Re: Host Enterprise Manager
- From: "Oliver aka v1k1ng0" <ofabelo@xxxxxxxxx>
- To: oracle-l@xxxxxxxxxxxxx
- Date: Sat, 19 Apr 2008 12:35:57 +0100
Hello,
2008/4/18, Oliver aka v1k1ng0 <ofabelo@xxxxxxxxx>:
>
> Hello,
> I have 1 node with a Oracle DB single instance in AIX 5L (64bits). When I
> created the DB I did put that it uses Enterprise Manager. The DB is created
> correctly. When I try access to EM, it shows well, I put user and password
> and it shows a error because the host is other ... Where may I configure
> (what file) for that it has the correct host?
> When I run "hostname", the output is correct, the /etc/hosts file is
> correct too. Thanks beforehand.
>
I'm sorry for my bad english. I have more time now for giving to you more
data.
It's 2 nodes, each one with its single oracle 10gr2 instance (not RAC, not
Dataguard). The servers have AIX5L, 2 physical disks and much SAN disks.
Both nodes see the hdisk's of SAN disks and each node see its 2 physical
disks. I have /u01 in a LV of physical disk in each node, with Oracle
software installed (/u01/app/oracle, oracle_base, and
$oracle_base/product/10.2.0/db_1 like oracle_home).
I have /u02/oradata, /u03/oradata, and so on in SAN disks. They have jfs2
filesystems, I know that it hasn't concurrent access, for that, one will be
active and the other standby (it is all done manually, varyoff/on vg's,
umount/mount filesystems).
Each node has its listener, DB instance and EM repository. Both instances
work well for separated. The directory
/u01/app/oracle/product/10.2.0/db_1/dbs (where is the spfile) is like mount
point in a SAN disk, for both instances have the same spfile (in both nodes
have that mount point), so both nodes share the same spfile. The control
files, redo logs, and the rest of datafiles are shared too in SAN disks.
When I created the DB in both nodes, I configured all datafiles for the same
path (of san disks). Can someone understand me for now? :-)
Ok, now the problem: I installed like last the second node, so the second DB
was created later than 1st database. When I access to Enterprise Manager, it
puts like host the second node. Ok, it is good. But now, I stop all
processes in 2 node, umount filesystems, varyoff vg's and I go to 1 node and
start all. All well, except the EM, that it puts like host the second node!
:-(. I've fixed recreating the EM, with emca tool (sure, the second node
will show the fist host now in the EM). I've been looking the tables of
sysman schema. Someone knows if touching some table of sysman, I can change
this issue? Changing the host that EM attack for default :-? Thanks
beforehand and thank you very much for people than can reply me.
Cheers...
Other related posts:
