trying to get heartbeat pumping bits

Geoff Nordli geoff at gnaa.net
Thu Apr 6 00:43:18 MDT 2000


There might be a bug in the code that leaves the lock file there.  Is
anyone else having that problem.  If not it might be somethat that
would be good to add to the FAQ's.

So in the ha.cf file I would need both servers.  Is that correct?
The ha.cf file almost gives you the wrong impression.  I thought it
was looking for a node "name" + the "uname -n"

There are some bits pumping now.

How do you specify which one is the master??

thanks,






> -----Original Message-----
> From: alanr at rmi.net [mailto:alanr at rmi.net]On Behalf Of Alan Robertson
> Sent: Wednesday, April 05, 2000 11:23 PM
> To: geoff at gnaa.net
> Cc: Linux-Ha List (E-mail)
> Subject: Re: trying to get heartbeat pumping bits
>
>
> Geoff Nordli wrote:
> >
> > I think I am really way out in left field on this one, but please
> > bear with me.  I am trying to create a fail-over system.  I thought
> > that I would start with apache--you know something easy ;-)
> > I think that there is something wrong with my design.  The error
> > may be a red herring.
> >
> > I get this error when I start up heartbeat on both systems:
> >
> > heartbeat: 1996/04/05_17:52:06 info: Configuration
> validated. Starting
> > heartbeat
> > heartbeat: 1996/04/05_17:52:07 error: cannot lock line /dev/ttyS0
> > heartbeat: 1996/04/05_17:52:07 error: cannot open serial /dev/ttyS0
>
> I assume you're running this as root.  It's trying to create a file
> named
> 	/var/lock/LCK..ttyS0
>
> I assume that it should be able to do so.  Do you maybe have
> a [m]getty
> or some other process running on ttyS0?  That's my guess.
>
> > I am using Heartbeat 0.47-1.i386.rpm, on a RH 6.1, and RH
> 6.2 System.
> >
> > The main machine is called moon.  The fail-over is called venus.
> >
> > I can echo "hello" between the two machines on /dev/ttys0.
> >
> > This is the config I have on moon:
> >
> > ha.cfg:
> > keepalive 2
> > deadtime 15
> > serial  /dev/ttyS0
> > node    moon    moon.highway-i.com
> > node    venus   venus.highway-i.com
>
> I'm sure this isn't what you want.  You should ONLY list one name for
> each machine in the "node" section.  You should list only the
> name that
> comes out when you say "uname -n".  With the current code, only two
> nodes should be listed.  You've listed four.  This error
> appears on the
> other machine also.
>
> Everything else looked fine.
>
> The unable to lock problem is a real problem.  You can check the file
> /var/lock/LCK..ttyS0 and look at the pid contained in it, and see what
> process has the tty locked and go from there.
>
> 	-- Alan Robertson
> 	   alanr at suse.com
>
> --------------------------------------------------------------
> ----------------
> Linux HA Web Site:
>   http://linux-ha.org/
> Linux HA HOWTO:
>
http://metalab.unc.edu/pub/Linux/ALPHA/linux-ha/High-Availability-HOWTO.html
----------------------------------------------------------------------------
--





More information about the Linux-HA mailing list