[zfs-discuss] ZFS

[cephas maposah]

hello team

i have an issue with my ZFS system, i have 5 file systems and i need to take
a daily backup of these onto tape. how best do you think i should do these?
the smallest filesystem is about 50GB

here is what i have been doing i take snapshots of the 5 file systems, i zfs
send these into a directory gzip the the files and then tar them onto tape.
this takes a considerable amount of time.
my question is there a faster and better way of doing this?
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] ZFS

[Peter Tribble]

On Tue, Sep 13, 2011 at 9:48 AM, cephas maposah  wrote:
> hello team
> i have an issue with my ZFS system, i have 5 file systems and i need to take
> a daily backup of these onto tape. how best do you think i should do these?
> the smallest filesystem is about 50GB
> here is what i have been doing i take snapshots of the 5 file systems, i zfs
> send these into a directory gzip the the files and then tar them onto tape.
> this takes a considerable amount of time.
> my question is there a faster and better way of doing this?

So you zfs send to a file and save the file to tape?

Personally, I use zfs send and zfs receive to replicate the data to
a second system, and then simply tar the replicated file system
to tape. That way I'm not dependent on zfs to get the data back
(and can get individual files/directories back if necessary, which
it often is). And you have the ability to slot that copy of the data
instantly into service if the primary copy fails.

For tar, you can substitute a free or commercial backup solution.
It works the same way.

-- 
-Peter Tribble
http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] ZFS

[Ian Collins]

 On 09/13/11 09:00 PM, Peter Tribble wrote:

On Tue, Sep 13, 2011 at 9:48 AM, cephas maposah  wrote:

hello team
i have an issue with my ZFS system, i have 5 file systems and i need to take
a daily backup of these onto tape. how best do you think i should do these?
the smallest filesystem is about 50GB
here is what i have been doing i take snapshots of the 5 file systems, i zfs
send these into a directory gzip the the files and then tar them onto tape.
this takes a considerable amount of time.
my question is there a faster and better way of doing this?

So you zfs send to a file and save the file to tape?

Personally, I use zfs send and zfs receive to replicate the data to
a second system, and then simply tar the replicated file system
to tape. That way I'm not dependent on zfs to get the data back
(and can get individual files/directories back if necessary, which
it often is). And you have the ability to slot that copy of the data
instantly into service if the primary copy fails.

I'd also add that unless you are sending tapes off-site, having a 
snapshots on a seconds system can save you a lot of incremental backups.


--
Ian.

___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] ZFS

[cephas maposah]

i do send the tapes offline. The idea of sending onto another system seems
an excellent idea but i would still need to send them to tape. if i tar or
dump the replicated filesystems this will chew up a lot of space since my
filesystems are almost 900GB in total

On Tue, Sep 13, 2011 at 11:07 AM, Ian Collins  wrote:

>  On 09/13/11 09:00 PM, Peter Tribble wrote:
>
>> On Tue, Sep 13, 2011 at 9:48 AM, cephas maposah
>>  wrote:
>>
>>> hello team
>>> i have an issue with my ZFS system, i have 5 file systems and i need to
>>> take
>>> a daily backup of these onto tape. how best do you think i should do
>>> these?
>>> the smallest filesystem is about 50GB
>>> here is what i have been doing i take snapshots of the 5 file systems, i
>>> zfs
>>> send these into a directory gzip the the files and then tar them onto
>>> tape.
>>> this takes a considerable amount of time.
>>> my question is there a faster and better way of doing this?
>>>
>> So you zfs send to a file and save the file to tape?
>>
>> Personally, I use zfs send and zfs receive to replicate the data to
>> a second system, and then simply tar the replicated file system
>> to tape. That way I'm not dependent on zfs to get the data back
>> (and can get individual files/directories back if necessary, which
>> it often is). And you have the ability to slot that copy of the data
>> instantly into service if the primary copy fails.
>>
>>  I'd also add that unless you are sending tapes off-site, having a
> snapshots on a seconds system can save you a lot of incremental backups.
>
> --
> Ian.
>
>
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] ZFS

[Fajar A. Nugraha]

On Tue, Sep 13, 2011 at 3:48 PM, cephas maposah  wrote:
> hello team
> i have an issue with my ZFS system, i have 5 file systems and i need to take
> a daily backup of these onto tape. how best do you think i should do these?
> the smallest filesystem is about 50GB

It depends.

