|
|
 |  |
Re: Slow server performanceFrom: Andreas Carlsson Date: Wednesday, July 9, 2003
Time: 5:13:09 pmI have serious problems accessing the webserver, so I can't read the
logs. I made some Top's though and noticed that Server Settings (I'm
on OS X) is taking around 100% all the time (it's a dual 1 GHz). Is
that normal?
However, I made some Ping statistics as well; My conclusion is that,
that everything on the DSL wire is making heavy packet losses. Since
I'm quite new to this I don't really know how much is much, but I
made some comparements:
ROUTER AT OFFICE:
--- 62.181.204.82 ping statistics ---
609 packets transmitted, 383 packets received, 37% packet loss
round-trip min/avg/max = 37.459/42.488/127.62 ms
DNS1
--- 62.181.204.83 ping statistics ---
669 packets transmitted, 411 packets received, 38% packet loss
round-trip min/avg/max = 34.842/38.91/137.342 ms
DNS2
--- 62.181.204.84 ping statistics ---
656 packets transmitted, 531 packets received, 19% packet loss
round-trip min/avg/max = 34.354/38.131/129.059 ms
WEBSERVER:
--- 62.181.204.85 ping statistics ---
642 packets transmitted, 550 packets received, 14% packet loss
round-trip min/avg/max = 34.487/38.163/190.505 ms
MAILSERVER:
--- 62.181.204.86 ping statistics ---
635 packets transmitted, 503 packets received, 20% packet loss
round-trip min/avg/max = 35.059/40.731/150.434 ms
SUNET IN SWEDEN
--- 192.36.125.2 ping statistics ---
573 packets transmitted, 571 packets received, 0% packet loss
round-trip min/avg/max = 23.266/25.201/174.237 ms
ANOTHER DNS IN SWEDEN:
--- 212.181.63.104 ping statistics ---
611 packets transmitted, 603 packets received, 1% packet loss
round-trip min/avg/max = 26.877/31.153/168.83 ms
BROADBAND PROVIDER'S GATEWAY:
--- 62.181.204.81 ping statistics ---
634 packets transmitted, 527 packets received, 16% packet loss
round-trip min/avg/max = 32.396/40.289/566.127 ms
Is there some way to ping just the http port etc?
Any thoughts are welcome, I am clueless.
Thanks,
/andreas
Len Conrad wrote:
>>Question is; What kind of problem? What can have happened without
>>us doing *anything* from this morning to afternoon?
>
>don't ask me, ask your machine. if it is Unix, then run "top" and
>see if their are any internal abnormalities.
>
>check your http logs for attacks that are consuming your http service.
>
>check all your logs for other error msgs.
>
>check memory and swapping (you can see both in top) since when a
>server goes into swapping, it can thresh itself into a very low
>level of service.
>
>Len
|

Return to Digital Point Solutions' Home Page |