averages
2.29 1.41 0.89 2/120 17589

[darren]

The TIME from ps output isn't a good indication of causing more load on one box than another. On my workstation, X is using 561:40:22 in the "TIME" column, but cpu usage is hanging at 0%-3% at all times. It's all relative to what else is on the machine using resources. The %CPU column would be a better indication of the server hitting the server harder than on another machine.

[SNN]

Well, either way, they're still the top two.

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
 2372 dopeman   15   0  240m 102m 2952 S 15.9 10.2  39:28.85 etded.x86
 2381 keith     15   0  157m  99m 3096 S 10.6  9.9  45:51.69 etded.x86

[darren]

That looks normal to me.. almost all these are empty servers:

ID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND                                                                                                                                                                                               
 3274 cpgs      15   0  703m 144m 4172 S   23  4.8  16172:04 server-bin                                                                                                                                                                                             
24625 cpgs      15   0 71420  51m 6336 S   13  1.7   2283:20 bf1942_lnxded                                                                                                                                                                                          
 3275 cpgs      15   0  364m 301m  12m R    6 10.0 431:38.13 ut3-bin                                                                                                                                                                                                
26038 cpgs      15   0  272m 131m 9820 S    5  4.4   5046:48 etqwded.x86                                                                                                                                                                                            
 3287 cpgs      15   0  103m  81m 7780 S    4  2.7   3749:57 bfv_linded                                                                                                                                                                                             
 3368 cpgs      15   0  162m  97m 2628 R    2  3.2   1943:16 hlds_amd

I'm not sure why armyops uses so many resources while idling, but I suspect it's due to the polling method it uses, trying to read from the network port (1716 by default) (too?) many times in a single second.
I'd have to look at the source to be sure of how it was working however.

[SNN]

That's quite strange cause we're getting lag spikes, and our VPS did not have the time from ps even on them... Could there be any relation from the CPU usage and lag?

[SNN]

Okay well after modifying things today we got things fixed besides this ..

top - 00:03:44 up  4:00,  1 user,  load average: 1.57, 2.05, 2.01
Tasks: 105 total,   4 running, 100 sleeping,   0 stopped,   1 zombie
Cpu(s): 98.3% us,  1.0% sy,  0.0% ni,  0.0% id,  0.0% wa,  0.0% hi,  0.7% si
Mem:   1034796k total,   930032k used,   104764k free,    18956k buffers
Swap:  2096472k total,      660k used,  2095812k free,   500308k cached

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
12519 dopeman   25   0  245m 114m 5536 R 65.5 11.4   9:08.33 etded.x86
13649 dopeman   15   0  151m  85m 3132 S  2.0  8.4   0:15.28 etded.x86
 4542 dopeman   15   0 19796 7124 1416 S  0.3  0.7   0:40.64 samp022svr
13504 dopeman   15   0  6180 1100  804 S  0.0  0.1   0:00.00 pure-ftpd

top - 00:16:29 up  4:13,  1 user,  load average: 2.28, 1.79, 1.86
Tasks: 107 total,   3 running, 102 sleeping,   1 stopped,   1 zombie
Cpu(s): 57.4% us,  1.0% sy,  0.0% ni, 40.9% id,  0.0% wa,  0.0% hi,  0.7% si
Mem:   1034796k total,   939464k used,    95332k free,    22240k buffers
Swap:  2096472k total,      660k used,  2095812k free,   516168k cached

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
14396 dopeman   15   0  243m 104m 5080 S 23.5 10.4   0:21.74 etded.x86
14581 dopeman   15   0  150m  84m 2784 S  4.3  8.4   0:02.23 etded.x86
 4542 dopeman   15   0 19796 7124 1416 S  0.3  0.7   0:42.50 samp022svr
14364 dopeman   15   0  6180 1172  856 S  0.0  0.1   0:00.01 pure-ftpd

is there something wrong there?

Top memory useage for the second server and keith's server is 15% CPU, I don't understand why dope's et_0 is causing so many problems =/

[darren]

How many players are on those ET servers ?
One of the things 2.60 brought was fixes for lower cpu usage:
http://www.gamershell.com/download_8686.shtml
Not sure how much affect that had but it was apparently a known issue.

[DaDopeman]

Hii Currently SNN is busy but im Da Dopeman as shown by records.. With old server we did not have major CPU hoggage i might have records of CPU logs i shall have to dig but when this happens not many are on and currently this cpanel is running 3 servers 2 mine and one is keiths keiths get occasional lagg and otherone we run doesnt have any lagg. The server is set at 54 people max like it was with old VPS. Anymore Questions or do not understand what i wrote ill clarify. Thx have a nice day.

[darren]

I'm not really sure why you might be seeing high cpu usage if you haven't in the past on another server. If you can build an install (not using cPGS, just normally) that doesn't have it, then we can try making a package from it to see if it performs well with cPGS too. You may have better luck tracking down game specific issues like this on forums centered on the game in question too as the traffic here isn't the highest.

[DaDopeman]

Ok thanks ill look into it