You can backup the files (so it'd be the same whatever filesystem the
files was on), or you can backup the send/recv stream.

> here is what i have been doing i take snapshots of the 5 file systems, i zfs
> send these into a directory gzip the the files and then tar them onto tape.
> this takes a considerable amount of time.
> my question is there a faster and better way of doing this?

yes, that sucks as you need to write it to a temporary file first. Is
your tape a real tape or VTL? If it's VTL, it might be easier to just
use it as disk so you can directly write the compressed zfs stream
there without the need of temporary file first. Also, using lzop is
faster than gzip (and less compression), so you might want to try that
as well.

-- 
Fajar
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] ZFS

[Fajar A. Nugraha]

On Tue, Sep 13, 2011 at 4:37 PM, Fajar A. Nugraha  wrote:
>> here is what i have been doing i take snapshots of the 5 file systems, i zfs
>> send these into a directory gzip the the files and then tar them onto tape.
>> this takes a considerable amount of time.
>> my question is there a faster and better way of doing this?
>
> yes, that sucks as you need to write it to a temporary file first.

EDIT: you CAN write the zfs stream directly to tape (using plain
output redirection, without using tar) so you don't need a temporary
file (at least this is true in Linux), but it's kinda awkward since
you need to remember (or store somewhere else) where the zfs stream
was from and what file number (on the tape) stores the stream.

So again I suggest you use disk as backup media when possible.

-- 
Fajar
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] ZFS

[Ian Collins]

 On 09/13/11 09:20 PM, cephas maposah wrote:
On Tue, Sep 13, 2011 at 11:07 AM, Ian Collins > wrote:


 On 09/13/11 09:00 PM, Peter Tribble wrote:

On Tue, Sep 13, 2011 at 9:48 AM, cephas
maposahmailto:mapo...@gmail.com>>  wrote:

hello team
i have an issue with my ZFS system, i have 5 file systems
and i need to take
a daily backup of these onto tape. how best do you think i
should do these?
the smallest filesystem is about 50GB
here is what i have been doing i take snapshots of the 5
file systems, i zfs
send these into a directory gzip the the files and then
tar them onto tape.
this takes a considerable amount of time.
my question is there a faster and better way of doing this?

So you zfs send to a file and save the file to tape?

Personally, I use zfs send and zfs receive to replicate the
data to
a second system, and then simply tar the replicated file system
to tape. That way I'm not dependent on zfs to get the data back
(and can get individual files/directories back if necessary, which
it often is). And you have the ability to slot that copy of
the data
instantly into service if the primary copy fails.

I'd also add that unless you are sending tapes off-site, having a
snapshots on a seconds system can save you a lot of incremental
backups.

i do send the tapes offline. The idea of sending onto another system 
seems an excellent idea but i would still need to send them to tape. 
if i tar or dump the replicated filesystems this will chew up a lot of 
space since my filesystems are almost 900GB in total


Pipe the output of tar or what ever you use through gzip (as you were 
originally?).


--
Ian.

___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] aclmode gone in S10u10?

[Paul Kraus]

On Mon, Sep 12, 2011 at 8:50 PM, Paul B. Henson  wrote:
> I recently saw a message posted to the sunmanagers list complaining
> about installing a kernel patch and suddenly having his ACL's disappear
> completely whenever a chmod occurred. I replied and asked him to check
> if the aclmode attribute was gone, as it sounded like the default
> discard that was (questionably) implemented in OpenSolaris/Solaris 11.
> He confirmed it was, so it looks like the removal of aclmode was
> backported to Solaris 10? I don't know exactly what kernel patch he
> installed; it doesn't look like update 10 is out yet.

Patch-ID# 144500-19
is the kernel update that is the kernel from 10U10

Thank you for flagging this, as Oracle support is telling me I have to
update to this release to get zpool 26 which fixes a zfs bug we are
running into, but if it breaks the ACL inheritance we have been using,
then it is non-starter.

> Can somebody in the know confirm whether or not aclmode is gone in
> update 10? I didn't think they'd backport such a feature disabling
> change to Solaris 10, seems to not line up with the "long term stability
> and compatibility" that's supposed to be the benefit there...

