Date: Tue, 6 Jul 2010 09:42:55 +0100 From: krad <kraduk@googlemail.com> To: Tim Daneliuk <tundra@tundraware.com> Cc: Liste FreeBSD <freebsd-questions@freebsd.org> Subject: Re: Global backup solution for FBSD & Ubuntu Message-ID: <AANLkTilhno2TYA5jhq_wWLxAhppY1lCbvgzIEt04O1aZ@mail.gmail.com> In-Reply-To: <4C3108B6.8070606@tundraware.com> References: <A5CD3711-70B2-4A07-836C-268AD11CA5D7@todoo.biz> <4C3108B6.8070606@tundraware.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 4 July 2010 23:18, Tim Daneliuk <tundra@tundraware.com> wrote: > On 07/04/10 16:43, bsd wrote: > >> Hello, >> >> I am trying to build a global backup solution for couple of strategic >> servers (7) based on two operating systems : >> >> - FreeBSD (6 - 7 // soon 7 - 8) >> - Ubuntu 8.04 LTS >> >> These servers are hosting some strategic components mainly related to DN= S >> infrastructure and databases. >> >> >> For the moment I am backing up these server using network based backup >> solution: >> >> - A "duplicity" based solution which backs up key directories in my >> infrastructure on a remote FTP server provided by my hosting company. >> - A "dump" of some key components which I am doing on regular basis for >> FreeBSD servers. >> - Duplicity is also used for the Ubuntu servers. >> - Databases are replicated "live" on a remote server using "slony" for t= he >> most strategic ones (Postgres DB) and using mysql dump export for MySQL. >> >> >> =95 I am not a 100% sure these solutions will allow me to restart rapidl= y >> from a crash, specially for Ubuntu servers. >> =95 I would like to know which solution(s) you have deployed at what cos= t >> for what results ? >> >> I am actually considering couple of different solutions >> >> - SAIT solution and backula. >> - Disk based solution (maybe also with backula). >> =85 >> >> >> I have couple of servers that will reach their end of life that could be >> recycled as backup solution at a very convenient price=85 >> >> >> > > I wrote a simple shell-based solution for this problem some time ago. It > (and FreeBSD > instructions) can be found: > > http://www.tundraware.com/Software/tbku/ > > > I am able to recover a production server (DNS, dhcp, http, sendmail, > etc...) in under > 30 minutes using this technique. > > _______________________________________________ > freebsd-questions@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to " > freebsd-questions-unsubscribe@freebsd.org" > we use an rsync based solution at work. All the files are basically rsyncd onto a big opensolaris filer backed with zfs. We then snapshot each hosts file system after the completed backup. It then gives us an incremental forever backup so is generally quite fast to do. Restores are also fairly fast depending on the size of the data set. For a full restore I boot into the new box on a liveusb os, partition/slice, newfs, mount and push the rsync back. All fairly easy and quick. With regard to database backups, we run all our mysql and oracle dbs on zfs= . This allows us to put a global write lock on the db and flush everything to disk. We then snapshot the db zfs fs and remove the write lock. Alternatively if its a mysql slave, we just stop the slave, flush and snap. This means we can take hot backups of all our dbs with minimal impact.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?AANLkTilhno2TYA5jhq_wWLxAhppY1lCbvgzIEt04O1aZ>