[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [cobalt-users] crashes



On Thu, 9 Mar 2000, Gary Peltola wrote:

> This is the 2nd time in 2 weeks (seems about one week ago the 1st went
> down), its a brand new server running only...3 weeks...it has about 230
> clients moved off a raq2
> 
> the 1st crash, i was able to get into TELNET but couldnt do anything...that
> may have been due to the logrotate...because i manuelly ran that before i
> went to bed.
> 
> This time around, all i did was setup a few clients and that was it...went
> to bed...my site alert saids it stopped responding at 3:30 am to 9:30 am
Seems to me as you have a runaway script...

Do you have any cgi-scripts running on the machine?

If you have, check those, they might have run OK on the RaQ2 but when
moved to a newer Perl-version they might behave differently. 

Best regards,

Rickard Osser
Manager
Osser Brosoft AB		Distributor of Cobalt Networks servers
Maria Bangata 6			Computer Consultants
S-118 63 Stockholm, Sweden	Networking, DOS/Win/Mac/Linux/Unix
Tel: +46-8-798 29 27		E-mail: ricky@xxxxxxxx
Fax: +46-8-668 89 10		WWW: http://www.brosoft.net

> Regards
> Gary
> 
> ----- Original Message -----
> From: "Mat Kovach" <mkovach@xxxxxxxxxxxxxxxxx>
> To: <cobalt-users@xxxxxxxxxxxxxxx>
> Sent: Thursday, March 09, 2000 10:46 AM
> Subject: Re: [cobalt-users] crashes
> 
> 
> > On Thu, Mar 09, 2000 at 09:52:20AM -0900, Gary Peltola wrote:
> > : 1. Where do i start in looking for answers
> >
> > Well, you started in the right spot, the log files....
> >
> > : 2. what do these log entrys mean (they are cut and pasted from various
> > : times)
> >
> > I'll skip the named (bind) error messages, while those might be part
> > of the cause, I want to focus on something.
> > : Mar  9 09:27:49 ns inetd[423]: fork: Resource temporarily unavailable
> >
> > This is not good.  Basically inetd is attemping to create a process and
> > it can.  There are serveral causes for this:  Lack of Memory, too many
> process
> > running, huge amounts of inetd connections.
> >
> > How often does this happen? Daily, Weekly, Monthly?
> >
> > I would create a crontask that e-mails the following in and see if that
> helps
> > find a cause.  Something like:
> >
> > #!/bin/bash
> >
> > TMPFILE="/tmp/crashdebug"
> >
> > netstat -a >> $TMPFILE
> > ps aux >> $TMPFILE
> > free >> $TMPFILE
> > df >> $TMPFILE
> >
> > cat $TMPFILE | mail -s 'crashdebugging' emailaddress@xxxxxxxxxxxx
> >
> > rm -f $TMPFILE
> >
> > exit 0
> >
> >
> > --
> > Mat Kovach                                      mkovach@xxxxxxxxxxxxxxxxx
> > Cleveland Linux User Group                       http://cleveland.lug.net
> >
> > _______________________________________________
> > cobalt-users mailing list
> > cobalt-users@xxxxxxxxxxxxxxx
> > To Subscribe or Unsubscribe, please go to:
> > http://list.cobalt.com/mailman/listinfo/cobalt-users
> >
> 
> _______________________________________________
> cobalt-users mailing list
> cobalt-users@xxxxxxxxxxxxxxx
> To Subscribe or Unsubscribe, please go to:
> http://list.cobalt.com/mailman/listinfo/cobalt-users
>