Re: Verifing Data Integrity
R Bankay wrote:
> To thoroughly verify data in a table slave/master, first stop the db applications. > > Then sum any numeric columns (I prefer a floating columns.) e.g., > > select sum(columnfp1), sum(colmnfp2) from table1; > > Do this on both master and slave. Should point out that if you choose an index with numeric value it will be much faster. >
When you have to scan the whole table, why will using the index will be
faster?
> > Bob B > > ----- Original Message ---- > From: Andrew Brosnan <andrew@broscom.com> > To: Paul Sindelar <paul.lists@mnwebhost.net> > Cc: replication@lists.mysql.com > Sent: Saturday, July 14, 2007 6:14:55 PM > Subject: Re: Verifing Data Integrity > > On 7/14/07 at 12:18 PM, paul.lists@mnwebhost.net (Paul Sindelar) wrote: > > >> Does anyone know a reliable/efficient way of thoroughly verifying >> the integrity of a slave? >> > > Would this be done on a routine basis? I think it's more realistic to do > routine sanity checks - perhaps you already do. What I do is check for > same number of records on master and slave, thoroughly verify *some* of > the data by checking random rows, compare file sizes, etc. I can't > imagine going through bit by bit without suspecting a problem. I think > I'd be more likely to reset the slave with a new snapshot of the masters > data. :-) > > Regards, > Andrew > > > -- > MySQL Replication Mailing List > For list archives: http://lists.mysql.com/replication > To unsubscribe: http://lists.mysql.com/replication?unsub=robertcb@yahoo.com > > > > > > > ____________________________________________________________________________________ > Park yourself in front of a world of choices in alternative vehicles. Visit the Yahoo! Auto Green Center. > http://autos.yahoo.com/green_center/ > >
--
MySQL Replication Mailing List
For list archives:
http://lists.mysql.com/replication
To unsubscribe:
http://lists.mysql.com/replication?unsub=lists@pantek.com
Received on Sun Jul 15 03:53:29 2007
This archive was generated by hypermail 2.1.8
: Thu Aug 02 2007 - 01:57:07 EDT
|