Date: Sat, 03 Jul 1999 09:53:09 -0700 From: JMS Internet <webmaster@jmsinternet.com> To: Alfred Perlstein <bright@rush.net> Cc: freebsd-stable@FreeBSD.ORG Subject: Re: Server Slowdown Message-ID: <4.1.19990703094749.00abc430@mail.sirius.com> In-Reply-To: <Pine.BSF.3.96.990703114850.14320x-100000@cygnus.rush.net> References: <4.1.19990703093016.00ceb180@mail.sirius.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Here is some more information about my system:
Intel PII 350, 384 Megs of RAM, 2 Multi-Gig hard drives (neither at capacity).
My co-location company and myself know that there is more transfer trying
to get through because we see it hitting a "roof" on it's transfer at about
3200kbps,
and it won't go past that, and there is no steady up and downs, it's flat
lining
at that type of transfer, so we know something is wrong, and speed for
transfers
is VERY slow... I ran a test at netmechanic.com, and it rated it as "poor". A
site to look at for yourself would be www.sitepalace.com. Thanks for your
help...
Here is the information requested below:
FROM TOP:
last pid: 89067; load averages: 0.62, 0.81, 0.75 up 3+21:40:19 09:53:38
279 processes: 1 running, 272 sleeping, 6 zombie
CPU states: 7.8% user, 0.0% nice, 8.2% system, 6.2% interrupt, 77.8% idle
Mem: 77M Active, 239M Inact, 45M Wired, 12M Cache, 8346K Buf, 1292K Free
Swap: 522M Total, 13M Used, 508M Free, 3% Inuse
FROM SYSTAT:
1 users Load 0.76 0.86 0.76 Sat Jul 3 09:52
Mem:KB REAL VIRTUAL VN PAGER SWAP PAGER
Tot Share Tot Share Free in out in out
Act 60648 724 141864 860 16868 count
All 382964 1784 3282820 2264 pages
145 cow Interrupts
Proc:r p d s w Csw Trp Sys Int Sof Flt 73 zfod 1389 total
3 241 171 346 2554 1389 71 340 46028 wire 100 clk0 irq0
76484 act 128 rtc0 irq8
7.1%Sys 13.7%Intr 4.2%User 0.0%Nice 75.0%Idl 243456 inact 343 pci irq11
| | | | | | | | | | 15360 cache fdc0 irq6
====++++++>>- 1508 free 818 wdc0
irq14
daefr wdc1
irq15
Namei Name-cache Dir-cache prcfr
atkbd0 irq
Calls hits % hits % react
14218 14183 100 6 0 pdwake
pdpgs
Discs fd0 wd0 wd1 intrn
KB/t 0.00 8.94 0.00 8348 buf
tps 0 46 0 29649 desiredvnodes
MB/s 0.00 0.40 0.00 20442 numvnodes
4759 freevnodes
At 11:51 AM 7/3/99 -0500, Alfred Perlstein wrote:
>On Sat, 3 Jul 1999, JMS Internet wrote:
>
>> I have been experiencing slow download times with my co-located server...
>> After changing out my HD, network card, and adding memory I have not seen
>much
>> improvement at all. But, after adjusting different Apache settings (max
>> users, max keep
>> alives, etc.) I have noticed a change of approximately 600kbps of transfer
>> higher then
>> my original settings. What I would like to know is if anyone has
>> experienced these types
>> of problems, or if anyone is willing to give me advice on httpd.conf fine
>> tuning. My server
>> currently transfers approximately 30 gigs a day, and is only using
>> approximately 30% of
>> it's allocated line. Any help or advice is greatly appreciated... I have
>> no idea where to turn
>> at this point..
>
>30 gigs a day is pretty nice, but i'm sure we can do better. :)
>
>More helpful would be :
>
>the exact configuration (ram, netcard, cpu, disks)
>a snapshot or two from "top" and "systat -vmstat"
>what is your current peak transfer rate?
>are you sure there is more than 30gigs/day demanded?
>
>-Alfred Perlstein - [bright@rush.net|bright@wintelcom.net]
>systems administrator and programmer
> Win Telecom - http://www.wintelcom.net/
>
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4.1.19990703094749.00abc430>