Interim backup solutions

Hi,

From reading this forum I can see that the universal backup system that linode is aiming for is a seriously tall order. I'm sure you guys will get there with flying colours.

In the meantime I'm taking your advice about doing my own backups, but have been having a bit of a crisis of confidence.

I'm playing around with rsync for doing incremental backups from one Linode to another, including rsync-ing binary log files from the DB, but am a bit concerned I might be screwing this up.

Yep, there's plenty of help on Google; in fact, rather too much. Every basic solution is obviously beset with edge cases and gotchas (e.g. backing up crontabs etc.)

Given the amount of experience on this forum, could anyone recommend a decent walkthrough for this kind of thing? It may be that backup schemes just have to be either 100% bespoke, or the Holy Grail that Linode's been after for almost a year now.

It would be great if a few backup-related walkthroughs could then be included in the Linode library. You never know - it might encourage the purchase a few more linodes for backup purposes.

Cheers

Peter

3 Replies

Hi Peter, you might find the following guides helpful (from this link in the Linode Library):

You can use the first link to "double check" your method against ours, and investigate the others as possible alternatives to what you're doing now. :) Hope it helps!

Rather than rsyncing mysql, why not just run a slave on the backup server?

Hi guys,

Thanks for the suggestions. Good ideas.

I was thinking that duplicating the whole disk image would be good to give a bulletproof backup every once in a while. One of my Linodes had to be moved as part of the power upgrading work Linode was doing and even moving it within their datacenter the estimated time to do this was over an hour - too long for my servers to be offline.

I guess if you had a couple of servers then you could take one offline and do this, but then you'd still need to sync it back from the live one when you'd finished….

Rsync looks like the way to go, and I take the point about slaving the mysql server. I've been doing this a bit and it seems to be okay. I found the process a bit manual, and in our development environment there are quite a few databases appearing and disappearing, so I'd need to use a decent script to scan this and turn on / off the syncing as necessary.

I had a few problems with rsync to do with the ownership of files (a bit like some of the problems you can have with using tar), but I think this is my lack of experience.

Looks like I need to spend some more time with Rdiff :-)

Thanks again for the suggestions.

Best wishes

Peter

Reply

Please enter an answer
Tips:

You can mention users to notify them: @username

You can use Markdown to format your question. For more examples see the Markdown Cheatsheet.

> I’m a blockquote.

I’m a blockquote.

[I'm a link] (https://www.google.com)

I'm a link

**I am bold** I am bold

*I am italicized* I am italicized

Community Code of Conduct