I cannot confirm this. So far I have one system at this kernel release
and have not yet been able to import the zpool with the problem. The
bug I am running into is that destruction of a snapshot runs the
system out of RAM. The snapshot is a partial zfs recv. I am told this
is a known bug (destruction of large snapshots can run the system out
of RAM as the destroy operation commits as one TXG). There is a fix,
but it in the on-disk format, so just zpool upgrading to version 26
will not fix existing snapshots. We only have 32 GB in this system and
the faulty snapshot *should* be about 2.5 TB.

-- 
{1-2-3-4-5-6-7-}
Paul Kraus
-> Senior Systems Architect, Garnet River ( http://www.garnetriver.com/ )
-> Sound Designer: Frankenstein, A New Musical
(http://www.facebook.com/event.php?eid=123170297765140)
-> Sound Coordinator, Schenectady Light Opera Company (
http://www.sloctheater.org/ )
-> Technical Advisor, RPI Players
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] ZFS

[Edward Ned Harvey]

> From: zfs-discuss-boun...@opensolaris.org [mailto:zfs-discuss-
> boun...@opensolaris.org] On Behalf Of cephas maposah
> 
> here is what i have been doing i take snapshots of the 5 file systems, i
zfs
> send these into a directory gzip the the files and then tar them onto
tape.
> this takes a considerable amount of time.
> my question is there a faster and better way of doing this?

zfs send | lzop | mbuffer
will write data stream directly to tape, no need for intermediate storage.
It has the disadvantage that you can't restore individual files, nor exclude
anything from backup, and if there's any corruption the whole stream is
useless.

zfs send | zfs receive (on disk)
Has the advantage that you can restore directly from snapshots if necessary
(much faster & easier than restoring from tape) and it verifies the
integrity of the data as it's written.  You can then use anything you want
(tar |lzop | mbuffer) or some commercial product, to write to tape.

___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] aclmode gone in S10u10?

[Peter Tribble]

On Tue, Sep 13, 2011 at 1:50 AM, Paul B. Henson  wrote:
> I recently saw a message posted to the sunmanagers list complaining
> about installing a kernel patch and suddenly having his ACL's disappear
> completely whenever a chmod occurred. I replied and asked him to check
> if the aclmode attribute was gone, as it sounded like the default
> discard that was (questionably) implemented in OpenSolaris/Solaris 11.
> He confirmed it was, so it looks like the removal of aclmode was
> backported to Solaris 10? I don't know exactly what kernel patch he
> installed; it doesn't look like update 10 is out yet.

Update 10 has been out for about 3 weeks.

> Can somebody in the know confirm whether or not aclmode is gone in
> update 10? I didn't think they'd backport such a feature disabling
> change to Solaris 10, seems to not line up with the "long term stability
> and compatibility" that's supposed to be the benefit there...

I just tried on a U9 and U10 box. On the U10 system, I did a
simple 'chmod g+s' on a directory with an ACL, and wham, the
ACL vanished. Same operation on U9, and the ACL is preserved.

(This doesn't affect me all that much, as ACLs on ZFS have never
really worked right, so anything where the ACL is critical gets stored
on ufs [yuck].)

Also, aclmode is no longer listed in the usage message you see
if you do 'zfs get'.

-- 
-Peter Tribble
http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] ZFS

[Paul Kraus]

On Tue, Sep 13, 2011 at 5:00 AM, Peter Tribble  wrote:

> Personally, I use zfs send and zfs receive to replicate the data to
> a second system, and then simply tar the replicated file system
> to tape. That way I'm not dependent on zfs to get the data back
> (and can get individual files/directories back if necessary, which
> it often is). And you have the ability to slot that copy of the data
> instantly into service if the primary copy fails.

tar or rsync will work for data, but you will lose any ACLs

The only tools I have found that work with zfs ACLs are the native zfs
tools (zfs send / recv), the native Solaris tools (cp, mv, etc.), and
Symantec NetBackup. I have not tried other commercial backup systems
as we already have NBU in house.

zfs send / recv is far, far faster than NBU.

-- 
{1-2-3-4-5-6-7-}
Paul Kraus
-> Senior Systems Architect, Garnet River ( http://www.garnetriver.com/ )
-> Sound Designer: Frankenstein, A New Musical
(http://www.facebook.com/event.php?eid=123170297765140)
-> Sound Coordinator, Schenectady Light Opera Company (
http://www.sloctheater.org/ )
-> Technical Advisor, RPI Players
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] aclmode gone in S10u10?

