Re: [cobalt-developers] Raq 3 Hangs! Websites down

["earthlink" <joelmon2002@xxxxxxxxxxxxx>]

Oh, no, that's fine. Thanks for the reply I noticed chiliasp was hung, the
service disabled, couldn't access the web admin panel
When I restarted the asp server and rebooted the cobalt, the 'top' command
showed no more 50+% cpu measurement
for chiliasp and 'netstat' no longer showed 9 bad connections. All was back
to normal, maybe that was the problem?

Anyway, here was the 'top' results before I did the restarts for the asp
server:

 12:02pm  up 1 day, 17:44,  0 users,  load average: 1.75, 1.33, 1.23
71 processes: 67 sleeping, 4 running, 0 zombie, 0 stopped
CPU states: 58.6% user, 41.3% system,  0.0% nice,  0.0% idle
Mem:  257816K av, 243872K used,  13944K free, 105288K shrd,  21440K buff
Swap: 131536K av,   3652K used, 127884K free                149312K cached

  PID USER     PRI  NI  SIZE  RSS SHARE STAT  LIB %CPU %MEM   TIME COMMAND
 1403 root      20   0  5756 5756  2984 R       0 52.5  2.2  2299m caspeng

Now after I restarted the asp server and fixed asp from hanging, I use top
and get this:
  8:15am  up 15:36,  0 users,  load average: 0.01, 0.02, 0.00
75 processes: 72 sleeping, 3 running, 0 zombie, 0 stopped
CPU states:  1.1% user,  1.1% system,  0.0% nice, 21.2% idle
Mem:  257816K av, 134216K used, 123600K free, 103776K shrd,  12824K buff
Swap: 131536K av,      0K used, 131536K free                 57196K cached


And the netstat was had 9 "CLOSE_WAIT" status reports for 9 chiliasp
connections

The reboots flushed that out too and no more 'close wait' status for any
connection. All are 'established'

Maybe that was the problem?




----- Original Message -----
From: Steve Werby <steve-lists@xxxxxxxxxxxx>
To: <cobalt-developers@xxxxxxxxxxxxxxx>
Sent: Wednesday, April 03, 2002 8:07 AM
Subject: Re: [cobalt-developers] Raq 3 Hangs! Websites down


> "earthlink" <joelmon2002@xxxxxxxxxxxxx> wrote:
> > Or could it be that only 16 megs of 258 ram are available?
> >
> > (I checked using 'top')
>
> Joel, please post the output of top, or since it's easier on the eyes
> consider posting the output of "cat /proc/meminfo" (which is where top
gets
> its data for the memory).  I think you may be misinterpreting what the
> output means.  If you post one of us can interpret it for you.  If you
fully
> understand what it all means, do not take this as a personal attack; it's
> just my experience that most people think they know what it means, but
> really don't.
>
> --
> Steve Werby
> President, Befriend Internet Services LLC
> http://www.befriend.com/
>
>
>