Discussion:
ZFS backup
(too old to reply)
Russell Aspinwall
2007-06-21 11:44:09 UTC
Permalink
Hi,

There is a plan to use ZFS to provide management of approx 5TB of disk
space.
I have been looking at the ZFS manual and found many interesting items.
The ZFS snapshot is very interesting allow for hot backups.
Is there an equivalent of ufsdump/ufsrestore for ZFS for backup to tape?
Data restore tends to be two different types :-
1) Selective file or directory restore
2) File system restore due to hardware failure
( 3 disk failed in a 13 disk RAID5 set + HS)

I like the ufsrestore due to the interactive mode allowing for complete
or selective restores.
--
Regards

Russell





Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
Al Hopper
2007-06-21 13:38:43 UTC
Permalink
Post by Russell Aspinwall
Hi,
There is a plan to use ZFS to provide management of approx 5TB of disk
space.
I have been looking at the ZFS manual and found many interesting items.
The ZFS snapshot is very interesting allow for hot backups.
Is there an equivalent of ufsdump/ufsrestore for ZFS for backup to tape?
No, unfortunately. Of course you can always do a tar - assuming that
you are OK with the zfs extended ACLs not being saved. The only ZFS
mechanism available right now are "zfs send" and "zfs receive".
Post by Russell Aspinwall
Data restore tends to be two different types :-
1) Selective file or directory restore
2) File system restore due to hardware failure
( 3 disk failed in a 13 disk RAID5 set + HS)
I like the ufsrestore due to the interactive mode allowing for complete
or selective restores.
To get the equivalent functionality you have to do something like:

a) re-arrange your data so that, instead of using sub directories to
organize your data, you use filesystems. For example:

existing data:
/bigRaid5/departmentA
/bigRaid5/departmentB
/bigRaid5/departmentC

new organization:

zfs create tank/departmentA
zfs create tank/departmentB
zfs create tank/departmentC
zfs set mountpoint=/deptC tank/departmentC
zfs create tank/departmentC/backups
zfs create tank/departmentC/projectA

b) you make snapshots of your individual filesystems (or recursive
snapshots for home directories etc) and save your "zfs send"
datastream (of these snapshots) to tape.

c) you have another ZFS pool available large enough to restore your
largest zfs filesystem

d) to do selective restores you "render" the 'zfs send' stream onto
your 2nd ZFS pool and get the subset of data you need from it.

You'll probably end up writing some helper shell scripts - search
around on blogs.sun.com for existing scripts that you can probably
adapt.

Another point: Ask yourself, with 5TB of data, do you really want to
mess with tapes - or build another ZFS box as the backup/restore
facility. Don't forget that you can turn on compression for zfs
filesystems.

Moving to ZFS requires that you re-think how you use storage. But
it's well worth the effort IMHO.

Feel free to email me offline if I can help.

Regards,

Al Hopper Logical Approach Inc, Plano, TX. ***@logical-approach.com
Voice: 972.379.2133 Fax: 972.379.2134 Timezone: US CDT
OpenSolaris Governing Board (OGB) Member - Apr 2005 to Mar 2007
http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/


Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
John Martinez
2007-06-22 20:35:34 UTC
Permalink
Post by Al Hopper
No, unfortunately. Of course you can always do a tar - assuming that
you are OK with the zfs extended ACLs not being saved. The only ZFS
mechanism available right now are "zfs send" and "zfs receive".
Interesting topic. I'll bite.

Using tar for large amounts of data is unrealistic. Many of us in
"enterprise" situations already have a backup product in-house. These
solutions are run by another group of people who have invested
millions into not only software, but also hardware solutions such as
tape libraries as well as disk-based backup solutions.
Post by Al Hopper
Post by Russell Aspinwall
Data restore tends to be two different types :-
1) Selective file or directory restore
2) File system restore due to hardware failure
( 3 disk failed in a 13 disk RAID5 set + HS)
I like the ufsrestore due to the interactive mode allowing for complete
or selective restores.
a) re-arrange your data so that, instead of using sub directories to
...
I like this solution, which is why I am looking into migrating to
ZFS. But I hear Russell's request above. It is essential to have the
functionality to restore sections of file systems. I would have to
restore a user's entire file system just because he mistakingly blew
away one little directory in his file system.
Post by Al Hopper
Another point: Ask yourself, with 5TB of data, do you really want to
mess with tapes - or build another ZFS box as the backup/restore
facility. Don't forget that you can turn on compression for zfs
filesystems.
Why yes. We have strict requirements for not only migrating data to
tape for off-site storage, we also move data around to other
geographic locations via the backup software. We are well into the
hundreds of terabytes.

We also have situations where we do off-host backups for database
applications. We freeze the database on the production server, take a
snapshot on the SAN storage array, unfreeze the database, mount the
file systems on a different host, create a database checkpoint and
back it up. This results in about <1 minute interruption for our
users, where it could take hours otherwise. I have ideas on how ZFS
would work in this scenario, and I don't think it would be different
from how our current solution works.
Post by Al Hopper
Moving to ZFS requires that you re-think how you use storage. But
it's well worth the effort IMHO.
Indeed it does. I think it still has a bit to mature for "enterprise"
applications, but it's getting there.
Post by Al Hopper
Feel free to email me offline if I can help.
Thanks for your insight, Al. Always helpful.

Does anybody know if ZFS works with the following list?

Veritas NetBackup
Legato Networker
IBM Tivoli backup
CommVault Galaxy

Maybe this is a good discussion for zfs-***@opensolaris.org?

-john


Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
Laurent Blume
2007-06-22 20:51:39 UTC
Permalink
Post by John Martinez
Veritas NetBackup
I heard 2 days ago from a Sun OS Embassador that this one was working.
Also Atempo Time Navigator, with the potential issue that ACL can't be
restored. I'm investigating that one further, since that's the one we're
using.
Atempo themselves said they support ZFS on "version 4.1, on Solaris 10
64 bit SPARC", which is not too clear since 4.1 also works on S10 AMD64.

Laurent
--
/ Leader de Projet & Communauté | I'm working, but not speaking for
\ G11N http://fr.opensolaris.org | Bull Services http://www.bull.com
/ FOSUG http://guses.org |


Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
Al Hopper
2007-06-22 22:14:17 UTC
Permalink
Post by John Martinez
Post by Al Hopper
No, unfortunately. Of course you can always do a tar - assuming that
you are OK with the zfs extended ACLs not being saved. The only ZFS
mechanism available right now are "zfs send" and "zfs receive".
Interesting topic. I'll bite.
Using tar for large amounts of data is unrealistic. Many of us in
"enterprise" situations already have a backup product in-house. These
solutions are run by another group of people who have invested
millions into not only software, but also hardware solutions such as
tape libraries as well as disk-based backup solutions.
Post by Al Hopper
Post by Russell Aspinwall
Data restore tends to be two different types :-
1) Selective file or directory restore
2) File system restore due to hardware failure
( 3 disk failed in a 13 disk RAID5 set + HS)
I like the ufsrestore due to the interactive mode allowing for complete
or selective restores.
a) re-arrange your data so that, instead of using sub directories to
...
I like this solution, which is why I am looking into migrating to
ZFS. But I hear Russell's request above. It is essential to have the
functionality to restore sections of file systems. I would have to
restore a user's entire file system just because he mistakingly blew
away one little directory in his file system.
Post by Al Hopper
Another point: Ask yourself, with 5TB of data, do you really want to
mess with tapes - or build another ZFS box as the backup/restore
facility. Don't forget that you can turn on compression for zfs
filesystems.
Why yes. We have strict requirements for not only migrating data to
tape for off-site storage, we also move data around to other
geographic locations via the backup software. We are well into the
hundreds of terabytes.
Understood - but my point is that for smaller datasets, in Russells'
case, 5 terabytes, ZFS, in combination with cost effective, high
capacity SATA disk drives, may very well have already obsoleted tape
based backups. Obviously, there is some user-scenario-specific
storage capacity (translation number-of-terabytes) where tape will be
the only economically viable backup storage medium currently
available.

The other point I did'nt make, is that, in many situations,
time-to-restore is much more important than the cost of the backup
subsystem. I know many people on this list have wasted years (OK
exaggeration) of their lives, waiting for a tape to "find" the
required dataset, before it even *starts* to stream off the required
data.

Everyone goes down at some time or another - for various reasons.
And downtime, while becoming increasingly untolerated by a typical
user community, can be explained and is understood and (somewhat)
accepted. But there is far less user community tolerance when it
takes, for example, a business day to restore a service[1]. The user
community expects the service provider to restore a service in a
matter of "N" hours - where N is often (in todays environment) less
than 2. Down the road, it is easy to imaging that N will continue to
shrink.
Post by John Martinez
We also have situations where we do off-host backups for database
applications. We freeze the database on the production server, take a
snapshot on the SAN storage array, unfreeze the database, mount the
file systems on a different host, create a database checkpoint and
back it up. This results in about <1 minute interruption for our
users, where it could take hours otherwise. I have ideas on how ZFS
would work in this scenario, and I don't think it would be different
from how our current solution works.
Post by Al Hopper
Moving to ZFS requires that you re-think how you use storage. But
it's well worth the effort IMHO.
Indeed it does. I think it still has a bit to mature for "enterprise"
applications, but it's getting there.
100% agreed. On zfs-discuss you can see the long running thread where
users have been asking for the ability to shrink a pool - for various
reasons. In one scenario, it's because they need to reallocate
physical disk drives due to changing business requirements, in other
scenarios, it's because rocket-scientist-turned-manager changes
his/her mind, after making a really bad decision, and needs to
reallocate physical disk resources. Or the unfortunate sysAdmin makes
a couple of "typos".... :)

There were also a few posts in the early days of zfs-discuss where
people thought that ZFS should never have been released before a
comprehensive and fully featured backup/restore feature set was
available. Personally, I'm glad those people did'nt get to stall the
release of ZFS! :)

I think its very healthy to *always* list the current deficits of a
product or technology - rather than sending the message that
technology "A" is the greatest thing since sliced bread - and then
many of the early adopters discover, to their frustration, the
limitations already known by the developers/proponents of it.

The really good news, is that team ZFS knows all this. And the
underlying technology has already been designed in such a way that the
currently missing functionality can be added with minimal disruption
to the zfs user community. So, your point that ZFS is still shy of a
fully featured enterprise storage subsystem is accurate - but there is
so much functionality and features currently available that it would
be shameful to not deploy it today, where it fits, and simply work
around the currently missing features/facilities.

In Russells scenario, he could build a backup server, with, say 4
terrabytes of storage for a reasonable cost. Store his snapshots
there, and be able to "render" a zfs filesystem to the same backup
server storage pool very, very quickly (at 400 to 500Mb/Sec) and
restore a specific user subdirectory structure in minutes. Way before
a tape based system would be able to even stream to the required
dataset!

[1] can anyone say docs.sun.com (or 24 hours)
Post by John Martinez
Post by Al Hopper
Feel free to email me offline if I can help.
Thanks for your insight, Al. Always helpful.
Does anybody know if ZFS works with the following list?
Veritas NetBackup
Legato Networker
IBM Tivoli backup
CommVault Galaxy
Yep - please subscribe to zfs-discuss. I've been on this list from
day-one!

Regards,

Al Hopper Logical Approach Inc, Plano, TX. ***@logical-approach.com
Voice: 972.379.2133 Fax: 972.379.2134 Timezone: US CDT
OpenSolaris Governing Board (OGB) Member - Apr 2005 to Mar 2007
http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/


Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
Ian Collins
2007-06-22 23:15:44 UTC
Permalink
Post by Al Hopper
In Russells scenario, he could build a backup server, with, say 4
terrabytes of storage for a reasonable cost. Store his snapshots
there, and be able to "render" a zfs filesystem to the same backup
server storage pool very, very quickly (at 400 to 500Mb/Sec) and
restore a specific user subdirectory structure in minutes. Way before
a tape based system would be able to even stream to the required
dataset!
I agree 100% with this, having suffered form the wait x hours for files
to come back from tape I always, both for my systems and client's, adopt
a two tier backup strategy. The first tier is in place to enable rapid
recovery of data, the second for archive and disaster recovery. I add
enough storage to hold backups to a second machine on the network, dump
onto that box and then archive those dumps to tape or DVD. ZFS makes an
ideal filesystem for the second box, it is the only file system I'd
trust for temporary backups on a single drive or a simple stripe.

Ian



Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
Anderson, Ryan C (US SSA)
2007-06-21 13:45:08 UTC
Permalink
You'd get a better response from zfs-discuss
(http://mail.opensolaris.org/mailman/listinfo/zfs-discuss). I can say
this though: You cannot do selective file or directory restores, though
restoring entire filesystems is very quick and efficient, eg:




zfs snapshot mypool/Foo01/***@today
rm -rf /export/Foo01/bar01/*
zfs rollback mypool/Foo01/***@today



You can do multiple snapshots without using a lot of disk space though.



RCA

--

UNIX Administrator, BAE Systems Land & Armaments
desk 763-572-6684 mobile 612-419-9362



________________________________

From: ***@yahoogroups.com [mailto:***@yahoogroups.com] On
Behalf Of Russell Aspinwall
Sent: Thursday, June 21, 2007 6:44 AM
To: ***@yahoogroups.com
Subject: [solarisx86] ZFS backup



Hi,

There is a plan to use ZFS to provide management of approx 5TB of disk
space.
I have been looking at the ZFS manual and found many interesting items.
The ZFS snapshot is very interesting allow for hot backups.
Is there an equivalent of ufsdump/ufsrestore for ZFS for backup to tape?
Data restore tends to be two different types :-
1) Selective file or directory restore
2) File system restore due to hardware failure
( 3 disk failed in a 13 disk RAID5 set + HS)

I like the ufsrestore due to the interactive mode allowing for complete
or selective restores.
--
Regards

Russell





[Non-text portions of this message have been removed]



Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
Russell Aspinwall
2007-06-21 13:54:41 UTC
Permalink
Hi Al,

Backup regime requires bi-monthly backups off site.
Not all of the 5TB is backed up, as the it can be regenerated,
the core is only about 1.5TB split across two servers.
It is a pity there is no "zfs ireceive" to allow interactive restoration.
Post by Russell Aspinwall
Hi,
There is a plan to use ZFS to provide management of approx 5TB of disk
space.
I have been looking at the ZFS manual and found many interesting items.
The ZFS snapshot is very interesting allow for hot backups.
Is there an equivalent of ufsdump/ufsrestore for ZFS for backup to tape?
No, unfortunately. Of course you can always do a tar - assuming that
you are OK with the zfs extended ACLs not being saved. The only ZFS
mechanism available right now are "zfs send" and "zfs receive".
Post by Russell Aspinwall
Data restore tends to be two different types :-
1) Selective file or directory restore
2) File system restore due to hardware failure
( 3 disk failed in a 13 disk RAID5 set + HS)
I like the ufsrestore due to the interactive mode allowing for complete
or selective restores.
a) re-arrange your data so that, instead of using sub directories to
/bigRaid5/departmentA
/bigRaid5/departmentB
/bigRaid5/departmentC
zfs create tank/departmentA
zfs create tank/departmentB
zfs create tank/departmentC
zfs set mountpoint=/deptC tank/departmentC
zfs create tank/departmentC/backups
zfs create tank/departmentC/projectA
b) you make snapshots of your individual filesystems (or recursive
snapshots for home directories etc) and save your "zfs send"
datastream (of these snapshots) to tape.
c) you have another ZFS pool available large enough to restore your
largest zfs filesystem
d) to do selective restores you "render" the 'zfs send' stream onto
your 2nd ZFS pool and get the subset of data you need from it.
You'll probably end up writing some helper shell scripts - search
around on blogs.sun.com for existing scripts that you can probably
adapt.
Another point: Ask yourself, with 5TB of data, do you really want to
mess with tapes - or build another ZFS box as the backup/restore
facility. Don't forget that you can turn on compression for zfs
filesystems.
Moving to ZFS requires that you re-think how you use storage. But
it's well worth the effort IMHO.
Feel free to email me offline if I can help.
Regards,
<mailto:al%40logical-approach.com>
Voice: 972.379.2133 Fax: 972.379.2134 Timezone: US CDT
OpenSolaris Governing Board (OGB) Member - Apr 2005 to Mar 2007
http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/
<http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/>
______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
______________________________________________________________________
--
Regards

Russell

Email: russell dot aspinwall at flomerics dot co dot uk
Network and Systems Administrator Flomerics Ltd
Telephone: 020-8941-8810 x3116 81 Bridge Road
Facsimile: 020-8941-8730 Hampton Court
Surrey, KT8 9HH
United Kingdom



Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
Adrian Saul
2007-06-21 14:16:38 UTC
Permalink
You could use Legato Networker 7.3.2 or one of the other commercial
backup suites that supports ZFS. At that size, its probably more
suitable for what you want, even a bare minimum legato setup.

We use Legato 7.3.2 on our few boxes with ZFs and aside from some really
baffling deficencies for ZFS, it works quite well. It wont backup
snapshots etc, but pretty much any mounted filesystem should be able to
backed and restored. The restores are interactive so you can pick and
choose what you want restored.

Your main issue is Legato wont do a saveset "ALL", but its not hard to
work around that (dummy entries in /etc/vfstab populated by a script, or
specify the filesystems to save in the client saveset list).
Post by Russell Aspinwall
Hi Al,
Backup regime requires bi-monthly backups off site.
Not all of the 5TB is backed up, as the it can be regenerated,
the core is only about 1.5TB split across two servers.
It is a pity there is no "zfs ireceive" to allow interactive restoration.
Post by Russell Aspinwall
Hi,
There is a plan to use ZFS to provide management of approx 5TB of disk
space.
I have been looking at the ZFS manual and found many interesting items.
The ZFS snapshot is very interesting allow for hot backups.
Is there an equivalent of ufsdump/ufsrestore for ZFS for backup to
tape?
No, unfortunately. Of course you can always do a tar - assuming that
you are OK with the zfs extended ACLs not being saved. The only ZFS
mechanism available right now are "zfs send" and "zfs receive".
Post by Russell Aspinwall
Data restore tends to be two different types :-
1) Selective file or directory restore
2) File system restore due to hardware failure
( 3 disk failed in a 13 disk RAID5 set + HS)
I like the ufsrestore due to the interactive mode allowing for complete
or selective restores.
a) re-arrange your data so that, instead of using sub directories to
/bigRaid5/departmentA
/bigRaid5/departmentB
/bigRaid5/departmentC
zfs create tank/departmentA
zfs create tank/departmentB
zfs create tank/departmentC
zfs set mountpoint=/deptC tank/departmentC
zfs create tank/departmentC/backups
zfs create tank/departmentC/projectA
b) you make snapshots of your individual filesystems (or recursive
snapshots for home directories etc) and save your "zfs send"
datastream (of these snapshots) to tape.
c) you have another ZFS pool available large enough to restore your
largest zfs filesystem
d) to do selective restores you "render" the 'zfs send' stream onto
your 2nd ZFS pool and get the subset of data you need from it.
You'll probably end up writing some helper shell scripts - search
around on blogs.sun.com for existing scripts that you can probably
adapt.
Another point: Ask yourself, with 5TB of data, do you really want to
mess with tapes - or build another ZFS box as the backup/restore
facility. Don't forget that you can turn on compression for zfs
filesystems.
Moving to ZFS requires that you re-think how you use storage. But
it's well worth the effort IMHO.
Feel free to email me offline if I can help.
Regards,
<mailto:al%40logical-approach.com>
<mailto:al%40logical-approach.com>
Voice: 972.379.2133 Fax: 972.379.2134 Timezone: US CDT
OpenSolaris Governing Board (OGB) Member - Apr 2005 to Mar 2007
http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/
<http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/>
<http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/
<http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/>>
__________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
<http://www.messagelabs.com/email>
__________________________________________________________
--
Regards
Russell
Email: russell dot aspinwall at flomerics dot co dot uk
Network and Systems Administrator Flomerics Ltd
Telephone: 020-8941-8810 x3116 81 Bridge Road
Facsimile: 020-8941-8730 Hampton Court
Surrey, KT8 9HH
United Kingdom
Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
Anderson, Ryan C (US SSA)
2007-06-22 18:15:39 UTC
Permalink
This will be available in a future Solaris release (update 5 maybe?),
something like ufsdump/ufsrestore, but for zfs instead. Given your
requirements, a zfs function does not exist, but an enterprise-level
backup/restore utility would be better.



RCA

--

UNIX Administrator, BAE Systems Land & Armaments
desk 763-572-6684 mobile 612-419-9362



________________________________

From: ***@yahoogroups.com [mailto:***@yahoogroups.com] On
Behalf Of Russell Aspinwall
Sent: Thursday, June 21, 2007 8:55 AM
To: ***@yahoogroups.com
Subject: Re: [solarisx86] ZFS backup



Hi Al,

Backup regime requires bi-monthly backups off site.
Not all of the 5TB is backed up, as the it can be regenerated,
the core is only about 1.5TB split across two servers.
It is a pity there is no "zfs ireceive" to allow interactive
restoration.
Post by Russell Aspinwall
Hi,
There is a plan to use ZFS to provide management of approx 5TB of disk
space.
I have been looking at the ZFS manual and found many interesting items.
The ZFS snapshot is very interesting allow for hot backups.
Is there an equivalent of ufsdump/ufsrestore for ZFS for backup to tape?
No, unfortunately. Of course you can always do a tar - assuming that
you are OK with the zfs extended ACLs not being saved. The only ZFS
mechanism available right now are "zfs send" and "zfs receive".
Post by Russell Aspinwall
Data restore tends to be two different types :-
1) Selective file or directory restore
2) File system restore due to hardware failure
( 3 disk failed in a 13 disk RAID5 set + HS)
I like the ufsrestore due to the interactive mode allowing for complete
or selective restores.
a) re-arrange your data so that, instead of using sub directories to
/bigRaid5/departmentA
/bigRaid5/departmentB
/bigRaid5/departmentC
zfs create tank/departmentA
zfs create tank/departmentB
zfs create tank/departmentC
zfs set mountpoint=/deptC tank/departmentC
zfs create tank/departmentC/backups
zfs create tank/departmentC/projectA
b) you make snapshots of your individual filesystems (or recursive
snapshots for home directories etc) and save your "zfs send"
datastream (of these snapshots) to tape.
c) you have another ZFS pool available large enough to restore your
largest zfs filesystem
d) to do selective restores you "render" the 'zfs send' stream onto
your 2nd ZFS pool and get the subset of data you need from it.
You'll probably end up writing some helper shell scripts - search
around on blogs.sun.com for existing scripts that you can probably
adapt.
Another point: Ask yourself, with 5TB of data, do you really want to
mess with tapes - or build another ZFS box as the backup/restore
facility. Don't forget that you can turn on compression for zfs
filesystems.
Moving to ZFS requires that you re-think how you use storage. But
it's well worth the effort IMHO.
Feel free to email me offline if I can help.
Regards,
<mailto:al%40logical-approach.com>
<mailto:al%40logical-approach.com>
Voice: 972.379.2133 Fax: 972.379.2134 Timezone: US CDT
OpenSolaris Governing Board (OGB) Member - Apr 2005 to Mar 2007
http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/
<http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/>
<http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/
<http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/> >
__________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
<http://www.messagelabs.com/email>
__________________________________________________________
--
Regards

Russell

Email: russell dot aspinwall at flomerics dot co dot uk
Network and Systems Administrator Flomerics Ltd
Telephone: 020-8941-8810 x3116 81 Bridge Road
Facsimile: 020-8941-8730 Hampton Court
Surrey, KT8 9HH
United Kingdom





[Non-text portions of this message have been removed]



Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
Russell Aspinwall
2007-06-22 07:09:34 UTC
Permalink
Used Legato Networker about 10 years ago for about 2 months.
Switched back to ufsdump.
Post by Adrian Saul
You could use Legato Networker 7.3.2 or one of the other commercial
backup suites that supports ZFS. At that size, its probably more
suitable for what you want, even a bare minimum legato setup.
We use Legato 7.3.2 on our few boxes with ZFs and aside from some really
baffling deficencies for ZFS, it works quite well. It wont backup
snapshots etc, but pretty much any mounted filesystem should be able to
backed and restored. The restores are interactive so you can pick and
choose what you want restored.
Your main issue is Legato wont do a saveset "ALL", but its not hard to
work around that (dummy entries in /etc/vfstab populated by a script, or
specify the filesystems to save in the client saveset list).
Post by Russell Aspinwall
Hi Al,
Backup regime requires bi-monthly backups off site.
Not all of the 5TB is backed up, as the it can be regenerated,
the core is only about 1.5TB split across two servers.
It is a pity there is no "zfs ireceive" to allow interactive
restoration.
Post by Russell Aspinwall
Post by Russell Aspinwall
Hi,
There is a plan to use ZFS to provide management of approx 5TB
of disk
Post by Russell Aspinwall
Post by Russell Aspinwall
space.
I have been looking at the ZFS manual and found many interesting
items.
Post by Russell Aspinwall
Post by Russell Aspinwall
The ZFS snapshot is very interesting allow for hot backups.
Is there an equivalent of ufsdump/ufsrestore for ZFS for backup to
tape?
No, unfortunately. Of course you can always do a tar - assuming that
you are OK with the zfs extended ACLs not being saved. The only ZFS
mechanism available right now are "zfs send" and "zfs receive".
Post by Russell Aspinwall
Data restore tends to be two different types :-
1) Selective file or directory restore
2) File system restore due to hardware failure
( 3 disk failed in a 13 disk RAID5 set + HS)
I like the ufsrestore due to the interactive mode allowing for
complete
Post by Russell Aspinwall
Post by Russell Aspinwall
or selective restores.
a) re-arrange your data so that, instead of using sub directories to
/bigRaid5/departmentA
/bigRaid5/departmentB
/bigRaid5/departmentC
zfs create tank/departmentA
zfs create tank/departmentB
zfs create tank/departmentC
zfs set mountpoint=/deptC tank/departmentC
zfs create tank/departmentC/backups
zfs create tank/departmentC/projectA
b) you make snapshots of your individual filesystems (or recursive
snapshots for home directories etc) and save your "zfs send"
datastream (of these snapshots) to tape.
c) you have another ZFS pool available large enough to restore your
largest zfs filesystem
d) to do selective restores you "render" the 'zfs send' stream onto
your 2nd ZFS pool and get the subset of data you need from it.
You'll probably end up writing some helper shell scripts - search
around on blogs.sun.com for existing scripts that you can probably
adapt.
Another point: Ask yourself, with 5TB of data, do you really want to
mess with tapes - or build another ZFS box as the backup/restore
facility. Don't forget that you can turn on compression for zfs
filesystems.
Moving to ZFS requires that you re-think how you use storage. But
it's well worth the effort IMHO.
Feel free to email me offline if I can help.
Regards,
<mailto:al%40logical-approach.com>
Post by Russell Aspinwall
<mailto:al%40logical-approach.com>
<mailto:al%40logical-approach.com>
Voice: 972.379.2133 Fax: 972.379.2134 Timezone: US CDT
OpenSolaris Governing Board (OGB) Member - Apr 2005 to Mar 2007
http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/
<http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/>
Post by Russell Aspinwall
<http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/
<http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/>>
Post by Russell Aspinwall
<http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/
<http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/>
Post by Russell Aspinwall
<http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/
<http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/>>>
Post by Russell Aspinwall
__________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
<http://www.messagelabs.com/email>
Post by Russell Aspinwall
<http://www.messagelabs.com/email <http://www.messagelabs.com/email>>
__________________________________________________________
--
Regards
Russell
Email: russell dot aspinwall at flomerics dot co dot uk
Network and Systems Administrator Flomerics Ltd
Telephone: 020-8941-8810 x3116 81 Bridge Road
Facsimile: 020-8941-8730 Hampton Court
Surrey, KT8 9HH
United Kingdom
______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
______________________________________________________________________
--
Regards

Russell

Email: russell dot aspinwall at flomerics dot co dot uk
Network and Systems Administrator Flomerics Ltd
Telephone: 020-8941-8810 x3116 81 Bridge Road
Facsimile: 020-8941-8730 Hampton Court
Surrey, KT8 9HH
United Kingdom



Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
Russell Aspinwall
2007-06-25 08:01:04 UTC
Permalink
Hi,

Thanks for the feedback, I did subscribe to the zfs-discuss list.
As I previously posted while having 5TB of disk, the actual
amount for backup is 600MB on Server A and 900MB on Server B.
Both servers have tape stackers.
The backup policy I use to perform a level 0 dump every night, as this
significantly
speeds up data restoration. I have found that I can restore any
directory or file within
2hours. I would expect with the 5TB disk subsystem this may increase to
4 hours
in some cases due to size of some directories which is acceptable to my
users.
The company mandates off site storage, so backing up to disk is
pointless unless it is off
site. The cost of implementing a high speed data connection 150 miles
to another office
would be very expensive.
I did receive a post the ufsrestore functionality may be implemented in
a later release of
Solaris U5
Post by Al Hopper
Post by John Martinez
No, unfortunately. Of course you can always do a tar - assuming that
you are OK with the zfs extended ACLs not being saved. The only ZFS
mechanism available right now are "zfs send" and "zfs receive".
Interesting topic. I'll bite.
Using tar for large amounts of data is unrealistic. Many of us in
"enterprise" situations already have a backup product in-house. These
solutions are run by another group of people who have invested
millions into not only software, but also hardware solutions such as
tape libraries as well as disk-based backup solutions.
Post by Russell Aspinwall
Data restore tends to be two different types :-
1) Selective file or directory restore
2) File system restore due to hardware failure
( 3 disk failed in a 13 disk RAID5 set + HS)
I like the ufsrestore due to the interactive mode allowing for complete
or selective restores.
a) re-arrange your data so that, instead of using sub directories to
...
I like this solution, which is why I am looking into migrating to
ZFS. But I hear Russell's request above. It is essential to have the
functionality to restore sections of file systems. I would have to
restore a user's entire file system just because he mistakingly blew
away one little directory in his file system.
Another point: Ask yourself, with 5TB of data, do you really want to
mess with tapes - or build another ZFS box as the backup/restore
facility. Don't forget that you can turn on compression for zfs
filesystems.
Why yes. We have strict requirements for not only migrating data to
tape for off-site storage, we also move data around to other
geographic locations via the backup software. We are well into the
hundreds of terabytes.
Understood - but my point is that for smaller datasets, in Russells'
case, 5 terabytes, ZFS, in combination with cost effective, high
capacity SATA disk drives, may very well have already obsoleted tape
based backups. Obviously, there is some user-scenario-specific
storage capacity (translation number-of-terabytes) where tape will be
the only economically viable backup storage medium currently
available.
The other point I did'nt make, is that, in many situations,
time-to-restore is much more important than the cost of the backup
subsystem. I know many people on this list have wasted years (OK
exaggeration) of their lives, waiting for a tape to "find" the
required dataset, before it even *starts* to stream off the required
data.
Everyone goes down at some time or another - for various reasons.
And downtime, while becoming increasingly untolerated by a typical
user community, can be explained and is understood and (somewhat)
accepted. But there is far less user community tolerance when it
takes, for example, a business day to restore a service[1]. The user
community expects the service provider to restore a service in a
matter of "N" hours - where N is often (in todays environment) less
than 2. Down the road, it is easy to imaging that N will continue to
shrink.
Post by John Martinez
We also have situations where we do off-host backups for database
applications. We freeze the database on the production server, take a
snapshot on the SAN storage array, unfreeze the database, mount the
file systems on a different host, create a database checkpoint and
back it up. This results in about <1 minute interruption for our
users, where it could take hours otherwise. I have ideas on how ZFS
would work in this scenario, and I don't think it would be different
from how our current solution works.
Moving to ZFS requires that you re-think how you use storage. But
it's well worth the effort IMHO.
Indeed it does. I think it still has a bit to mature for "enterprise"
applications, but it's getting there.
100% agreed. On zfs-discuss you can see the long running thread where
users have been asking for the ability to shrink a pool - for various
reasons. In one scenario, it's because they need to reallocate
physical disk drives due to changing business requirements, in other
scenarios, it's because rocket-scientist-turned-manager changes
his/her mind, after making a really bad decision, and needs to
reallocate physical disk resources. Or the unfortunate sysAdmin makes
a couple of "typos".... :)
There were also a few posts in the early days of zfs-discuss where
people thought that ZFS should never have been released before a
comprehensive and fully featured backup/restore feature set was
available. Personally, I'm glad those people did'nt get to stall the
release of ZFS! :)
I think its very healthy to *always* list the current deficits of a
product or technology - rather than sending the message that
technology "A" is the greatest thing since sliced bread - and then
many of the early adopters discover, to their frustration, the
limitations already known by the developers/proponents of it.
The really good news, is that team ZFS knows all this. And the
underlying technology has already been designed in such a way that the
currently missing functionality can be added with minimal disruption
to the zfs user community. So, your point that ZFS is still shy of a
fully featured enterprise storage subsystem is accurate - but there is
so much functionality and features currently available that it would
be shameful to not deploy it today, where it fits, and simply work
around the currently missing features/facilities.
In Russells scenario, he could build a backup server, with, say 4
terrabytes of storage for a reasonable cost. Store his snapshots
there, and be able to "render" a zfs filesystem to the same backup
server storage pool very, very quickly (at 400 to 500Mb/Sec) and
restore a specific user subdirectory structure in minutes. Way before
a tape based system would be able to even stream to the required
dataset!
[1] can anyone say docs.sun.com (or 24 hours)
Post by John Martinez
Feel free to email me offline if I can help.
Thanks for your insight, Al. Always helpful.
Does anybody know if ZFS works with the following list?
Veritas NetBackup
Legato Networker
IBM Tivoli backup
CommVault Galaxy
<mailto:zfs-discuss%40opensolaris.org>?
Yep - please subscribe to zfs-discuss. I've been on this list from
day-one!
Regards,
<mailto:al%40logical-approach.com>
Voice: 972.379.2133 Fax: 972.379.2134 Timezone: US CDT
OpenSolaris Governing Board (OGB) Member - Apr 2005 to Mar 2007
http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/
<http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/>
______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
______________________________________________________________________
--
Regards

Russell

Email: russell dot aspinwall at flomerics dot co dot uk
Network and Systems Administrator Flomerics Ltd
Telephone: 020-8941-8810 x3116 81 Bridge Road
Facsimile: 020-8941-8730 Hampton Court
Surrey, KT8 9HH
United Kingdom



Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
Chad Leigh -- Shire.Net LLC
2007-06-25 09:13:18 UTC
Permalink
Post by Russell Aspinwall
The company mandates off site storage, so backing up to disk is
pointless unless it is off
site.
Well, your tapes are not off site until after the backup us run and
the tapes are fedex-ed or whatever to the offsite location.

So do the same with some 750GB drives. Do your monthly backup to
some drives and ship them offsite.
Post by Russell Aspinwall
I did receive a post the ufsrestore functionality may be
implemented in
a later release of
Solaris U5
I am interested in this myself as I backup to disk but do it using
dump/restore which is very fast to restore specific files from
without having to have large spare disks to hold a 200gb file system
restore.

I know that as I migrate more to ZFS the actual layout/allocation/use
of disks/partitions/filesystems will change and these will be
smaller, but it is still easier for me to restore from a dump
interactively than to restore the whole thing to scratch space
somewhere and then have to search through the results to get the file
and move it back to the live file system. That takes ME more time
and my time is valuable.

