Re: Weird SCAN Issue

From: Steven Andrew <postora_at_gmail.com>
Date: Fri, 2 Nov 2012 15:40:53 -0700
Message-ID: <CAE6QNMt_pDHY51OtPyK2OSpn7nbV2Fy0C8v96eroB7=Z24Vcvg_at_mail.gmail.com>



Resending, as it was caught for over quoting it. Hopefully this get through. Frits, you hit the jackpot! Your guess was absolutely on the money. Well this was so weird. From the appNode1, i couldn't resolve the hostname of the remote server handler in the SCAN listener. However, i could resolve the same from appNode2. Both app nodes and db nodes had the same name server and same entries in /etc/resolve.conf Our Sysadmins had to make a change to remove an extra search list entry from resolve.conf (search $FQDN) and restart nscd to make it work. Not sure if that line messed up the lookup but looks like. After restarts, i was able to connect from appNode1 just fine. I still don't know how it worked on appNode2 as both appNodes had the same resolve.conf entries, perhaps the old caching was good..?. /etc/hosts on appNodes don't have any references to DB nodes/vips.
Anyways, thanks to all who replied.
--
http://www.freelists.org/webpage/oracle-l
Received on Fri Nov 02 2012 - 23:40:53 CET

Original text of this message