| 1 |
On Tue, 16 Jun 2009 18:58:56 +0600 |
| 2 |
Mike Kazantsev <mk.fraggod@×××××.com> wrote: |
| 3 |
|
| 4 |
> On Tue, 16 Jun 2009 12:40:46 +0200 |
| 5 |
> Renat Golubchyk <ragermany@×××.net> wrote: |
| 6 |
> |
| 7 |
> > Hi all! |
| 8 |
> > |
| 9 |
> > Short: |
| 10 |
> > What is the best way to setup something similar to RAID 1 over a |
| 11 |
> > WAN? |
| 12 |
> |
| 13 |
> ... |
| 14 |
> |
| 15 |
> > One purpose of the setup is to have data redundancy. Thus we have to |
| 16 |
> > ensure that the data is replicated in a timely manner. Replicating |
| 17 |
> > MySQL data is not difficult. The problem is the file system data |
| 18 |
> > like uploaded documents and pictures. We can monitor changes in the |
| 19 |
> > file system and initiate rsync to copy files over the network, but |
| 20 |
> > I think it's not a good solution. What we are after is a network |
| 21 |
> > equivalent of RAID 1. Are there any viable solutions that could |
| 22 |
> > work over a WAN? |
| 23 |
> |
| 24 |
> If replication after file is completely hit the disk is not acceptable |
| 25 |
> either because files are quite large or have to be really synchronous |
| 26 |
> then I'll second DRBD-with-HA-cluster-fs suggestion. |
| 27 |
> |
| 28 |
|
| 29 |
DRBD is HA solution which is achieved by switching the role of the |
| 30 |
nodes in case the "active node" goes offline. I think DRBD is not meant |
| 31 |
for the schema OP has described, because only the "active node" is |
| 32 |
accessible via FS. DRBD works between the FS and block device layers. |
| 33 |
It catches the FS writes from the active node and sends them over the |
| 34 |
network. DRBD on the backup node receives those and replicates them |
| 35 |
directly to the disk driver. Thus you can't have mounted FS on the |
| 36 |
backup node. If the active node goes offline, the backup node takes |
| 37 |
over which means DRBD switches roles and the FS has to be mounted |
| 38 |
afterwards. |
| 39 |
|
| 40 |
-- |
| 41 |
Best regards, |
| 42 |
Daniel |
Archives