[zfs-discuss] Replace Drives in RAIDZ Array

[Chris]

I found this posted on another site and its exactly what I want to do.  I have 
6 drives I want to replace with 6 larger drives.  Can anyone help me with what 
commands I would need to run?  Do I replace, Offline, etc?

==
2) replace all the drives in the raidz, 1 by 1 (waiting for zfs to reconstruct 
the data on the new drive), and when you have replaced the final drive the 
zpool will "magically" increase in size. i.e replace the 500GB drives, 1 by 1, 
with 750GB drives, and when you finish the zpool effective storage will jump to 
2250GB.
==

thanks,
Chris
-- 
This message posted from opensolaris.org
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

[zfs-discuss] Backup/Restore root pool : SPARC and x86/x64

[Marlanne DeLaSource]

I have been impressed to see my blade1500 booting ZFS for the first time. Even 
if the OBP does not boot EFI disk yet, we can already greatly simplify the 
everyday life of the sysadmin with this awaited feature.

I was checking in the official Sun doc and on opensolaris doc and I didn't 
notice any clear backup/restore procedure for the root pool (or maybe my search 
was not accurate enough).

Is there any procedure for backup (I imagine zfs send is the way, but how can I 
export the metadata - pool/fs) and for restore (like boot cdrom and so on).

Thanks for your answers.

PS: We still badly miss the remove for vdev in a pool ;)
-- 
This message posted from opensolaris.org
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] OpenStorage GUI

[Clayton Wheeler]

> On Tue, Nov 11, 2008 at 11:26:46AM -0800, Adam
> Leventhal wrote:
> > On Nov 11, 2008, at 10:41 AM, Brent Jones wrote:
> > > Wish I could get my hands on a beta of this
> GUI...
> > 
> > Take a look at the VMware version that you can run
> on any machine:
> > 
> http://www.sun.com/storage/disk_systems/unified_storage
> /resources.jsp
> 
> So long as "any machine" != ESX, then yes. ;)
> 
> -brian

Actually, it works pretty well on ESX 3.5 U3 for me. I just had to import the 
VM as an "Other" type VM using VMware Converter Enterprise, the same way you 
would for VMware Server VMs or whatever. For some reason I had to remove two of 
the network interfaces, also, but after that it fired right up.
-- 
This message posted from opensolaris.org
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

[zfs-discuss] Possible ZFS panic on Solaris 10 Update 6

[Ian Collins]

The system was an x4540 running Solaris 10 Update 6 acting as a
production Samba server.

The only unusual activity was me sending and receiving incremental dumps
to and from another system.

When the system came back, the filesystem/local service was in
maintenance and had to be manually restarted.

Unfortunately I don't have the crash dump (it failed), but the panic
message was:

Nov 12 17:36:43 ares ^Mpanic[cpu1]/thread=fe8d2c087460:
Nov 12 17:36:43 ares genunix: [ID 683410 kern.notice] BAD TRAP: type=e (#pf 
Page fault) rp=fe800429a6e0 addr=fe893f924000
Nov 12 17:36:43 ares unix: [ID 10 kern.notice]
Nov 12 17:36:43 ares unix: [ID 839527 kern.notice] zfs:
Nov 12 17:36:43 ares unix: [ID 753105 kern.notice] #pf Page fault
Nov 12 17:36:43 ares unix: [ID 532287 kern.notice] Bad kernel fault at 
addr=0xfe893f924000
Nov 12 17:36:43 ares unix: [ID 243837 kern.notice] pid=20825, 
pc=0xfb8303ba, sp=0xfe800429a7d8, eflags=0x10216
Nov 12 17:36:43 ares unix: [ID 211416 kern.notice] cr0: 
8005003b cr4: 6f0
Nov 12 17:36:43 ares unix: [ID 354241 kern.notice] cr2: fe893f924000 cr3: 
420275000 cr8: c
Nov 12 17:36:43 ares unix: [ID 592667 kern.notice]  rdi: fe8b461d2640 
rsi: fe893f924000 rdx:  108
Nov 12 17:36:43 ares unix: [ID 592667 kern.notice]  rcx:   19  
r8:0  r9:0
Nov 12 17:36:43 ares unix: [ID 592667 kern.notice]  rax:  100 
rbx: fe89658356c0 rbp: fe800429a850
Nov 12 17:36:44 ares unix: [ID 592667 kern.notice]  r10:1 
r11:0 r12:  108
Nov 12 17:36:44 ares unix: [ID 592667 kern.notice]  r13: fe800429a864 
r14: fe8d014b1c70 r15:0
Nov 12 17:36:44 ares unix: [ID 592667 kern.notice]  fsb: 8000 
gsb: 8b708000  ds:   43
Nov 12 17:36:44 ares unix: [ID 592667 kern.notice]   es:   43  
fs:0  gs:  1c3
Nov 12 17:36:44 ares unix: [ID 592667 kern.notice]  trp:e 
err:0 rip: fb8303ba
Nov 12 17:36:44 ares unix: [ID 592667 kern.notice]   cs:   28 
rfl:10216 rsp: fe800429a7d8
Nov 12 17:36:44 ares unix: [ID 266532 kern.notice]   ss:   30
Nov 12 17:36:44 ares unix: [ID 10 kern.notice]
Nov 12 17:36:44 ares genunix: [ID 655072 kern.notice] fe800429a5f0 
unix:real_mode_end+7201 ()
Nov 12 17:36:44 ares genunix: [ID 655072 kern.notice] fe800429a6d0 
unix:trap+5e6 ()
Nov 12 17:36:44 ares genunix: [ID 655072 kern.notice] fe800429a6e0 
unix:_cmntrap+140 ()
Nov 12 17:36:44 ares genunix: [ID 655072 kern.notice] fe800429a850 
unix:bcopy+a ()
Nov 12 17:36:44 ares genunix: [ID 655072 kern.notice] fe800429a890 
zfs:dbuf_read+95 ()
Nov 12 17:36:44 ares genunix: [ID 655072 kern.notice] fe800429a8c0 
zfs:dmu_bonus_hold+8f ()
Nov 12 17:36:44 ares genunix: [ID 655072 kern.notice] fe800429a910 
zfs:restore_object+191 ()
Nov 12 17:36:44 ares genunix: [ID 655072 kern.notice] fe800429aa60 
zfs:dmu_recv_stream+5e6 ()
Nov 12 17:36:44 ares genunix: [ID 655072 kern.notice] fe800429ad30 
zfs:zfs_ioc_recv+247 ()
Nov 12 17:36:44 ares genunix: [ID 655072 kern.notice] fe800429ad80 
zfs:zfsdev_ioctl+14c ()
Nov 12 17:36:44 ares genunix: [ID 655072 kern.notice] fe800429ad90 
genunix:cdev_ioctl+1d ()
Nov 12 17:36:45 ares genunix: [ID 655072 kern.notice] fe800429adb0 
specfs:spec_ioctl+50 ()
Nov 12 17:36:45 ares genunix: [ID 655072 kern.notice] fe800429ade0 
genunix:fop_ioctl+25 ()
Nov 12 17:36:45 ares genunix: [ID 655072 kern.notice] fe800429aec0 
genunix:ioctl+ac ()
Nov 12 17:36:45 ares genunix: [ID 655072 kern.notice] fe800429af10 
unix:brand_sys_syscall32+1a3 ()
Nov 12 17:36:45 ares unix: [ID 10 kern.notice]
Nov 12 17:36:45 ares genunix: [ID 672855 kern.notice] syncing file systems...
Nov 12 17:36:45 ares genunix: [ID 733762 kern.notice]  43
Nov 12 17:36:47 ares genunix: [ID 904073 kern.notice]  done
Nov 12 17:36:48 ares genunix: [ID 111219 kern.notice] dumping to 
/dev/zvol/dsk/rpool/dump, offset 65536, content: kernel
Nov 12 17:57:03 ares genunix: [ID 409368 kern.notice] ^M 81% done: 3324566 
pages dumped, compression ratio 3.28,
Nov 12 17:57:03 ares genunix: [ID 495082 kern.notice] dump failed: error 28

-- 
Ian.

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

Re: [zfs-discuss] Rewrite vdev label?

[Dale Ghent]

On Nov 11, 2008, at 11:13 PM, Dan Mick wrote:

> I have a situation where the first part of the first label on the  
> vdev is
> corrupted (from 8K to 16K on the slice).
>
> Can I use a zfs/zpool command to rewrite that label from one of the  
> other three?
> So far I've tried

I'm not aware of something like that being in zdb.

If you're sure of the mangled zone (offset, length) you might just  
want to resort to dd'ing what you need off a good disk and then dd'ing  
those bits on to the ailing disk.

Or if the ailing disk is in a mirror set, just wipe and reattach it.

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

[zfs-discuss] Rewrite vdev label?

[Dan Mick]

I have a situation where the first part of the first label on the vdev is
corrupted (from 8K to 16K on the slice).

Can I use a zfs/zpool command to rewrite that label from one of the other three?
So far I've tried

zpool update
zpool export
zpool import

Some zdb magic?

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

Re: [zfs-discuss] OpenStorage GUI

[Ed Saipetch]

Boyd,

That's exactly what I was getting at.  This list probably isn't the  
place to discuss but this is the first real instance aside of maybe  
xVM Ops Center where it was pretty much put out in the open that you  
can expect to pay to get the goods.

Fishworks seems to have much more than just a nice wrapper put around  
Solaris, ZFS, NFS, FMA, AVS etc.  A lot of my ability to evangelize  
the benefits of Solaris in the storage world to my customers hinges on  
me being able to say "Try it... you'll like it...".  I know Try-and- 
buy exists but in the grand scheme of things, adoption of Solaris  
hinges on easy accessibility.

I apologize for the tangent and the VM instance is a good start but  
the stance on opensourcing right or wrong seems like it has changed.

On Nov 11, 2008, at 8:30 PM, Boyd Adamson wrote:

> Bryan Cantrill <[EMAIL PROTECTED]> writes:
>
>> On Tue, Nov 11, 2008 at 02:21:11PM -0500, Ed Saipetch wrote:
>>> Can someone clarify Sun's approach to opensourcing projects and
>>> software?  I was under the impression the strategy was to charge for
>>> hardware, maintenance and PS.  If not, some clarification would be  
>>> nice.
>>
>> There is no single answer -- we use open source as a business  
>> strategy,
>> not as a checkbox or edict.  For this product, open source is an  
>> option
>> going down the road, but not a priority.  Will our software be open
>> sourced in the fullness of time?  My Magic 8-Ball tells me "signs
>> point to yes" (or is that "ask again later"?) -- but it's certainly
>> not something that we have concrete plans for at the moment...
>
> I think that's fair enough. What Sun choose to do is, of course, up to
> Sun.
>
> One can, however, understand that people might have expected otherwise
> given statements like this:
>
>> "With our announced intent to open source the entirety of our  
>> software
>> offerings, every single developer across the world now has access to
>> the most sophisticated platform available for web 1.0, 2.0 and  
>> beyond"
>
> - Jonathan Schwartz
> http://www.sun.com/smi/Press/sunflash/2005-11/sunflash.20051130.1.xml
>
> -- 
> Boyd

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

Re: [zfs-discuss] Strange zfs receive error

[Ian Collins]

Richard Elling wrote:
> Ian Collins wrote:
>> I've been replicating a number of filesystems from a Solaris 10
>> update 6 system to an update 5 one.  All of the filesystems receive
>> fine except for one, which fails with
>> cannot receive: invalid backup stream
>>   
>
> What are the zfs versions?  ("zfs upgrade" command output)
Update 5 us V4, Update 6 is V10.

-- 
Ian.

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

Re: [zfs-discuss] OpenStorage GUI

[Boyd Adamson]

Bryan Cantrill <[EMAIL PROTECTED]> writes:

> On Tue, Nov 11, 2008 at 02:21:11PM -0500, Ed Saipetch wrote:
>> Can someone clarify Sun's approach to opensourcing projects and  
>> software?  I was under the impression the strategy was to charge for  
>> hardware, maintenance and PS.  If not, some clarification would be nice.
>
> There is no single answer -- we use open source as a business strategy,
> not as a checkbox or edict.  For this product, open source is an option
> going down the road, but not a priority.  Will our software be open
> sourced in the fullness of time?  My Magic 8-Ball tells me "signs
> point to yes" (or is that "ask again later"?) -- but it's certainly 
> not something that we have concrete plans for at the moment...

I think that's fair enough. What Sun choose to do is, of course, up to
Sun.