Chad


---
Chad Leigh -- Shire.Net LLC
Your Web App and Email hosting provider
chad at shire.net





Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
John Martinez
2007-06-25 15:03:45 UTC
Permalink
Post by Chad Leigh -- Shire.Net LLC
Post by Russell Aspinwall
The company mandates off site storage, so backing up to disk is
pointless unless it is off
site.
Well, your tapes are not off site until after the backup us run and
the tapes are fedex-ed or whatever to the offsite location.
So do the same with some 750GB drives. Do your monthly backup to
some drives and ship them offsite.
What about failure rates of disks vs. tapes? Disks tend to break if
handled incorrectly (dropped, etc.), but tapes can handle a bit more
abuse. I've never worked, or seen, an off-site data warehouse, but my
guess is that not everyone is going to treat the media gingerly.
Post by Chad Leigh -- Shire.Net LLC
Post by Russell Aspinwall
I did receive a post the ufsrestore functionality may be
implemented in
a later release of
Solaris U5
I am interested in this myself as I backup to disk but do it using
dump/restore which is very fast to restore specific files from
without having to have large spare disks to hold a 200gb file system
restore.
I *think* I mentioned it before, but we implement a three tier backup
system. Important data gets backed up to disk *and* to tape
simultaneously. The tapes are sent off-site for safe keeping. Less
important data is backed up to disk initially, then is backed up to
tape for archiving, as a "just in case." Other data is only backed up
to disk and rolls off when the data retention policy for that data
expires it. The idea is that backing up to disk is cheap and fast.
Especially true when you use devices that do compression, etc. Hey,
sounds like a good application for ZFS, don't it? ;-)
Post by Chad Leigh -- Shire.Net LLC
I know that as I migrate more to ZFS the actual layout/allocation/use
of disks/partitions/filesystems will change and these will be
smaller, but it is still easier for me to restore from a dump
interactively than to restore the whole thing to scratch space
somewhere and then have to search through the results to get the file
and move it back to the live file system. That takes ME more time
and my time is valuable.
It's been years since I've used ufs{dump,restore}, but I don't
believe it kept an index, or if it did, at the beginning of the tape,
but that is also an advantage of a commercial backup utility. It
keeps an index of files backed up somewhere in a database or tape,
where it can quickly look up a file or set of files for restore. If
you go disk to disk (regardless of filesystem), using find to get to
the right file can take up a bunch of time, as you say.

Really, I don't concern myself with backups at my current place,
because we have specialists that have it under control. It's just my
background in storage and servers that keeps me in that loop.
Sometimes much closer than I want to be. :-(

I am very interested in learning about people's ZFS implementations,
as I am researching it for implementation myself. It is very
compelling for its features and benefits. I am in no rush, though.

-john


Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
rogerfujii
2007-06-28 16:23:37 UTC
Permalink
Post by John Martinez
What about failure rates of disks vs. tapes? Disks tend to break if
handled incorrectly (dropped, etc.), but tapes can handle a bit more
abuse. I've never worked, or seen, an off-site data warehouse, but my
guess is that not everyone is going to treat the media gingerly.
true, but if you can ups a drive that you bought from
newegg/your_favorite_etailer, I think you can pack something secure
enough that can survive off-site data storage. And it's not like
tapes are the most reliable of media to do long term storage either.
Post by John Martinez
Post by Chad Leigh -- Shire.Net LLC
I am interested in this myself as I backup to disk but do it using
dump/restore which is very fast to restore specific files from
without having to have large spare disks to hold a 200gb file system
restore.
uh, isn't this the whole point of snapshots so you don't have to do this?
Post by John Martinez
I *think* I mentioned it before, but we implement a three tier backup
system. Important data gets backed up to disk *and* to tape
simultaneously. The tapes are sent off-site for safe keeping. Less
important data is backed up to disk initially, then is backed up to
tape for archiving, as a "just in case." Other data is only backed up
to disk and rolls off when the data retention policy for that data
expires it. The idea is that backing up to disk is cheap and fast.
Especially true when you use devices that do compression, etc. Hey,
sounds like a good application for ZFS, don't it? ;-)
would be cool to be able to attach a write-only mirror.
Post by John Martinez
Post by Chad Leigh -- Shire.Net LLC
I know that as I migrate more to ZFS the actual layout/allocation/use
of disks/partitions/filesystems will change and these will be
smaller, but it is still easier for me to restore from a dump
interactively than to restore the whole thing to scratch space
somewhere and then have to search through the results to get the file
and move it back to the live file system. That takes ME more time
and my time is valuable.
... which is why you want to use snapshots. As mentioned somewhere
else, it is really worthwhile to re-think backup strategies with ZFS,
as the cost for certain backup operations (oops recovery) has gone
down to zero.
Post by John Martinez
I am very interested in learning about people's ZFS implementations,
as I am researching it for implementation myself. It is very
compelling for its features and benefits. I am in no rush, though.
Though I'm not sure if this is a good forum to do this, I'm also
curious about what people's perceived weaknesses of zfs are. The only
one I know of is not having a utility to do multi-volume tape backups
to (though one could easily hack ufsdump/ufsrestore to do this with rf
- and xf - ).

-r



Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
pcsol1996
2007-06-28 20:03:28 UTC
Permalink
Post by rogerfujii
Though I'm not sure if this is a good forum to do this, I'm also
Sorry for highjacking the thread somewhat, but do many insist that
questions be directed to Opensolaris discussions. Is this not still a
place to ask any question related to Solaris?

I have visited the Opensolaris lists and a lot of it is that
pseudo-religious license/opensource-mantra Linux-like B.S.

I like this list much better and it would be sad to see it die.









Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
John Martinez
2007-06-29 15:10:17 UTC
Permalink
Post by pcsol1996
Sorry for highjacking the thread somewhat, but do many insist that
questions be directed to Opensolaris discussions. Is this not still a
place to ask any question related to Solaris?
Absolutely. This list has been a great resource over the years, in
it's many forms.
Post by pcsol1996
I have visited the Opensolaris lists and a lot of it is that
pseudo-religious license/opensource-mantra Linux-like B.S.
opensolaris-discuss can be, yes. But that is the general discussion
list. Many a religious flame war on there. The dedicated lists like
zfs-discuss tend to be more technical. There is a mix of developer
type questions/discussions and operational/administration related
questions/discussions.
Post by pcsol1996
I like this list much better and it would be sad to see it die.
I don't think it's going to die, but I think the more resources there
are, the better. Also, if you want to be "in the know" about Solaris
and what's coming (OpenSolaris), joining the other lists is a good
thing. I'm an OpenSolaris pilot member, so I am biased, also I know
there are some others on here who are pilot members or have been
involved in the CAB of OpenSolaris.

