wow Transparent Application Failover seems pretty cool. It seems like all the tools are out there to make a swappable db and swappable server environment possible without clustering the servers but nobody seems to have the tools put together.
Sean
wow Transparent Application Failover seems pretty cool. It seems like all the tools are out there to make a swappable db and swappable server enviroment possible without clustering the servers but nobody seems to have the tools put together. Don't get me wrong I'm not opposed to clustering.
Sean
>>> "john" <john.[Email address protected] 09/30/04 03:41PM >>>
**A LazyDBA.com subscriber has responded to your lazydba.com post**
**LazyDBA.com mail shield has forwarded you this email,
**and removed any attachments, and kept your email address secret
**from this person, and any viruses/trojans.
**If you reply to this email, the person will see your email address as normal
**Anything below this line is the original email text
Sean if clustering is out then Dataguard is a good way to go (or replication).
You don't need to switchover to the standby to take a backup - if you use RMAN and a CATALOG
you can backup the standby database (nice huh !!).
Additionally you can open the standby database for querying/reporting ....................... and then reinstantiate it to
managed recovery mode. Another plus !!!
It takes a bit more management, and you want a good network connect to the standby machine (private NIC)
You are correct in that this will not give you 100% availability- however it can be highly robust if implemented correctly. Look at your network and your application - also read up on Transparent Application Failover (TAF) - generally used in RAC but a good overview of a nice networking feature.
Hope that helps
John
________________________________
From: Sean Kenny [mailto:oracledba-ezmlmshield-x93496499.[Email address protected]
Sent: Thu 30/09/2004 21:28
To: LazyDBA Discussion
Subject: 24/7 solution 9i on W2k
I posted this last week with no response. :-(
anyway I need some advice. We are currently on 8.1.7 in a W2k enviroment. Db is about 80g and will be around 250 by 2006. Our database and user base has grown and is now at the point that downtime is a big problem. Even being down 3-4 hrs for a cold backup is unacceptable. With our previous lemon server we lost the db due to hardware problems 3 times in less then a year. Im presently looking at all possible options for a 24/7 system but it seems like a lot of overhead. Honestly I don't see anyway to have a 100% uptime environment maybe 95%.
Here are my thoughts so far.
The only server based hardware failover solution seems to be clustering. Not an option since we already have new non identical servers.
The only way to reboot without taking down the db is if you are clustered. I guess I'm down to 93% uptime already. I can improve that by getting off Windows. Anyway...
To eliminate DB failure you need a standby db. This is possible but how can you seamlessly switch from a production to a standby. Logicly is seems as if you would have to move all connections to the db first then flush the redo logs. Then bring the standby into production mode. If it was possible I would like to swap db's every few days to backup one of them them.
I'm currently reading up on datagard. Any info is appreciated.
Thanks
Sean
--------
website: http://www.LazyDBA.com
Please don't reply to RTFM questions
Oracle documentation is here: http://tahiti.oracle.com
To unsubscribe: see http://www.lazydba.com/unsubscribe.html
To subscribe: see http://www.lazydba.com
By using this list you agree to these terms:http://www.lazydba.com/legal.html
This message is for the designated recipient only and may contain privileged, proprietary, or otherwise private information. If you have received it in error, please notify the sender immediately and delete the original. Any other use of the email by you is prohibited.
Oracle LazyDBA home page