[Hung-Sheng Tsao (Lao Tsao 老曹) Ph.D.]

hi
I do not have s10u10 but in s10u9 zfs get does have aclmode
but in  s11x(express) zfs get does not have aclmode any more, there was 
big discussion and seems that  illumos has resurrect aclmode

regards
On 9/13/2011 8:21 AM, Peter Tribble wrote:

On Tue, Sep 13, 2011 at 1:50 AM, Paul B. Henson  wrote:

I recently saw a message posted to the sunmanagers list complaining
about installing a kernel patch and suddenly having his ACL's disappear
completely whenever a chmod occurred. I replied and asked him to check
if the aclmode attribute was gone, as it sounded like the default
discard that was (questionably) implemented in OpenSolaris/Solaris 11.
He confirmed it was, so it looks like the removal of aclmode was
backported to Solaris 10? I don't know exactly what kernel patch he
installed; it doesn't look like update 10 is out yet.

Update 10 has been out for about 3 weeks.


Can somebody in the know confirm whether or not aclmode is gone in
update 10? I didn't think they'd backport such a feature disabling
change to Solaris 10, seems to not line up with the "long term stability
and compatibility" that's supposed to be the benefit there...

I just tried on a U9 and U10 box. On the U10 system, I did a
simple 'chmod g+s' on a directory with an ACL, and wham, the
ACL vanished. Same operation on U9, and the ACL is preserved.

(This doesn't affect me all that much, as ACLs on ZFS have never
really worked right, so anything where the ACL is critical gets stored
on ufs [yuck].)

Also, aclmode is no longer listed in the usage message you see
if you do 'zfs get'.

<>___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] ZFS

[Bryan Hodgson]

Concerning native Solaris tools:

Solaris tar takes a 'p' flag that stores ACL information, and
re-creates same when extracted.  I've not made much use of it yet,
but it apparently works on UFS and ZFS (although there are obvious
difficulties restoring ZFS ACLs to UFS, etc.)

Gtar on Linux extracts the resulting tar file, but doesn't know
what to do with the ACL entries.

Current versions of Amanda support a 'suntar' backup type.  Haven't
explored that yet, but Amanda might be useful.

Bryan

On Tue, Sep 13, 2011 at 08:41:37AM -0400, Paul Kraus wrote:
> On Tue, Sep 13, 2011 at 5:00 AM, Peter Tribble  
> wrote:
> 
> > Personally, I use zfs send and zfs receive to replicate the data to
> > a second system, and then simply tar the replicated file system
> > to tape. That way I'm not dependent on zfs to get the data back
> > (and can get individual files/directories back if necessary, which
> > it often is). And you have the ability to slot that copy of the data
> > instantly into service if the primary copy fails.
> 
> tar or rsync will work for data, but you will lose any ACLs
> 
> The only tools I have found that work with zfs ACLs are the native zfs
> tools (zfs send / recv), the native Solaris tools (cp, mv, etc.), and
> Symantec NetBackup. I have not tried other commercial backup systems
> as we already have NBU in house.
> 
> zfs send / recv is far, far faster than NBU.
> 
> -- 
> {1-2-3-4-5-6-7-}
> Paul Kraus
> -> Senior Systems Architect, Garnet River ( http://www.garnetriver.com/ )
> -> Sound Designer: Frankenstein, A New Musical
> (http://www.facebook.com/event.php?eid=123170297765140)
> -> Sound Coordinator, Schenectady Light Opera Company (
> http://www.sloctheater.org/ )
> -> Technical Advisor, RPI Players
> ___
> zfs-discuss mailing list
> zfs-discuss@opensolaris.org
> http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

-- 

Bryan Hodgson   Lehigh University
bryan.hodg...@lehigh.eduPackard Lab 019 room 115
610-758-500119 Memorial Drive West
Bethlehem, PA 18015-3016
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] aclmode gone in S10u10?

[Paul B. Henson]

On 9/13/2011 5:07 AM, Paul Kraus wrote:


Patch-ID# 144500-19 is the kernel update that is the kernel from
10U10


Yep, the guy posting on sunmanagers confirmed that was the patch he
installed which broke aclmode.

Did update 10 sneak out under cover of darkness or what? I didn't see
any announcements or chatter about it, google doesn't find anything, and
the Oracle download site still only shows update 9:

http://www.oracle.com/technetwork/server-storage/solaris/overview/solaris-latest-version-170418.html


Thank you for flagging this, as Oracle support is telling me I have
to update to this release to get zpool 26 which fixes a zfs bug we
are running into, but if it breaks the ACL inheritance we have been
using, then it is non-starter.


Yup. Fortunately, there are no critical bugs (that Oracle is willing to
fix 8-/) that I've been waiting for. I thought I'd be able to keep an
up-to-date S10 install going while I figure out what to do next; guess
not :(.

He already opened a support ticket, they responded:

-
ZFS appears to be the only file system supporting NFSv4 ACLs
that attempts to preserve ACLs during chmod(2) operations.
Unfortunately, this requires the ACL to be modified in ways that are
confusing to customers and the time has come to stop the confusion and 
to just "discard" the ACL during chmod(2) operations. This implies that 
the ZFS aclmode property will no longer be needed and will be removed 
from ZFS.


This functionality is targetted to be back in Solaris 11 - as per
CR7002239 want ZFS aclmode property back
-

Interesting that there's already a new CR to put it back -- I thought 
that bridge was already burned. It's already back in Illumos. I wonder 
how long it will take to get put back in S10; or I guess it could be one 
of those CR's that never gets resolved. I suppose I'll open my own 
ticket to voice support.



runs the system out of RAM. The snapshot is a partial zfs recv. I am
told this is a known bug (destruction of large snapshots can run the
system out of RAM as the destroy operation commits as one TXG). There
is a fix, but it in the on-disk format, so just zpool upgrading to
version 26 will not fix existing snapshots. We only have 32 GB in
this system and the faulty snapshot *should* be about 2.5 TB.


Hmm, if updating the zpool won't fix the existing snapshot, how is 
support telling you to recover? Is it going to be one of those wipe and 
rebuild resolutions 8-/? Good luck...


I imagine a pool upgraded to version 26 will no longer be compatible 
with other zfs implementations.


--
Paul B. Henson  |  (909) 979-6361  |  http://www.csupomona.edu/~henson/
Operating Systems and Network Analyst  |  hen...@csupomona.edu
California State Polytechnic University  |  Pomona CA 91768
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] aclmode gone in S10u10?

[Paul B. Henson]

On 9/13/2011 5:21 AM, Peter Tribble wrote:


Update 10 has been out for about 3 weeks.


Where was any announcement posted? I haven't heard anything about it. As 
far as I can tell, the Oracle site still only has update 9 available for 
download:


http://www.oracle.com/technetwork/server-storage/solaris/downloads/index.html


I just tried on a U9 and U10 box. On the U10 system, I did a
simple 'chmod g+s' on a directory with an ACL, and wham, the
ACL vanished. Same operation on U9, and the ACL is preserved.


Meh, bogus :(. Thanks for the confirmation.


--
Paul B. Henson  |  (909) 979-6361  |  http://www.csupomona.edu/~henson/
Operating Systems and Network Analyst  |  hen...@csupomona.edu
California State Polytechnic University  |  Pomona CA 91768
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] aclmode gone in S10u10?

[Ian Collins]

 On 09/14/11 12:21 AM, Peter Tribble wrote:

I just tried on a U9 and U10 box. On the U10 system, I did a
simple 'chmod g+s' on a directory with an ACL, and wham, the
ACL vanished. Same operation on U9, and the ACL is preserved.

(This doesn't affect me all that much, as ACLs on ZFS have never
really worked right, so anything where the ACL is critical gets stored
on ufs [yuck].)


Not work on what way?

I have a client who makes extensive (more like excessive!) use of ACLs 
on ZFS and we don't see any problems.  Other than the ridiculous 
complexity of some of the ACLs that have grown over time.


--
Ian.

___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] aclmode gone in S10u10?

[Peter Tribble]

On Tue, Sep 13, 2011 at 8:34 PM, Paul B. Henson  wrote:
> On 9/13/2011 5:21 AM, Peter Tribble wrote:
>
>> Update 10 has been out for about 3 weeks.
>
> Where was any announcement posted? I haven't heard anything about it. As far
> as I can tell, the Oracle site still only has update 9 available for
> download:
>
> http://www.oracle.com/technetwork/server-storage/solaris/downloads/index.html

Hm. They updated that a few weeks ago with a new release but you're right,
it's now back to S10U9. Which leaves me with a whole slew of boxes running
a release that doesn't exist.