One can, however, understand that people might have expected otherwise
given statements like this:

> "With our announced intent to open source the entirety of our software
> offerings, every single developer across the world now has access to
> the most sophisticated platform available for web 1.0, 2.0 and beyond"

- Jonathan Schwartz
http://www.sun.com/smi/Press/sunflash/2005-11/sunflash.20051130.1.xml

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

Re: [zfs-discuss] Lost space in empty pool (no snapshots)

[Anton B. Rang]

The "deferred free" indicates that these blocks are supposed to be freed at a 
future time. A quick glance at the code would seem to indicate that this is 
supposed to happen when the next transaction group is pushed. Apparently it's 
not happening on your system ... presumably a bug.
-- 
This message posted from opensolaris.org
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] grub prompt after aborted ufs to zfs live upgrade

[Robert Buick]

Since the above post, I downloaded Super Grub and managed to recover a trashed 
(grub) FC8 installation that was lost when nevada was installed some time back. 
This now has as well as the FC 8 prompt  in the Grub menu:

title Solaris - BE1
rootnoverify (hd0,2,a)
chainloader +1
makeactive

title Solaris - BE2
rootnoverify (hd0,2,e)
chainloader +1
makeactive

Selecting either of the Solaris choices receives:
Error13: Invalid or unsupported executable format

Back with Super Grub, selecting Partitions presents, amongst the other 
partitions the two 15GB BE's I was using for live upgrade, namely:
N   IDE SCSIGRUBHURDTYPESIZEOS
.
.
3   hda3sda3(hd0,2,a)   hd0s3   ufs 15GB
.
.
3   hda3sda3(hd0,2,e)   hd0s3   swap15GB


dropping back to a  grub prompt and entering root returns:

root (hd0,2,e):Filesystem type is zfs, partition type 0x82

As I mentioned in my earlier post, the zfs BE had been deleted, but not the 
zpool.

Any of the other Super Grub options to get the solaris running ends in failure.

Thanks for your time.
-- 
This message posted from opensolaris.org
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] Strange zfs receive error

[Richard Elling]

Ian Collins wrote:
> I've been replicating a number of filesystems from a Solaris 10 update 6 
> system to an update 5 one.  All of the filesystems receive fine except for 
> one, which fails with 
>
> cannot receive: invalid backup stream
>   

What are the zfs versions?  ("zfs upgrade" command output)
 -- richard

> I can receive this stream on another update 6 system.
>
> Are there any zfs options that may cause this?  The troublesome fs is the 
> only one I have with sharenfs options (rw,root=*,anon=0).
>
> Thanks.
>
>   

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

Re: [zfs-discuss] OpenStorage GUI

[Brent Jones]

On Tue, Nov 11, 2008 at 11:26 AM, Adam Leventhal <[EMAIL PROTECTED]> wrote:
> On Nov 11, 2008, at 10:41 AM, Brent Jones wrote:
>>
>> Wish I could get my hands on a beta of this GUI...
>
>
> Take a look at the VMware version that you can run on any machine:
>
>  http://www.sun.com/storage/disk_systems/unified_storage/resources.jsp
>
> Adam
>
> --
> Adam Leventhal, Fishworkshttp://blogs.sun.com/ahl
>
>

Tried loading this under Virtualbox and encountered some issues with
all the virtual disks it required, but VMWare player worked fine.

The system looks pretty sweet, I'll be sure to show this off here to
get a head start purchasing!

-- 
Brent Jones
[EMAIL PROTECTED]
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] OpenStorage GUI

[Brian Hechinger]

On Tue, Nov 11, 2008 at 11:26:46AM -0800, Adam Leventhal wrote:
> On Nov 11, 2008, at 10:41 AM, Brent Jones wrote:
> > Wish I could get my hands on a beta of this GUI...
> 
> Take a look at the VMware version that you can run on any machine:
> 
>http://www.sun.com/storage/disk_systems/unified_storage/resources.jsp

So long as "any machine" != ESX, then yes. ;)

-brian
-- 
"Coding in C is like sending a 3 year old to do groceries. You gotta
tell them exactly what you want or you'll end up with a cupboard full of
pop tarts and pancake mix." -- IRC User (http://www.bash.org/?841435)
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] OpenStorage GUI

[Miles Nordin]

> "es" == Ed Saipetch <[EMAIL PROTECTED]> writes:

es> Sun's approach to opensourcing projects

AIUI, only the development branch is open-sourced.  The stable
branches remain proprietary: no source for sol10u, and no stable
branch for opensolaris ips.  sol10u is free-as-in-free-beer for
now, but you cannot make one small change and rebuild it, nor can you
duplicate the DVD and give it to others.  Sun loves to explain their
release strategy, but the explanation is always so careful I wind up
more confused after I've heard it than before.

