The best way that I know to protect against hardware failure and data
loss (the rm -rf / instead of rm -rf . type) is to run incremented
backups. This, of course, isn't a live filesystem level backup solution,
but this is what I do:
I have a virtual server. I am concerned about /home and /etc.
On
Disclaimer: I have very little experience in this area, and none with
RAID arrays. Proceed at your own risk.
Usually, I'll use the diagnostic utilities that hard-disk manufacturers
provide to determine hardware-level failures. I'd be very surprised if
they were destructive, but I can't say for
DRBD, disconnected mirrored file system, works really good. Can only view
the mirror after the primary fails though.
Robert
On 10/22/07 12:23 PM, "Alberto Treviño" <[EMAIL PROTECTED]> wrote:
> I have a problem that I would like to solve the cheap way. I'm thinking
> of buying two identical ser
On Monday 22 October 2007 01:20:25 pm Michael L Torrie wrote:
> My only problem with this scheme is that it while it can account for
> a hard drive failure, it can't do anything about my own stupidity, or
> data corruption. If the data on disk 1 is stirred, it's stirred on
> disk 2. This is the ma
Matthew J. Probst wrote [-1 top posting, -1 no trimming]:
>LVM can be run on top of drbd.. so you can have the added protection
>of lvm snapshots to protect from user errors/file system corruption.
Sure. But my point was that, for the reasons I gave, drbd isn't the best
thing for what I w
LVM can be run on top of drbd.. so you can have the added protection
of lvm snapshots to protect from user errors/file system corruption.
-matt
Michael L Torrie wrote:
Matthew J. Probst wrote:
Hi Alberto,
What you are looking for is DRBD ([1]http://www.drbd.org/).
It can do the on
Matthew J. Probst wrote:
> Hi Alberto,
>
> What you are looking for is DRBD (http://www.drbd.org/).
>
> It can do the one way synchronous replication of the block device
> beneath a file system you are looking for. You can also use version 8
> of drbd for bi-directional replication of a block de
I don't know of anything comparable. And I wish I could a afford a
copy myself to see it perform.
Adam
On 10/22/07, Daniel Dilts <[EMAIL PROTECTED]> wrote:
> Yeah, that looks like what I need, however I don't have $89 to purchase it.
> I don't suppose that there is some free alternative, is ther
I've never actually used it, but you might look into DRBD
(http://www.drbd.org/). My understanding is that it mirrors over the
network at the block device level. I've been told, though, that you
have to be careful, especially when the amount of I/O ops is pretty high.
Again, I'm completely in
Hi Alberto,
What you are looking for is DRBD (http://www.drbd.org/).
It can do the one way synchronous replication of the block device
beneath a file system you are looking for. You can also use version 8
of drbd for bi-directional replication of a block device below a
clustered file system (suc
I have a problem that I would like to solve the cheap way. I'm thinking
of buying two identical servers and virtualize all my servers. I
looked into buying two servers with no storage and a storage server,
but that is too expensive. It is cheaper to buy two servers with
storage on each one.
Yeah, that looks like what I need, however I don't have $89 to purchase it.
I don't suppose that there is some free alternative, is there?
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
Behalf Of Adam Findley
Sent: Monday, October 22, 2007 2:14 AM
To: BYU Unix User
Hands down, I think this is the best program out there. It's not
free, but it sure is thorough and non-destructive:
http://www.grc.com/sr/spinrite.htm
It's saved more people's drives than probably any other thing out there.
Adam
On 10/21/07, Daniel Dilts <[EMAIL PROTECTED]> wrote:
> I have 2x
13 matches
Mail list logo