-john


Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
Chad Leigh -- Shire.Net LLC
2007-06-29 07:12:36 UTC
Permalink
Post by rogerfujii
Post by Chad Leigh -- Shire.Net LLC
I am interested in this myself as I backup to disk but do it using
dump/restore which is very fast to restore specific files from
without having to have large spare disks to hold a 200gb file system
restore.
uh, isn't this the whole point of snapshots so you don't have to do this?
No

Snapshots still take disk space when the files in the snapshot are
modified. Backups take that off disk

Chad


---
Chad Leigh -- Shire.Net LLC
Your Web App and Email hosting provider
chad at shire.net





Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
rogerfujii
2007-06-29 19:42:27 UTC
Permalink
Post by Chad Leigh -- Shire.Net LLC
Post by rogerfujii
Post by Chad Leigh -- Shire.Net LLC
I am interested in this myself as I backup to disk but do it using
dump/restore which is very fast to restore specific files from
without having to have large spare disks to hold a 200gb file system
restore.
uh, isn't this the whole point of snapshots so you don't have to do this?
No
Snapshots still take disk space when the files in the snapshot are
modified. Backups take that off disk
s/files/pages/ (will make a big difference if you are incrementally
modding a large database file). Granted, YMMV, but it is certainly
arguable that keeping a base snapshot and "zfs send -i" images around
is still better than dealing with ufsdump images. And once sun does
the next obvious zfs extension (a vdev to store only snapshot data),
it becomes a no-brainer.

Personally, I think that non-live backup schemes will go away if for
no other reason than it takes too long to restore. Is there any
backup technology today that are cheaper than using drives for
capacities > 10G?

-r



Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com

Chad Leigh -- Shire.Net LLC
2007-06-29 07:15:23 UTC
Permalink
Post by John Martinez
Post by Chad Leigh -- Shire.Net LLC
I know that as I migrate more to ZFS the actual layout/allocation/use
of disks/partitions/filesystems will change and these will be
smaller, but it is still easier for me to restore from a dump
interactively than to restore the whole thing to scratch space
somewhere and then have to search through the results to get the file
and move it back to the live file system. That takes ME more time
and my time is valuable.
It's been years since I've used ufs{dump,restore}, but I don't
believe it kept an index, or if it did, at the beginning of the tape,
That is OK, all my dumps live on a backup disk. dump/restore works
well

Chad


---
Chad Leigh -- Shire.Net LLC
Your Web App and Email hosting provider
chad at shire.net





Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
Russell Aspinwall
2007-06-27 06:20:23 UTC
Permalink
Hi,

My backup script add a text file of the backup partitions on the first
tape, the backup logs
also give information, allowing me to find the tape the backup is on.
If anyone wants a copy, send me an email of list.
Post by John Martinez
Post by Chad Leigh -- Shire.Net LLC
Post by Russell Aspinwall
The company mandates off site storage, so backing up to disk is
pointless unless it is off
site.
Well, your tapes are not off site until after the backup us run and
the tapes are fedex-ed or whatever to the offsite location.
So do the same with some 750GB drives. Do your monthly backup to
some drives and ship them offsite.
What about failure rates of disks vs. tapes? Disks tend to break if
handled incorrectly (dropped, etc.), but tapes can handle a bit more
abuse. I've never worked, or seen, an off-site data warehouse, but my
guess is that not everyone is going to treat the media gingerly.
Post by Chad Leigh -- Shire.Net LLC
Post by Russell Aspinwall
I did receive a post the ufsrestore functionality may be
implemented in
a later release of
Solaris U5
I am interested in this myself as I backup to disk but do it using
dump/restore which is very fast to restore specific files from
without having to have large spare disks to hold a 200gb file system
restore.
I *think* I mentioned it before, but we implement a three tier backup
system. Important data gets backed up to disk *and* to tape
simultaneously. The tapes are sent off-site for safe keeping. Less
important data is backed up to disk initially, then is backed up to
tape for archiving, as a "just in case." Other data is only backed up
to disk and rolls off when the data retention policy for that data
expires it. The idea is that backing up to disk is cheap and fast.
Especially true when you use devices that do compression, etc. Hey,
sounds like a good application for ZFS, don't it? ;-)
Post by Chad Leigh -- Shire.Net LLC
I know that as I migrate more to ZFS the actual layout/allocation/use
of disks/partitions/filesystems will change and these will be
smaller, but it is still easier for me to restore from a dump
interactively than to restore the whole thing to scratch space
somewhere and then have to search through the results to get the file
and move it back to the live file system. That takes ME more time
and my time is valuable.
It's been years since I've used ufs{dump,restore}, but I don't
believe it kept an index, or if it did, at the beginning of the tape,
but that is also an advantage of a commercial backup utility. It
keeps an index of files backed up somewhere in a database or tape,
where it can quickly look up a file or set of files for restore. If
you go disk to disk (regardless of filesystem), using find to get to
the right file can take up a bunch of time, as you say.
Really, I don't concern myself with backups at my current place,
because we have specialists that have it under control. It's just my
background in storage and servers that keeps me in that loop.
Sometimes much closer than I want to be. :-(
I am very interested in learning about people's ZFS implementations,
as I am researching it for implementation myself. It is very
compelling for its features and benefits. I am in no rush, though.
-john
______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
______________________________________________________________________
--
Regards

Russell

Email: russell dot aspinwall at flomerics dot co dot uk
Network and Systems Administrator Flomerics Ltd
Telephone: 020-8941-8810 x3116 81 Bridge Road
Facsimile: 020-8941-8730 Hampton Court
Surrey, KT8 9HH
United Kingdom



Please check the Links page before posting:
http://groups.yahoo.com/group/solarisx86/links
Post message: ***@yahoogroups.com
UNSUBSCRIBE: solarisx86-***@yahoogroups.com
Continue reading on narkive:
Loading...