Nothing stops another distribution like Nexenta to make their own
hopefully-more-stable branch from the unstable solaris source, but
copyright ownership will prevent the competing community member from
taking their branch proprietary as Sun can do with theirs.  Sun is
selling their patch maintenance in a CentOS-ification-proof way.
Fewer people than usual seem to care among ZFS users because the
stable branch isn't stable, but if it were it'd suck to be stuck
choosing between freedom and a stable release.

then, CDDL is intentionally GPL-incompatible which makes a fork of the
unstable trunk less likely.

Also it is a hybrid closed/open thing.  They have stated ``all new
components added to Solaris will be open-source,'' but this doesn't
include new drivers.  There are lots of proprietary 3rd party drivers,
as there are with Linux, but unlike Linux many of the inbuilt drivers
for hardware Sun sells are proprietary.  ``All new components'' seems
to mean their intent is for anything on which their marketing
spotlight falls to beat an open-source drum---the promise has fallen
rather flat for cut-off dates or product boundaries compared to
something like Ubuntu.

Unlike most Linuxes the compiling of bundled 'sfw' freeware into SYSV
packages is a manual process, and there've been reports that the
source package sometimes doesn't match the binary package or simply
doesn't build.  Remember ``including all build scripts'' from the GPL?
if your source doesn't build, make sure you build it by hand so you
can try to claim the recipient jsut doesn't have enough skill to type
the right incantations.  This is another way to become the main source
for stable releases---you can't add a huge subsystem to Samba this way
because you'll get caught, but you can let a few bug fixes and build
tricks slip through the cracks, gaining the advantage while still
appearing in good faith.

And they've done things that look legal to me but very questionable
from a software freedom advocate's standpoint, like their franken-gcc
which, unlike free-as-in-freedom gcc, makes SPARC executables of
reasonable performance (good!) by combining gcc's front-end with Forte
proprietary back-end (upsetting.).

From a user who cares about software freedom's standpoint, it's much,
*much* better than Mac OS X, but meaningfully worse than Linux on
several fronts.  It's hard to tell what's strategic from what's just
unfortunate.  So far publicity seems to be one of the successful
prongs of the strategy---virtually all the Sun criticism you'll read
is inane stock market punditry, business-porn, predictions threats
unsolicited-advice---that sort of nonsense.  Criticism on the software
freedom front is minimal or hazily-informed from what I've seen.
Critics keep saying ``let's wait and see'' for the last three years
but refuse to actually see anything and talk about it.


pgpew166OQRPx.pgp
Description: PGP signature
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] OpenStorage GUI

[Scott Williamson]

We would be very interested in this. I have just gotten my thumpers
(x4500) into preproduction testing as NFS NAS servers. Our real challenge
has been lack of a management GUI so that long complex lists of commands do
not need to be followed.

I do not run my own apps on them, so I would be interested in an upgrade
path.
On Tue, Nov 11, 2008 at 12:38 PM, Bryan Cantrill <[EMAIL PROTECTED]> wrote:

> On Tue, Nov 11, 2008 at 09:31:26AM -0800, Adam Leventhal wrote:
> > > Is this software available for people who already have thumpers?
> >
> > We're considering offering an upgrade path for people with existing
> > thumpers. Given the feedback we've been hearing, it seems very likely
> > that we will. No word yet on pricing or availability.
>
> Just to throw some ice-cold water on this:
>
>  1.  It's highly unlikely that we will ever support the x4500 -- only the
>  x4540 is a real possibility.
>
>  2.  If we do make something available, your data and any custom
>  software won't survive the journey:  you will be forced to
>  fresh-install your x4540 with our stack.
>
>  3.  If we do make something available, it will become an appliance:  you
>  will permanently lose the ability to run your own apps on the x4540.
>
>  4.  If we do make something available, it won't be free.
>
> If you are willing/prepared(/eager?) to abide by these constraints, please
> let us ([EMAIL PROTECTED]) know -- that will help us build the business
> case for doing this...
>
>- Bryan
>
> --
> Bryan Cantrill, Sun Microsystems Fishworks.   http://blogs.sun.com/bmc
>
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] OpenStorage GUI

[Andy Lubel]

-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Bryan Cantrill
Sent: Tuesday, November 11, 2008 12:39 PM
To: Adam Leventhal
Cc: zfs-discuss@opensolaris.org
Subject: Re: [zfs-discuss] OpenStorage GUI

On Tue, Nov 11, 2008 at 09:31:26AM -0800, Adam Leventhal wrote:
> > Is this software available for people who already have thumpers?
> 
> We're considering offering an upgrade path for people with existing 
> thumpers. Given the feedback we've been hearing, it seems very likely 
> that we will. No word yet on pricing or availability.

Just to throw some ice-cold water on this:

  1.  It's highly unlikely that we will ever support the x4500 -- only
the
  x4540 is a real possibility.

  2.  If we do make something available, your data and any custom
  software won't survive the journey:  you will be forced to
  fresh-install your x4540 with our stack.

  3.  If we do make something available, it will become an appliance:
you
  will permanently lose the ability to run your own apps on the
x4540.

  4.  If we do make something available, it won't be free.

If you are willing/prepared(/eager?) to abide by these constraints,
please let us ([EMAIL PROTECTED]) know -- that will help us build the
business case for doing this...

- Bryan


--
Bryan Cantrill, Sun Microsystems Fishworks.
http://blogs.sun.com/bmc
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss


--



1. Can someone let me know who to talk to about upgrading our 4500's
(offline is fine).  We did buy them about 3 weeks before the new 4540
was announced and were pretty stung that our rep didn't tell us what was
on the horizon.. Im an apple user so im pretty used to that anyways.  I
just didn't know that there was a path for upgrading to what a 4540 is.

2. That's why we bought 2 and keep each half full :)

3. I like that, even better would be a way to install it without
dedicating spindles to OS.

4. I am ok with value added software being sold by Sun.  We don't mind
paying money if it makes our job actually less complex each workday!

Im going to give this vmware image a whirl and see what its missing :)  

-Andy

PS - sorry for the incorrect format for responding, I didn't have a real
email client available today.
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] OpenStorage GUI

[Bryan Cantrill]

On Tue, Nov 11, 2008 at 02:21:11PM -0500, Ed Saipetch wrote:
> Can someone clarify Sun's approach to opensourcing projects and  
> software?  I was under the impression the strategy was to charge for  
> hardware, maintenance and PS.  If not, some clarification would be nice.

There is no single answer -- we use open source as a business strategy,
not as a checkbox or edict.  For this product, open source is an option
going down the road, but not a priority.  Will our software be open
sourced in the fullness of time?  My Magic 8-Ball tells me "signs
point to yes" (or is that "ask again later"?) -- but it's certainly 
not something that we have concrete plans for at the moment...

- Bryan

--
Bryan Cantrill, Sun Microsystems Fishworks.   http://blogs.sun.com/bmc
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] ZFS on Fit-PC Slim?

[Francois Dion]

That's why I said Nexenta. It's text mode after the fact, from the get go, 
after the install and reboot. 

http://www.nexenta.org/os

-Original Message-
>> Nexenta and Belenix will also run on this machine
>> with 512MB. Nexenta will probably be what you want if
>> you are saying you are running ubuntu on this
>> box.

>Perhaps I wasn't sufficiently clear.

>I *installed* in text mode of course.

>However on booting the system it wants to start graphics.
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] OpenStorage GUI

[Adam Leventhal]

On Nov 11, 2008, at 10:41 AM, Brent Jones wrote:
> Wish I could get my hands on a beta of this GUI...


Take a look at the VMware version that you can run on any machine:

   http://www.sun.com/storage/disk_systems/unified_storage/resources.jsp

Adam

--
Adam Leventhal, Fishworkshttp://blogs.sun.com/ahl

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

Re: [zfs-discuss] OpenStorage GUI

[Ed Saipetch]

Can someone clarify Sun's approach to opensourcing projects and  
software?  I was under the impression the strategy was to charge for  
hardware, maintenance and PS.  If not, some clarification would be nice.

On Nov 11, 2008, at 12:38 PM, Bryan Cantrill wrote:
>
>  4.  If we do make something available, it won't be free.
>
> If you are willing/prepared(/eager?) to abide by these constraints,  
> please
> let us ([EMAIL PROTECTED]) know -- that will help us build the  
> business
> case for doing this...
>
>   - Bryan
>
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] OpenStorage GUI

[Brent Jones]

On Tue, Nov 11, 2008 at 9:52 AM, Adam Leventhal <[EMAIL PROTECTED]> wrote:
> On Nov 11, 2008, at 9:38 AM, Bryan Cantrill wrote:
>
>> Just to throw some ice-cold water on this:
>>
>>  1.  It's highly unlikely that we will ever support the x4500 --
>> only the
>>  x4540 is a real possibility.
>
>
> And to warm things up a bit: there's already an upgrade path from the
> x4500 to the x4540 so that would be required before any upgrade to the
> equivalent of the Sun Storage 7210.
>
> Adam
>
> --
> Adam Leventhal, Fishworkshttp://blogs.sun.com/ahl
>
> ___
> zfs-discuss mailing list
> zfs-discuss@opensolaris.org
> http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
>

We just ordered several X4540's, excited to get them in place soon.
Having the Openstorage GUI as an option down the road is very
appealing for our VM/hosted side, after we install this bulk storage
environment.

Wish I could get my hands on a beta of this GUI...

-- 
Brent Jones
[EMAIL PROTECTED]
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] OpenStorage GUI

[Bryan Cantrill]

On Tue, Nov 11, 2008 at 09:31:26AM -0800, Adam Leventhal wrote:
> > Is this software available for people who already have thumpers?
> 
> We're considering offering an upgrade path for people with existing
> thumpers. Given the feedback we've been hearing, it seems very likely
> that we will. No word yet on pricing or availability.

Just to throw some ice-cold water on this:

  1.  It's highly unlikely that we will ever support the x4500 -- only the
  x4540 is a real possibility.

  2.  If we do make something available, your data and any custom
  software won't survive the journey:  you will be forced to
  fresh-install your x4540 with our stack.

  3.  If we do make something available, it will become an appliance:  you
  will permanently lose the ability to run your own apps on the x4540.

  4.  If we do make something available, it won't be free.

If you are willing/prepared(/eager?) to abide by these constraints, please
let us ([EMAIL PROTECTED]) know -- that will help us build the business
case for doing this...

- Bryan

--
Bryan Cantrill, Sun Microsystems Fishworks.   http://blogs.sun.com/bmc
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Re: [zfs-discuss] OpenStorage GUI

[Adam Leventhal]

On Nov 11, 2008, at 9:38 AM, Bryan Cantrill wrote:

> Just to throw some ice-cold water on this:
>
>  1.  It's highly unlikely that we will ever support the x4500 --  
> only the
>  x4540 is a real possibility.


And to warm things up a bit: there's already an upgrade path from the
x4500 to the x4540 so that would be required before any upgrade to the
equivalent of the Sun Storage 7210.

Adam

--
Adam Leventhal, Fishworkshttp://blogs.sun.com/ahl

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

Re: [zfs-discuss] OpenStorage GUI

[Adam Leventhal]

> Is this software available for people who already have thumpers?

We're considering offering an upgrade path for people with existing
thumpers. Given the feedback we've been hearing, it seems very likely
that we will. No word yet on pricing or availability.

Adam

--
Adam Leventhal, Fishworkshttp://blogs.sun.com/ahl

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

[zfs-discuss] OpenStorage GUI

[Scott Williamson]

Hi,

Is this 
softwareavailable
for people who already have thumpers?

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

Re: [zfs-discuss] ZFS on Fit-PC Slim?

[Vincent Fox]

> Solaris 10 U6 and Solaris Express
> Community Edition can both be installed in text mode.
> Nexenta and Belenix will also run on this machine
> with 512MB. Nexenta will probably be what you want if
> you are saying you are running ubuntu on this
> box.

Perhaps I wasn't sufficiently clear.

I *installed* in text mode of course.

However on booting the system it wants to start graphics.

Why do people keep beating on this point?

Yes I could have done various workarounds to disable graphics.  My point was 
during the INSTALL there is no way to specify you just want a plain text 
console for the later running system.  You are expected to boot, then login and 
disable it later.  Is that clear now?
-- 
This message posted from opensolaris.org
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

[zfs-discuss] ZFS capable GRUB install from within Linux?

[Mario Goebbels]

Is it possible to install a GRUB that can boot a ZFS root, but installing it 
from within Linux?

I was planning on getting a new unmanaged dedicated server, which however only 
comes with Linux preinstalled. The thing has a software RAID1, so the cunning 
plan was to break up the RAID1, install VirtualBox RDP and install OpenSolaris 
onto the freed drive. The only roadblock is GRUB at that point, which would 
still reside on the existing Linux disk. Once OpenSolaris has been booted, I 
can recreate the mirror and do a proper GRUB install.

Getting a IP-KVM attached to change BIOS settings is kind of the cost I'd want 
to avoid here.

Any ideas?

Regards,
-mg
-- 
This message posted from opensolaris.org
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

[zfs-discuss] Autoreplace not working

[Bluenix]

First of all: I'm very new to Opensolaris. After reading a lot, I'm mostly 
interested in the autoreplace function of zfs. I've got a Sun Fire X2100 M2 and 
now run Opensolaris on it.

I've created a mirror in the rpool pool of 2 sata drives. I've also enable the 
autoreplace function of that pool in zfs.
When the pool is up and running, and I physically remove one of the two hard 
drives, the '[b]zpool status[/b]' command shows the drive has been removed and 
the mirror/pool is now degraded. However, when I plug the drive back in and run 
'zpool status' again, the drive remains [b]"REMOVED"[/b].

The output of dmesg is showing: 
[b]"Application(s) accessing previously attached SATA device have to release it 
before newly inserted device can be made accessible."[/b]

As far as I understand, the 'autoreplace' function should take care of the 
drive replacement, so no manual intervention should be needed.
Does anybody have an idea how to accomplish that, or what can be wrong now?
Your help is greatly appreciated!
-- 
This message posted from opensolaris.org
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

[zfs-discuss] Q. ZFS requires freespace in file system ?

[Wolfgang Pungartnik]

Hi,

i recently encountered a database crash on ZFS with MaxDB and
Solaris 10. The reason was that the ZFS file system was 100% full
(quota, reservation set and exceeded). Unfortunately using "df -k" and
counting the bytes only 860GB out of 870GB were used for files. No
hidden files, no ZFS snapshots or anything else i could see. The big
question is who's using the missing 10GB.

Does ZFS requires a reserved amount of space per file system ? I know
that due to COW ZFS requires a little bit more spaces until the update
is done. But we are still missing 10GB.

Any ideas ?


Thx.

++
sapph10:ph1adm 10% ls -laR
.:
total 1824522213
drwxr-xr-x   2 sdb  sdba  45 Sep  5 17:10 .
drwxrwxr-x   6 sdb  sdba  11 Jul  2 09:05 ..
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0001
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0002
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0003
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0004
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0005
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0006
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0007
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0008
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0009
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0010
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0011
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0012
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0013
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0014
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0015
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0016
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0017
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0018
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0019
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0020
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0021
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0022
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0023
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0024
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0025
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0026
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0027
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0028
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0029
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0030
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0031
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0032
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0033
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0034
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0035
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0036
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0037
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0038
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0039
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0040
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0041
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0042
-rw-rw   1 sdb  sdba 21474844672 Sep 30 09:47 DISKD0043

sapph10:ph1adm 12% df -k
Filesystemkbytesused   avail capacity  Mounted on
/ 976367  278453  68815129%/
/dev  976367  278453  68815129%/dev
/lib 15493250 4193500 1114481828%/lib
/platform15493250 4193500 1114481828%/platform
/sbin15493250 4193500 1114481828%/sbin
/usr 15493250 4193500 1114481828%/usr
proc   0   0   0 0%/proc
ctfs   0   0   0 0%/system/contract
mnttab 0   0   0 0%/etc/mnttab
objfs  0   0   0 0%/system/object
swap 120834608 328 120834280 1%/etc/svc/volatile
fd 0   0   0 0%/dev/fd
swap 1048576  48 1048528 1%/tmp
swap 120834320  40 120834280 1%/var/run
ph1_sap  1023934464  24 106873770 1%/ph1_sap
ph1_saplog   51351552  24 25859951 1%/ph1_saplog
ph1_sap/sapdb2097152 2097236   0   100%/sapdb
ph1_sap/ph1  4194304  151045 4043259 4%/sapdb/PH1
ph1_sap/sapdata1 912261120 9