Am 25.01.2011 15:56, schrieb Johan De Meersman:
>> jesus christ nobody cares if they are binary replica as long
>> as the data is consistent and ident
> Actually, I can see this being an issue if you're using LVM snapshot backups
> or another similar technique - if the datafiles aren't all identical you
> won't be able to restore to any machine from a single backup.
Where exactly do you see any problem?
* the master writes his bin-log in row format
* the slaves writes his relay-log and managing inserts updates
* lvm makes a snapshot of relay-log / datafiles
There is none and if there could be one LVM must be broken
because a snapshot has to be consistent
So if you stop the slave, make the snahpshot and start the slave
again there are all buffers written to the vfs-layer and the snapshot
must have a defined state.
Only if the db-server is running and have some data in memory cache
you could have any troubles and that is why not copy the files
as long the server is running
Nobody out there makes a copy of database files while the server
is running, really nobody!
Mit besten Grüßen, Reindl Harald
the lounge interactive design GmbH
A-1060 Vienna, Hofmühlgasse 17
CTO / software-development / cms-solutions
p: +43 (1) 595 3999 33, m: +43 (676) 40 221 40
icq: 154546673, http://www.thelounge.net/
Attachment: [application/pgp-signature] OpenPGP digital signature signature.asc