[Gluster-infra] Backup strategy for our infrastructure?

[Justin Clift]

On 01/09/2014, at 5:56 PM, Louis Zuckerman wrote:
> Ok I enabled scheduled image creation for all the cloud servers.
> 
> For some reason a few servers asked me what time of day to take the snapshot, but not how long to retain them.  For everything else I wasn't asked the time of day, and was only asked for retention period.  No idea why that is.
> 
> Also noteworthy, the images only capture the system disk, not any attached block storage.  So that's pretty sad.
> 
> Here's details for all the systems:
> 
> backups.c.g.o - 7 day schedule
> download.g.o - only let me pick time of day, not retention
> freebsd0.c.g.o - 7 day, warned that attached storage won't be imaged, only system disk
> g4s* - 7 day
> gfs-rackspace-f20-1 - 7 day, warned that attached storage won't be imaged, only system disk
> gluster-mirror2 - only let me pick time of day, not retention
> gluster-swift2 - 7 day
> gluster.com-mirror1 - only let me pick time of day, not retention
> netbsd0 - 7 day, warned that attached storage won't be imaged, only system disk
> netbsd7-gfs2 - 7 day
> slave* - 7 day, warned that attached storage won't be imaged, only system disk
> staging - 7 day
> tiny - 7 day
> www.g.o - 7 day

Looking in the Rackspace UI now, they're all showing up.  So, this
seems to be working. (not that we've testing restoring yet)

With those three server which only let you pick the time of day,
and not the retention, it's likely due to their server type.

download.gluster.org, gluster-mirror2, and gluster.com-mirror1
are all "old" VM's using Rackspace\s previous generation
infrastructure.  They're also really expensive (taking almost
1/3 of entire budget by themselves).  We should probably migrate
anything on them to new hosts in the next few months. :)

+ Justin

--
GlusterFS - http://www.gluster.org

An open source, distributed file system scaling to several
petabytes, and handling thousands of clients.

My personal twitter: twitter.com/realjustinclift