Re: Monitoring server
- From: Martin Bach <development@xxxxxxxxxxxxxxxxx>
- To: adar666@xxxxxxxxxxxx
- Date: Sat, 09 Jan 2010 12:15:48 +0000
Hi Ram,
the company I am currently working for have thought about implementing
Grid Control as well, with the additional requirement to monitor
databases across continents (mainly UK and USA right now).
The setup we came up with would have been similar to this, per
continent. As a side note, DBConsole is in use for many databases right
now and every developer deems having access to the performance page (yes
there are licenses :) is as essential as the availability of the
database itself which opens a Pandora's box....
2 node RAC cluster for the repository database (11.1.0.7 projected, but
might as well be 11.2 as soon as it's certified)
2 OMS installations on 2 nodes with a hardware load balancer. This has
the advantage that only 1 URL needs to be bookmarked by users.
We are also considering an identical setup in a remote data centre using
a physical standby for the repository but that will most likely not be
implemented due to budget restrictions.
Hope this helps,
Martin
--
Martin Bach
OCM 10g
http://martincarstenbach.wordpress.com
On 23/12/09 07:31, Yechiel Adar wrote:
> We use a 2 node rac.
> The Grid repository is Rac database that is shared on the two nodes.
> Each node has grid and each grid monitors the other grid.
>
> The monitored servers are divided between the two grids, so each one
> upload the data to one grid.
> Since both grids update the same database, the whole data is available
> from each grid.
>
> Adar Yechiel
> Rechovot, Israel
>
>
>
> Ram Raman wrote:
>> Listers,
>>
>> While configuring a monitoring server like Grid control server (or any
>> other type), how much of High availability or redundancy factors
>> should we consider. What happens when the monitoring server itself
>> goes down. I am curious to know how it is handled.
>> Thanks.
>>
> --
> //www.freelists.org/webpage/oracle-l
>
>
--
//www.freelists.org/webpage/oracle-l
Other related posts:
