[TriLUG] Anyone here know much about kernel programming?
igor at igorpartola.com
Mon Jul 11 14:15:38 EDT 2011
If I understand correctly, you still would need to deal with the
applications that are using the file system, at least on some level. Unless
all the applications involved only use atomic transactions on the file
system (which is pretty rare, from what I've seen), getting a consistent
snapshot will always involve shutting down/pausing some of the applications
first, then taking the snapshot (through whatever means necessary) and then
starting the applications back up. I think from this point of view LVM is
pretty good since it only takes a few moments to create the snapshot. You
could also look at rdiff and rdiff-backup for more ad-hoc setups where speed
is not as much of an issue.
On Mon, Jul 11, 2011 at 2:10 PM, Paul Bennett <paul.w.bennett at gmail.com>wrote:
> On Mon, 11 Jul 2011 13:57:36 -0400, Cristóbal Palmer <cmp at cmpalmer.org>
> If you are using LVM, as you should be, it is pretty simple:
> There's a number of different systems in play here. I'm not the sysadmin
> for them (thank the maker), but there are some cases where we have to (e.g.)
> take down one half of a two-node cluster for a few seconds in order to make
> a consistent backup of the data.
> I'm looking for a kind of "One Ring" solution that can be dropped in place
> for situations like this, where for whatever reason the system is
> adminsitered in such a way as to make LVM and CLVM snapshot backups tedious
> or difficult or annoying or something.
> This message was sent to: Igor Partola <igor at igorpartola.com>
> To unsubscribe, send a blank message to trilug-leave at trilug.org from that
> TriLUG mailing list : http://www.trilug.org/mailman/**listinfo/trilug<http://www.trilug.org/mailman/listinfo/trilug>
> Unsubscribe or edit options on the web : http://www.trilug.org/mailman/**
> TriLUG FAQ : http://www.trilug.org/wiki/**
More information about the TriLUG