Oracle Solaris 10 8/11 s10x_u10wos_17a X86
  Copyright (c) 1983, 2011, Oracle and/or its affiliates. All rights reserved.
 Assembled 13 July 2011

-- 
-Peter Tribble
http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] aclmode gone in S10u10?

[Paul B. Henson]

On 9/13/2011 12:46 PM, Peter Tribble wrote:


Hm. They updated that a few weeks ago with a new release but you're right,
it's now back to S10U9. Which leaves me with a whole slew of boxes running
a release that doesn't exist.

 Oracle Solaris 10 8/11 s10x_u10wos_17a X86
   Copyright (c) 1983, 2011, Oracle and/or its affiliates. All rights reserved.
  Assembled 13 July 2011


Interesting. For some reason it just doesn't surprise me 8-/.

Maybe it got recalled because it broke aclmode ;).

On a different but related note, I went to try and open a support ticket 
about update 10, and for the life of me I couldn't figure out how to 
tell it the product in question was "Solaris". I'm not sure if something 
has changed since the last time I opened a bug (many months ago), but 
today the support site went beyond its usual painful all the way to 
unusable .



--
Paul B. Henson  |  (909) 979-6361  |  http://www.csupomona.edu/~henson/
Operating Systems and Network Analyst  |  hen...@csupomona.edu
California State Polytechnic University  |  Pomona CA 91768
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] aclmode gone in S10u10?

[Paul Kraus]

On Tue, Sep 13, 2011 at 3:31 PM, Paul B. Henson  wrote:

> Did update 10 sneak out under cover of darkness or what? I didn't see
> any announcements or chatter about it, google doesn't find anything, and
> the Oracle download site still only shows update 9:

It was supposed to be released on August 17, but I have not had
support point me at it yet. I suspect (based on some chatter) that
they found a bug that has forced them back to the drawing board for a
few weeks.

>> Thank you for flagging this, as Oracle support is telling me I have
>> to update to this release to get zpool 26 which fixes a zfs bug we
>> are running into, but if it breaks the ACL inheritance we have been
>> using, then it is non-starter.
>
> Yup. Fortunately, there are no critical bugs (that Oracle is willing to
> fix 8-/) that I've been waiting for. I thought I'd be able to keep an
> up-to-date S10 install going while I figure out what to do next; guess
> not :(.

Uhhh, not being able to destroy snapshots that are "too big" is a
pretty big one for us :-(

> He already opened a support ticket, they responded:
>
> -
> ZFS appears to be the only file system supporting NFSv4 ACLs
> that attempts to preserve ACLs during chmod(2) operations.
> Unfortunately, this requires the ACL to be modified in ways that are
> confusing to customers and the time has come to stop the confusion and to
> just "discard" the ACL during chmod(2) operations. This implies that the ZFS
> aclmode property will no longer be needed and will be removed from ZFS.

So the default behavior will be aclmode = discard, wonderful as
that will probably seriously break what I am working on right now.

> This functionality is targetted to be back in Solaris 11 - as per
> CR7002239 want ZFS aclmode property back
> -

>> runs the system out of RAM. The snapshot is a partial zfs recv. I am
>> told this is a known bug (destruction of large snapshots can run the
>> system out of RAM as the destroy operation commits as one TXG). There
>> is a fix, but it in the on-disk format, so just zpool upgrading to
>> version 26 will not fix existing snapshots. We only have 32 GB in
>> this system and the faulty snapshot *should* be about 2.5 TB.
>
> Hmm, if updating the zpool won't fix the existing snapshot, how is support
> telling you to recover? Is it going to be one of those wipe and rebuild
> resolutions 8-/? Good luck...

I am having that discussion with Oracle Support right now :-) One
option is to load the system with enough RAM to destroy the legacy
snapshots and then not let them grow too big (in other words, take
more frequent snapshots). But I have been having trouble getting
support to tell me how to estimate the amount of RAM necessary to
delete a snapshot based on *something* I can measure.

P.S. The backup server ran into the bug full steam ahead and is off
line right now. On production, once I realized the root cause I
stopped the snapshot destroy script, so they are just piling up now
:-) Production is 20 TB and 400 million objects, we can't reload that,
the outage would kill us.

