February 11, 2006 at 10:53 am
My company has one Sql 2000 server that supports a Point-Of-Sale system that is critical to its business. If this system crashes what is the best way to restore it with minimum down time?
Any ideas or suggestions will be greatly appreciated.
February 12, 2006 at 7:22 pm
You should plan for this type of disaster, I've implemented something simple but serves its purpose
1. change recovery model of the database to full
2. create jobs that will do a full backup every start of the day
3. create jobs that will do 15 minute interval log backup until next full backup
4. schedule both jobs to run automatically, make sure that you receive notification like emails when either of the jobs fail
Note: a backup file naming convention will help a lot in automating the entire process, not to mention stamps each file with a date and time or date atleast e.g. [backupname]yyyymmdd.bak
With this plan, you get to do point-in-time restoration
HTH
February 13, 2006 at 3:05 am
This needs to be driven by the business. If there are no formal SLAs agreed, then this must be the first step. When you have a written and signed-off document giving the business expectations, you have something to work with.
If the business can cope with the PoS system being down for 7 days, then you have time to buy, install and configure a new server. If they think a 30-minute outage is a disaster then you need something a tad more robust.
For 30-minute availability you must have a second server fully configured, with all changes being replicated to it. If the business expects 30-minute recovery even if the data centre catches fire, then your backup server must be off-site. When you have got what looks like a resilient solution, you must test it.
Where I work, we have GB and US datacenters, with all business data replicated between them mainly via a SAN. We normally run using GB servers as most staff are GB based, but we have at least 2 days per month running from US. Failover typically takes about 20 - 30 minutes. This may sound fairly good, but I have seen organisations achieving failover of bigger stuff than ours in under 5 minutes.
Original author: https://github.com/SQL-FineBuild/Common/wiki/ 1-click install and best practice configuration of SQL Server 2019, 2017 2016, 2014, 2012, 2008 R2, 2008 and 2005.
When I give food to the poor they call me a saint. When I ask why they are poor they call me a communist - Archbishop Hélder Câmara
February 13, 2006 at 8:39 am
Thanks all for the advice. Since I am new to this situation management is listening to me to purchase a backup server.
February 13, 2006 at 8:56 pm
All of the above options are good, and there are a number of mechanism to physically achieve this depending on cost & paranoia needed (ranging from log backups, RAID/SAN drives, log shipping, clustering etc etc). But I suspect the real answer is to talk to the vendor that supplied the system, and if it was developed internally then find any documentation you can and start from there!
Viewing 5 posts - 1 through 4 (of 4 total)
You must be logged in to reply to this topic. Login to reply