Mailing List Archives
Authenticated access
|
|
![[Computer Systems Lab]](http://www.cs.wisc.edu/pics/csl_logo.gif)
|
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [Condor-users] Slow response
- Date: Thu, 12 May 2005 15:21:46 +0100
- From: "Sandy Spence" <axs@xxxxxxxxxx>
- Subject: RE: [Condor-users] Slow response
Hi Rob,
here is the output from time condor_status
[axs@wig-01 ~]$ time condor_status
Name OpSys Arch State Activity LoadAv Mem
ActvtyTime
wig-01.cluste LINUX INTEL Owner Idle 0.030 1024
22+15:25:16
wig-02.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+00:50:04
wig-03.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+00:35:04
wig-04.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+00:35:04
wig-05.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+00:35:04
wig-06.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:05:08
wig-07.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:05:21
wig-08.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:05:17
wig-09.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:05:16
wig-10.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:05:28
wig-11.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:04:59
wig-12.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+01:00:04
wig-13.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:05:51
wig-14.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:06:46
wig-15.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:06:32
wig-16.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:06:26
wig-17.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:05:23
wig-18.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:06:29
wig-19.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:06:18
wig-20.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:05:20
wig-21.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:05:39
wig-22.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:07:16
wig-23.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:05:47
wig-24.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:07:10
wig-25.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:08:03
wig-26.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:03:56
wig-27.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:05:23
wig-28.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:04:28
wig-29.cluste LINUX INTEL Unclaimed Idle 0.040 1024
0+02:06:32
wig-30.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:03:13
wig-31.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+01:30:04
wig-32.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+00:50:04
wig-33.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:06:48
wig-34.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+00:50:04
wig-35.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:05:23
wig-36.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:08:09
wig-37.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:03:58
wig-38.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:03:21
wig-39.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:03:24
wig-40.cluste LINUX INTEL Unclaimed Idle 0.000 1024
0+02:06:59
Machines Owner Claimed Unclaimed Matched Preempting
INTEL/LINUX 40 1 0 39 0 0
Total 40 1 0 39 0 0
real 3m4.741s
user 0m0.028s
sys 0m0.011s
As you can see it is a little bit longer, this is not always the case
though.
Cheers,
Sandy
Computer Officer, RA Certification Manager
Department of Computer Science - UWA
Llandinam Building
Penglais Campus
Aberystwyth
Ceredigion
Wales - UK
SY23 3DB
Tel: (01970)-622433
Fax: (01970)-628536
-----Original Message-----
From: Rob Fletcher [mailto:rpf1@xxxxxxxxxx]
Sent: 12 May 2005 13:57
To: Sandy Spence
Subject: Re: [Condor-users] Slow response
On Thu, 12 May 2005, Sandy Spence wrote:
> Hi,
>
> I have a private condor pool, 1 master with 2 NICs and 39 slaves,
> there are times when entering condor_status takes a long time to show
> any output and if running a test job the job takes a long time to
> start, are there any tweeks I can perform that would speed up these
> processes, I am running Condor 6.6.8. Which logs are best to view to
> get a better feel for what might be happening.
>
Hi,
I am running with master with 2 nics, and 4 slaves (master WBEL, slaves
WinXP), and I find condor_status takes and age! Or am I simply just
expecting too much?
Using a new HP switch too, but the boxes are old PIII 700's with 384 Mb RAM
(so I sort of expect a slow response).
Also, condor_submit seems to take age too.
Running 6.6.9
e.g.
[root@WEREWOLF transit]# time condor_status
Name OpSys Arch State Activity LoadAv Mem
ActvtyTime
localhost.loc LINUX INTEL Unclaimed Idle 0.000 373
0+00:45:36
xpnode0 WINNT51 INTEL Unclaimed Idle 0.000 384
0+03:09:30
xpnode1 WINNT51 INTEL Unclaimed Idle 0.010 384
0+02:19:47
xpnode2 WINNT51 INTEL Unclaimed Idle 0.020 384
0+02:19:59
xpnode3 WINNT51 INTEL Unclaimed Idle 0.010 384
0+02:24:57
Machines Owner Claimed Unclaimed Matched Preempting
INTEL/LINUX 1 0 0 1 0 0
INTEL/WINNT51 4 0 0 4 0 0
Total 5 0 0 5 0 0
real 0m39.773s
user 0m0.010s
sys 0m0.010s
Cheers,
Rob
+==========================+=======================================+
| Dr R P Fletcher (Rob) | e R.Fletcher@xxxxxxxxxx |
| Graphics Coordinator | t +44 (0)1904 433816 |
| Computing Service | f +44 (0)1904 433740 |
| University of York | w http://www-users.york.ac.uk/~rpf1/ |
| YORK YO10 5DD, UK | |
+==================================================================+