-- 
{1-2-3-4-5-6-7-}
Paul Kraus
-> Senior Systems Architect, Garnet River ( http://www.garnetriver.com/ )
-> Sound Designer: Frankenstein, A New Musical
(http://www.facebook.com/event.php?eid=123170297765140)
-> Sound Coordinator, Schenectady Light Opera Company (
http://www.sloctheater.org/ )
-> Technical Advisor, RPI Players
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] aclmode gone in S10u10?

[Paul B. Henson]

On Tue, Sep 13, 2011 at 12:41:36PM -0700, Ian Collins wrote:
> Not work on what way?
> 
> I have a client who makes extensive (more like excessive!) use of ACLs 
> on ZFS and we don't see any problems.  Other than the ridiculous 
> complexity of some of the ACLs that have grown over time.

>From my perspective, the fact that any random process can come along and
do a legacy chmod() and destroy your acl falls pretty clearly into the
"not work" camp ... That and the fact that an NFSv4 exclusive open
destroys an inherited ACL :(. If you can somehow keep things from trying
to poke legacy mode bits and don't use NFS, maybe you're ok.

-- 
Paul B. Henson  |  (909) 979-6361  |  http://www.csupomona.edu/~henson/
Operating Systems and Network Analyst  |  hen...@csupomona.edu
California State Polytechnic University  |  Pomona CA 91768
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

[zfs-discuss] NexentaStor auto-sync - zfs snapshot destroys times

[Evgueni Martynov]

Does anyone see long timeouts during zfs snapshot destroys?
How long does it take to destroy in your situation?
If yes, in what kind of configuration (NexentaStor version if applicable, snapshot size, type of IO to the pool, zpool 
type)?

I'm interested to hear if anyone uses NexentaStor/Opensolaris ZFS platform to 
provide storage (datastores)
for VMware vSphere environments via NFS/iscsi.

Thanks
Evgueni
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] NexentaStor auto-sync - zfs snapshot destroys times

[Ian Collins]

 On 09/14/11 01:59 PM, Evgueni Martynov wrote:

It's not a good idea to hijack an existing thread!


Does anyone see long timeouts during zfs snapshot destroys?


Define long in this context.


How long does it take to destroy in your situation?


No more than a second or two.


If yes, in what kind of configuration (NexentaStor version if applicable, 
snapshot size, type of IO to the pool, zpool
type)?


You should start off describing your configuration, the most common 
cause of long snapshot destroys is dedup and insufficient RAM/cache.


--
Ian.

___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] NexentaStor auto-sync - zfs snapshot destroys times

[Evgueni Martynov]

On 13/09/2011 10:12 PM, Ian Collins wrote:

On 09/14/11 01:59 PM, Evgueni Martynov wrote:

It's not a good idea to hijack an existing thread!

Does anyone see long timeouts during zfs snapshot destroys?


Define long in this context.


~1-2 minutes, may be longer




How long does it take to destroy in your situation?


No more than a second or two.

That's what I saw before and the numbers that I would be comfortable with.
I saw bigger timeouts from the NFS client side (ESX servers).
I saw ESX servers loosing datastores for 50 sec and more interested if anyone saw that as well an what kind of IO going 
to that zpool. We have lots of VMs. And I'm looking for ideas of how to replicate that in none production environment.

Maybe record the IO on a production server and replay during snapshot destroy 
in lab environment.




If yes, in what kind of configuration (NexentaStor version if applicable, 
snapshot size, type of IO to the pool, zpool
type)?


You should start off describing your configuration, the most common cause of 
long snapshot destroys is dedup and
insufficient RAM/cache.

no dedupe and 96 GB of RAM




___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] aclmode gone in S10u10?

[Florian Wagner]

> [...]
> 
> He already opened a support ticket, they responded:
> 
> -
> ZFS appears to be the only file system supporting NFSv4 ACLs
> that attempts to preserve ACLs during chmod(2) operations.
> Unfortunately, this requires the ACL to be modified in ways that are
> confusing to customers and the time has come to stop the confusion
> and to just "discard" the ACL during chmod(2) operations. This
> implies that the ZFS aclmode property will no longer be needed and
> will be removed from ZFS.
> 
> This functionality is targetted to be back in Solaris 11 - as per
> CR7002239 want ZFS aclmode property back
> -
> 
> [...]

Just so you know and can respond to Oracles wrong statements with
facts: ZFS is not the only filesystem doing something like that.
NetApps WAFL provides an option to make it behave very similar.


Regards
Florian Wagner


signature.asc
Description: PGP signature
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss