[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

RE: [linux-lvm] snapshot questions



I will try to quicky summarize what _I_ do:

- Yes snapshots work.
- No, it's not a 1 to 1 disk usage ratio between the vg and the snapshot
vol.  It's much less.  I dont know how to test for fullness using
Linux/LVM yet.
- I use Oracle, your DB may be different.
1) put the tablespaces in hot backup mode
2) take a snapshot (mount the snapshot vol) of the volume/s that the
tablespace is on
3) take the db out of hot backup mode.
- I also do some checkpointing, etc in my script.
- the tape drive can then get the backup from the snapshot.
- You would restore the image from tape, to a new or the orig. volume
when you recover.
- no noticable effect on end users, it's a 24/7 thing.  Restores work
fine.

It's really not all that scary. Sorry for not going into more technical
detail.

Kenny Gorman
DBA
Kenamea


-----Original Message-----
From:	Kyle Hayes
Sent:	Fri 11/2/2001 1:41 PM
To:	linux-lvm sistina com
Cc:	
Subject:	Re: [linux-lvm] snapshot questions


Hmm, seems to be quite a range of opinion.  However, I must say that
having so many negative options is not a good sign.  Do snapshots
work at all?

If I get a corrupted filesystem in my snapshot, then it doesn't work.

If I have to bend over backwards and change all my apps, then it does
not work.

It is good to hear that it works more like Veritas and less like the
SuSE whitepaper claims.  That paper claimed that the changes will go
to the snapshot partition.  This implied that there would be a terrible
IO hit when you deleted the snapshot partition and the changes need to
be written back to the original partition.

I am confused by the claim (see below) that you need the same amount
of space for the snapshot partition as you had on the original LV.
>From what I understood in the HOWTOs and the whitepapers, you only need
as much space as you will have changes during the lifetime of the
snapshot LV.  And, if it works like Veritas, that implies that once
a PE was copied, any further data changes to it would not be copied.
This means that in situations like database systems, you'd actually
need less than the number of PEs that changed.  Is this correct?

Under what circumstances do snapshots work?  I am confused by the
claims that a snapshot partition is going to be "corrupt" as far as
the OS is concerned.  In my scenario, here is what I am planning on
doing:

1) lock all tables against writes.

2) flush all logs and tables to disk.

3) create a snapshot partition for the filesystem (ext2 or probably)
on which the database data resides.  These are not raw disks.

4) unlock the tables.

5) mount the snapshot partition and copy the database data out.


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]