[ngw] I hope its not a repost - End of Support - GW7

Tue Apr 13 15:13:27 UTC 2010

Paul,   I'm assuming Linux here,  but may be wrong.    The big
difference between OES Linux and Netware clustering is that on NetWare,
if an app fails, it usually brings the whole server (or in the case of
clustering, the address space) down,  then NCS (cluster services) can
detect it and fail the cluster resource to another node.   On the Linux
side of the fence,   app fails with a seg fault, NCS can't detect it to
force a failover.   THEREFORE,   I'd suggest putting on some type of app
monitor solution to restart said app if it fails.   One way to go is
with GW Monitor with it's restart capabilities for the GW agents using
GWHA.   The other is a monitor script running as a cron job.   Here's an
example of a script I've used for a MTA named utah and it's GWIA:

#!/bin/sh
CNT=0
CNT=`ps -aef | grep -v grep | grep -c '@utah.mta'`
if [ $CNT -eq 0 ]
then
echo "restarting utah MTA"
echo "Restarted utah MTA" `! date` >> /var/log/gw/gwagentchk.log
        /etc/init.d/grpwise start utah
mail -s "utah MTA restarted" user at domain.com <
/var/log/gw/gwagentchk.log
else
        echo "utah MTA is running"
fi
CNT=0
CNT=`ps -aef | grep -v grep | grep -c '@gwia.cfg'`
if [ $CNT -eq 0 ]
then
echo "restarting utah gwia"
echo "Restarted utah gwia" `! date` >> /var/log/gw/gwagentchk.log
        /etc/init.d/grpwise start utah.gwia
mail -s "utah GWIA restarted" user at domain.com <
/var/log/gw/gwagentchk.log
else
        echo "utah GWIA is running"
fi

ciao,

--

Morris

Messaging Architects
Risk Free Email = Business Driven Email

Voice: 514.392.9220 ext. 253
Fax: 514.392.9120
Email: morris.blackham at messagingarchitects.com
 >>> Paul Allen <pallenr32 at gmail.com> wrote:
> Hi Danita
>
> Thanks for the confirmation...I am still waiting to hear back from my
local
> Novell Account Manager.
>
> I just thought it was strange to release 7.0.4 and a few weeks later
drop
> Standard support...I just haven't been hanging around the GroupWise
> coolsolutions page that much lately.
>
> The Forums are great source of Knowledge, and I really appreciate the
help I
> have had over the years from NGW !
>
> For the time being, I have loaded the cluster resource for the
MTA/GWIA on a
> server node without any other cluster resources and will just bounce
that
> server node whenever I get the segfault issue....its only happening
with
> this GWIA. might be related to the load as its the only pop/imap GWIA
we
> have servicing 750 users.
>
> best regards
> Paul
>
> On 13 April 2010 01:59, Danita Zanre <dzanre.ngwlist at gmail.com>
wrote:
>
> > I can't help you on your cluster issue, but indeed, there is no
more
> > support for GroupWise 7 since April 2.  I've had a number of "new"

> customers
> > come to me for support since then!  In fact, that reminds me that I
need to
> > make a blog entry about it <heehee>.
> >
> > The only Novell support available for GW7 now is through
forums.novell.com.
> > which as you know is volunteer driven, but does have at least some
backend
> > support from Novell so while we can't get Novell to fix anything
that's
> > broken in GW7 now, if there is some "knowledge" that we are lacking
to
> solve
> > a GW7 problem we do still have some hope of finding someone from
Novell who
> > might be able to point us in the right direction.
> >
> > But, no more SR support for GW7 - period.
> >
> > Danita
> >
> >
> > On Sun, Apr 11, 2010 at 7:12 PM, Paul Allen <pallenr32 at gmail.com>
wrote:
> >
> >> Hello
> >>
> >> I tried opening a SR with Novell today for a segfault issue I am
having
> >> with GW7.0.3 GWIA (pop/imap) that has happened twice this  week on
my
> >> SLES10sp3 OES2sp2 cluster.
> >>
> >> No can do - only GW8 supported by chat - since 2 April...
> >>
> >> If anyone here can shed some light on this it would be great...I
was
> >> hoping to find out during the SR process how to remove the
secondary
> >> ipaddress of the cluster resources when the segfault occurs...if
you
> offline
> >> the cluster resources with the MTA/GWIA the secondary ipaddress
stays up
> and
> >> the next time you try to load the cluster resources it goes
comatose as the
> >> ipaddress is in use...only solution is to reboot the server node -
or in
> >> more extreme cases the entire cluster.
> >>
> >> cheers
> >> Paul
> >>
> >>
> >>
> >>
> >> --
> >> Any intelligent fool can make things bigger and more complex... It
takes a
> >> touch of genius - and a lot of courage to move in the opposite
direction..
> >> Albert Einstein
> >>
> >> _______________________________________________
> >> ngw mailing list
> >> ngw at ngwlist.com
> >> http://ngwlist.com/mailman/listinfo/ngw
> >>
> >>
> >
> >
> > --
> > Danita Zanrè
> > Keep in touch!
> > http://www.twitter.com/GWGoddess
> > http://www.facebook.com/Caledonia.net
> > http://www.linkedin.com/in/danitazanre
> >
> >
> > _______________________________________________
> > ngw mailing list
> > ngw at ngwlist.com
> > http://ngwlist.com/mailman/listinfo/ngw
> >
> >
>
>
> --
> Any intelligent fool can make things bigger and more complex... It
takes a
> touch of genius - and a lot of courage to move in the opposite
direction.
> Albert Einstein

Considering a Migration to GroupWise 8 or Exchange? 
Learn more at http://www.messagingarchitects.com/webcasts