Re: [OpenIndiana-discuss] Looking for concise doc on REALLY ALL the steps the live install actually performs

2013-03-17 Thread Nathan Kunkee

On 03/16/13 12:01 PM, Hans J. Albertsson wrote:

Looking for concise doc on REALLY ALL the steps the live install
actually performs.

Yeah, really, just for a checklist.

Tried looking in various places. Found nothing. Maybe just me needing a
new brain.

___
OpenIndiana-discuss mailing list
OpenIndiana-discuss@openindiana.org
http://openindiana.org/mailman/listinfo/openindiana-discuss




If you haven't found it, check out the different minimal OpenSolaris 
projects. The one I found had some minimal install scripts that I think 
are what you want:


http://alexeremin.blogspot.com/2009/04/minimal-opensolaris-network-install.html
http://alexeremin.blogspot.com/2008/12/minimum-opensolaris-200811-install.html

Hope that helps,
Nathan


___
OpenIndiana-discuss mailing list
OpenIndiana-discuss@openindiana.org
http://openindiana.org/mailman/listinfo/openindiana-discuss


Re: [OpenIndiana-discuss] Tape backup

2013-03-17 Thread dormitionsk...@hotmail.com
After the embezzlement debacle, and a couple of complete turnovers (except for 
me) of the IS staff, my (new) boss and I went to a class specifically about 
backups.  They were saying that not only do we need to keep your normal daily, 
weekly and monthly backups; we should also keep snapshots of our data for years 
upon years.

They were saying we should do this not only for financial data, but for any 
data, because you might have a disgruntled employee decide to monkey with your 
data, and it may take years upon years to discover it.  While it would most 
likely be significantly more difficult to restore that old data -- especially 
if the tampering were done to a database -- at least you would have some 
opportunity to recover your uncorrupted information.

Just food for thought.



On Mar 17, 2013, at 4:36 PM, Doug Hughes wrote:

> On 3/17/2013 6:23 PM, Reginald Beardsley wrote:
>> Tape as an archival medium has significant issues.  Reading poorly stored 
>> tapes is a "one try" proposition w/ no assurance of success.  The first high 
>> volume commercial application for digital tape was seismic data acquisition 
>> for the oil industry.  The oil companies had very detailed  cleaning and 
>> retensioning schedules w/ a large staff to perform them on the tape 
>> archives. Absent that level of care, reading old tapes is very difficult and 
>> requires great skill.  Old tape is NOT fun to work with.
>> 
>> High capacity tape drives and tapes are not cheap either.  Blank LTO tape is 
>> almost as expensive as SATA disk. A ZFS based remote replicating server 
>> using triple parity RAIDZ is probably cheaper than tape.  For extremely 
>> large volumes and long archival periods, optical tape is probably the best 
>> choice.  But then you're probably working for the government.
>> 
>> I would strongly urge comparing the cost of a ZFS backup server w/ daily 
>> snapshots to the cost of conventional tape backup. I think you'll be quite 
>> surprised at the implications.
>> 
> 
> reading old disks is a just as significant if not more of an issue, by my 
> estimation. Try to find a machine that you can do low voltage differential 
> disks with these days. That was only 15 years ago. what about SMD? The 
> controllers keep changing over time. Also, after a disk has been in use for a 
> significant period of time (say years), the lubrication on the platters tends 
> to evaporate a little bit so that when you leave it off for a long period of 
> time (days/weeks) it will stick to the heads and the platters. Tape doesn't 
> have that issue.
> 
> An LTO5 tape is about $30 each, better in quantity. 1.5TB, more depending 
> upon compression. That's an enterprise quality tape with much longer shelf 
> life than a cheap deskstar disk if cared for properly. Even the cheap 1TB 
> disks are $70. That's almost a 3X advantage. There's still a place for tape 
> for archival, and yes, you do have to care for it properly, just like you 
> have to care for everything else. But, cared for properly, tape should still 
> outlive disk.
> 
> But, used it in its proper place. zfs snapshots make a lot of sense for 
> online backups!
> 
> 
> 
> ___
> OpenIndiana-discuss mailing list
> OpenIndiana-discuss@openindiana.org
> http://openindiana.org/mailman/listinfo/openindiana-discuss
> 


___
OpenIndiana-discuss mailing list
OpenIndiana-discuss@openindiana.org
http://openindiana.org/mailman/listinfo/openindiana-discuss


Re: [OpenIndiana-discuss] Tape backup

2013-03-17 Thread Doug Hughes

On 3/17/2013 6:23 PM, Reginald Beardsley wrote:

Tape as an archival medium has significant issues.  Reading poorly stored tapes is a 
"one try" proposition w/ no assurance of success.  The first high volume 
commercial application for digital tape was seismic data acquisition for the oil 
industry.  The oil companies had very detailed  cleaning and retensioning schedules w/ a 
large staff to perform them on the tape archives. Absent that level of care, reading old 
tapes is very difficult and requires great skill.  Old tape is NOT fun to work with.

High capacity tape drives and tapes are not cheap either.  Blank LTO tape is 
almost as expensive as SATA disk. A ZFS based remote replicating server using 
triple parity RAIDZ is probably cheaper than tape.  For extremely large volumes 
and long archival periods, optical tape is probably the best choice.  But then 
you're probably working for the government.

I would strongly urge comparing the cost of a ZFS backup server w/ daily 
snapshots to the cost of conventional tape backup. I think you'll be quite 
surprised at the implications.



reading old disks is a just as significant if not more of an issue, by 
my estimation. Try to find a machine that you can do low voltage 
differential disks with these days. That was only 15 years ago. what 
about SMD? The controllers keep changing over time. Also, after a disk 
has been in use for a significant period of time (say years), the 
lubrication on the platters tends to evaporate a little bit so that when 
you leave it off for a long period of time (days/weeks) it will stick to 
the heads and the platters. Tape doesn't have that issue.


An LTO5 tape is about $30 each, better in quantity. 1.5TB, more 
depending upon compression. That's an enterprise quality tape with much 
longer shelf life than a cheap deskstar disk if cared for properly. Even 
the cheap 1TB disks are $70. That's almost a 3X advantage. There's still 
a place for tape for archival, and yes, you do have to care for it 
properly, just like you have to care for everything else. But, cared for 
properly, tape should still outlive disk.


But, used it in its proper place. zfs snapshots make a lot of sense for 
online backups!




___
OpenIndiana-discuss mailing list
OpenIndiana-discuss@openindiana.org
http://openindiana.org/mailman/listinfo/openindiana-discuss


Re: [OpenIndiana-discuss] Tape backup

2013-03-17 Thread Reginald Beardsley
Tape as an archival medium has significant issues.  Reading poorly stored tapes 
is a "one try" proposition w/ no assurance of success.  The first high volume 
commercial application for digital tape was seismic data acquisition for the 
oil industry.  The oil companies had very detailed  cleaning and retensioning 
schedules w/ a large staff to perform them on the tape archives. Absent that 
level of care, reading old tapes is very difficult and requires great skill.  
Old tape is NOT fun to work with.

High capacity tape drives and tapes are not cheap either.  Blank LTO tape is 
almost as expensive as SATA disk. A ZFS based remote replicating server using 
triple parity RAIDZ is probably cheaper than tape.  For extremely large volumes 
and long archival periods, optical tape is probably the best choice.  But then 
you're probably working for the government.

I would strongly urge comparing the cost of a ZFS backup server w/ daily 
snapshots to the cost of conventional tape backup. I think you'll be quite 
surprised at the implications.


Have Fun!
Reg



___
OpenIndiana-discuss mailing list
OpenIndiana-discuss@openindiana.org
http://openindiana.org/mailman/listinfo/openindiana-discuss


Re: [OpenIndiana-discuss] What to do with 4k sector SATA drives... All fixes are for SCSI drives only!!??

2013-03-17 Thread Reginald Beardsley



--- On Sun, 3/17/13, Hans J Albertsson  wrote:

> From: Hans J Albertsson 
> Subject: Re: [OpenIndiana-discuss] What to do with 4k sector SATA drives... 
> All fixes are for SCSI drives only!!??
> To: "Discussion list for OpenIndiana" 
> Date: Sunday, March 17, 2013, 1:31 PM
> I read somewhere that you can
> 
> Set sata to ahci in bios.
> Boot a live dvd
> In there  i can just do a zpool import -f rpool.
> Could this be true? Really?

4k sector SATA drives work fine w/ the sd.conf configured properly.  Make sure 
you read George Wilson's blog on the subject:

http://blog.delphix.com/gwilson/2012/11/15/4k-sectors-and-zfs/

In particular, the part about using cfgadm to reattach and reconfigure the the 
disk.  4k USB disks don't appear to work properly at present, though they do 
work, albeit slowly.


Some operations are a bit like a "Towers of Hanoi" problem.  You might get some 
insight into the situation from my exposition on building a bootable, 3-4 drive 
RAIDZ using the GUI installer.

http://wiki.openindiana.org/oi/Building+a+small+NAS+using+OI

That's gotten a bit confused between my edits and Jim's.  I think all the info 
is there, just not very well organized.  I had lots of trouble until I wrote a 
step by step outline (one sheet of paper).  That was sufficient to get me there.

Probably the safest approach would be to remove the existing drives, install 
the 4k drive and install onto it.  Then replace the old drives and import the 
old pools, copy the data and then reconfigure the interfaces and build new 
pools on the old drives.

You *might* be able to form a mirror w/ 512 and 4k drives,  but I'd test that 
w/ a fresh install onto a 512 sector disk rather than a system I wanted to keep 
stuff from.  There are warnings about that not working, but it might work as a 
migration pathway.

For example:

install and configure 4k drive

form mirror w/ 512 sector drive

wait for it to resilver

detach 512 sector drive from mirror

install another 4k drive

form mirror w/ pair of 4k drives


NB: format(1m) will create a cylinder that doesn't start on a 4k boundary, 
however, it would appear that ZFS knows how to avoid getting in trouble in such 
cases.

Mostly I think we need better documentation for the sorts of things experienced 
admins expect to be able to do.  I'd love to see a clear statement of what the 
gui installer does.  So far as I know, the text installer script is as close to 
a statement of how to install as there is.  It's not easy to follow though.

Have Fun!
Reg

___
OpenIndiana-discuss mailing list
OpenIndiana-discuss@openindiana.org
http://openindiana.org/mailman/listinfo/openindiana-discuss


Re: [OpenIndiana-discuss] What to do with 4k sector SATA drives... All fixes are for SCSI drives only!!??

2013-03-17 Thread Sašo Kiselkov
On 03/17/2013 04:06 PM, Hans J. Albertsson wrote:
> Your counter-question baffles me a bit... but:
> 
> Can you point at any part of that wiki page that actually deals with how
> to produce a 4kblocksize pool on a SATA, not a SCSI, drive that is
> actually 4k physical blocksize but reports having 512 bytes blocks.

I have a bunch of 4k WD Red drives, they work just fine (zpool created
with ashift=13 automatically).

> The trick of editing sd.conf shouldn't work on SATA drives, since
> they're not scsi, and sd.conf deals with SCSI harddisks plus some ATAPI
> CD/DVDs and such..
> 
> Or am I missing something?

It will work, native SATA devices are handled through sd, since the ATA
translation layer makes them appear to sd as if they were speaking SCSI.

> P.S. I can't just test right now, since the machines are miles away,
> otherwise I would have just plugged the drives in and tested. I'd like
> to know more before spending a couple of hours on a possibly meaningless
> test.

Make sure your SATA controller is in AHCI (native SATA) mode, no IDE
(emulated PATA).

--
Saso

___
OpenIndiana-discuss mailing list
OpenIndiana-discuss@openindiana.org
http://openindiana.org/mailman/listinfo/openindiana-discuss


Re: [OpenIndiana-discuss] What to do with 4k sector SATA drives... All fixes are for SCSI drives only!!??

2013-03-17 Thread Hans J Albertsson
I read somewhere that you can

Set sata to ahci in bios.
Boot a live dvd
In there  i can just do a zpool import -f rpool.
Could this be true? Really?


Skickat från min Samsung Mobil

 Originalmeddelande 
Från: "Hans J. Albertsson"  
Datum:  
Till: Discussion list for OpenIndiana  
Rubrik: Re: [OpenIndiana-discuss] What to do with 4k sector SATA drives... All 
fixes are for SCSI drives only!!?? 
 
i think I've just got a problem because of machine age!

if I can muster enough courage  ok, well, being retired, what can 
harm me now?
I have my sata controllers in that machine in ide mode... gaah..

There was a problem in Solaris 11/Opensolaris around builds << 97 or so 
that meant I had to do that, or else my machines would reboot at random. 
I think I know the problem is gone since b134 at least, and I almost 
remember,  I think, doing some kind of reconfiguring and recapturing the 
rpool from a live CD on another box on account of that problem.
It's years ago so my memory of the event has been almost erased
I hope that will work without reinstalliing. All the devices and dev 
trees naming looks very different.

Will I have to export/backup the entire thing to another disk, and 
reconfigure and import/recv/something back, or can I  just change the 
SATA mode and recreate the devices and dev trees from a live cd w/o 
loosing the rpool?

I realised this when I happened to check things on another machine, by 
mistake, a machine with proper settings and seeing very different 
behaviour in the inquire command.

My  face is crimson!


Reinstall needed???  St!!!


On 2013-03-17 16:38, Jan Owoc wrote:
> On Sun, Mar 17, 2013 at 9:06 AM, Hans J. Albertsson
>  wrote:
>>> On Sun, Mar 17, 2013 at 8:30 AM, Hans J. Albertsson
>>>  wrote:
 When I look around in all the various places where the 4k blocksize issue
 is
 discussed, it turns out all the advice only ever deals with SCSI or SAS
 drives.
>>> After reading this page [1], what specific further questions do you
>>> have? The goal is to have that page a catch-all for your kind of
>>> question.
>>>
>>> [1] http://wiki.illumos.org/display/illumos/ZFS+and+Advanced+Format+disks
>> Your counter-question baffles me a bit... but:
>>
>> Can you point at any part of that wiki page that actually deals with how to
>> produce a 4kblocksize pool on a SATA, not a SCSI, drive that is actually 4k
>> physical blocksize but reports having 512 bytes blocks.
>>
>> The trick of editing sd.conf shouldn't work on SATA drives, since they're
>> not scsi, and sd.conf deals with SCSI harddisks plus some ATAPI CD/DVDs and
>> such..
> >From what I understood, both SATA and SCSI drives can be modified
> using sd.conf, but I don't have any drives I can test at the moment,
> so either of us could be wrong ;-).
>
> However, there is a sample listing of drives that can be entered into
> sd.conf [2], and many (all?) of them are SATA, so that is where I
> presumed it would work with SATA. Let me know if it doesn't, as I will
> may to work with such drives in the near future.
>
> [2] 
> http://wiki.illumos.org/display/illumos/List+of+sd-config-list+entries+for+Advanced-Format+drives
>
> Cheers,
> Jan
>
> ___
> OpenIndiana-discuss mailing list
> OpenIndiana-discuss@openindiana.org
> http://openindiana.org/mailman/listinfo/openindiana-discuss


___
OpenIndiana-discuss mailing list
OpenIndiana-discuss@openindiana.org
http://openindiana.org/mailman/listinfo/openindiana-discuss
___
OpenIndiana-discuss mailing list
OpenIndiana-discuss@openindiana.org
http://openindiana.org/mailman/listinfo/openindiana-discuss


Re: [OpenIndiana-discuss] What to do with 4k sector SATA drives... All fixes are for SCSI drives only!!??

2013-03-17 Thread Hans J. Albertsson

i think I've just got a problem because of machine age!

if I can muster enough courage  ok, well, being retired, what can 
harm me now?

I have my sata controllers in that machine in ide mode... gaah..

There was a problem in Solaris 11/Opensolaris around builds << 97 or so 
that meant I had to do that, or else my machines would reboot at random. 
I think I know the problem is gone since b134 at least, and I almost 
remember,  I think, doing some kind of reconfiguring and recapturing the 
rpool from a live CD on another box on account of that problem.

It's years ago so my memory of the event has been almost erased
I hope that will work without reinstalliing. All the devices and dev 
trees naming looks very different.


Will I have to export/backup the entire thing to another disk, and 
reconfigure and import/recv/something back, or can I  just change the 
SATA mode and recreate the devices and dev trees from a live cd w/o 
loosing the rpool?


I realised this when I happened to check things on another machine, by 
mistake, a machine with proper settings and seeing very different 
behaviour in the inquire command.


My  face is crimson!


Reinstall needed???  St!!!


On 2013-03-17 16:38, Jan Owoc wrote:

On Sun, Mar 17, 2013 at 9:06 AM, Hans J. Albertsson
 wrote:

On Sun, Mar 17, 2013 at 8:30 AM, Hans J. Albertsson
 wrote:

When I look around in all the various places where the 4k blocksize issue
is
discussed, it turns out all the advice only ever deals with SCSI or SAS
drives.

After reading this page [1], what specific further questions do you
have? The goal is to have that page a catch-all for your kind of
question.

[1] http://wiki.illumos.org/display/illumos/ZFS+and+Advanced+Format+disks

Your counter-question baffles me a bit... but:

Can you point at any part of that wiki page that actually deals with how to
produce a 4kblocksize pool on a SATA, not a SCSI, drive that is actually 4k
physical blocksize but reports having 512 bytes blocks.

The trick of editing sd.conf shouldn't work on SATA drives, since they're
not scsi, and sd.conf deals with SCSI harddisks plus some ATAPI CD/DVDs and
such..

>From what I understood, both SATA and SCSI drives can be modified
using sd.conf, but I don't have any drives I can test at the moment,
so either of us could be wrong ;-).

However, there is a sample listing of drives that can be entered into
sd.conf [2], and many (all?) of them are SATA, so that is where I
presumed it would work with SATA. Let me know if it doesn't, as I will
may to work with such drives in the near future.

[2] 
http://wiki.illumos.org/display/illumos/List+of+sd-config-list+entries+for+Advanced-Format+drives

Cheers,
Jan

___
OpenIndiana-discuss mailing list
OpenIndiana-discuss@openindiana.org
http://openindiana.org/mailman/listinfo/openindiana-discuss



___
OpenIndiana-discuss mailing list
OpenIndiana-discuss@openindiana.org
http://openindiana.org/mailman/listinfo/openindiana-discuss


Re: [OpenIndiana-discuss] Mapping LSI SAS Target numbers to disk SAS WWNs (Log Info 0x31120303)

2013-03-17 Thread Richard Elling
On Mar 16, 2013, at 5:02 PM, Richard Elling  
wrote:

> there is a way to get this info from mdb... I added a knowledge base article 
> on this at Nexenta a few years ago, lemme see if I can dig it up from my 
> archives…

And the winner is:
echo "::mptsas -t" | mdb -k
mptsas_t inst ncmds suspend  power




ff07037850000 0   0 ON=D0


 
The SCSI target information

 
devhdl 12, sasaddress 5000c5002128f332, phymask ff,devinfo 401
 
throttle 20, dr_flag 0, m_t_ncmds 0

 
devhdl 11, sasaddress 5000c50021295cb2, phymask ff,devinfo 401
 
throttle 20, dr_flag 0, m_t_ncmds 0

mptsas reports the devhdl and you can cross-reference to the SAS address (WWN)
 -- richard

> 
> On Mar 15, 2013, at 11:22 PM, "Richard L. Hamilton"  wrote:
> 
>> Running on something older (SXCE snv_97 on SPARC, or thereabouts), with an 
>> LSI SAS controller using the mpt driver:
>> 
>> #! /bin/ksh
>> for dev in $( find /devices -type c -name 'sd@*:a,raw'|grep LSILogic,sas)
>> do
>>   echo $dev
>>   prtconf -v ${dev}|grep id1,
>> done
>> 
>> 
>> produced the following output
>> 
>> /devices/pci@8,60/LSILogic,sas@1/sd@0,0:a,raw
>>   value='id1,sd@n5000c5000682fbef'
>> /devices/pci@8,60/LSILogic,sas@1/sd@1,0:a,raw
>>   value='id1,sd@n5000c5000682fb0f'
>> /devices/pci@8,60/LSILogic,sas@1/sd@2,0:a,raw
>>   value='id1,sd@n5000c500104a589f'
>> /devices/pci@8,60/LSILogic,sas@1/sd@3,0:a,raw
>>   value='id1,sd@n5000c500104aa29b'
>> /devices/pci@8,60/LSILogic,sas@1/sd@4,0:a,raw
>>   value='id1,sd@n5000c50041e49faf'
>> /devices/pci@8,60/LSILogic,sas@1/sd@5,0:a,raw
>>   value='id1,sd@n5000c50041d6455f'
>> /devices/pci@8,60/LSILogic,sas@1/sd@6,0:a,raw
>>   value='id1,sd@n5000c50041ddfebf'
>> /devices/pci@8,60/LSILogic,sas@1/sd@7,0:a,raw
>>   value='id1,sd@n5000c50041e2be0b'
>> 
>> I assume the lines beginning with value='id1,sd@n have the WWN following
>> that, right?
>> 
>> This is a 1068, that I lucked into on eBay for my SB2K.
>> 
>> If this approach could be adapted or cleaned up a bit for what you want,
>> the advantage is that it doesn't require lsiutil, but just uses plain old
>> prtconf.  Presumably someone fluent in perl could figure out a way to
>> parse and format it more elegantly.
>> 
>> 
>> 
>> On Mar 15, 2013, at 11:20 PM, Peter Tripp wrote:
>> 
>>> No questions...just information for how to translate a Target ID to a SAS 
>>> WWN on LSI MPT2 SAS2 controllers under Illumos/Solaris.  My apologies for 
>>> cross posting or if this is old hat, but I've been running an LSI SAS2 
>>> controller with SATA disks and had my logs fill up with repeated cryptic 
>>> entries, but never found a troubleshooting strategy until tonight. Under 
>>> load (scrub) the following will repeatedly show up: 
>>> 
>>> scsi: [ID 365881 kern.info] /pci@0,0/pci15ad,7a0@15/pci1000,3030@0 
>>> (mpt_sas0):
>>>   Log info 0x31120311 received for target 20.
>>>   scsi_status=0x0, ioc_status=0x804b, scsi_state=0xc
>>> scsi: [ID 243001 kern.warning] WARNING: 
>>> /pci@0,0/pci15ad,7a0@15/pci1000,3030@0 (mpt_sas0):
>>>   mptsas_handle_event_sync: IOCStatus=0x8000, IOCLogInfo=0x31120303
>>> 
>>> I assumed this was due to a disk failing or a cabling problem (my LSI SAS 
>>> 9200-16e SAS_2116 has directly cabled WD RE4 SATA disks) but which disk is 
>>> Target 20? I took the time to label my drive caddies with their WWN to ease 
>>> trouble shooting, but was never able to figure out how to translate Target 
>>> ID into a SAS WWN so I narrow down my troubleshooting to a single 
>>> deviceuntil now.
>>> 
>>> Run lsituil.i386 as root
>>> (LSIUtil Kit 1.63.zip\Solaris\lsiutil.i386 from 
>>> http://www.juhonkoti.net/media/LSIUTIL-1.63.zip )
>>> * Select your MPT device
>>> * e (Enable expert mode in menus)
>>> * 20 (Diagnostics)
>>> * 1 (Inquiry Test)
>>> * 0 (Bus 0)
>>> * XX (Target number from above, Target 20 for me)
>>> * 0 (Lun 0)
>>> * 83 or 80 (SAS WWN and Disk Serial Number respectively)
>>> Output:
>>> 
>>> VPD Page:  [00-FF or RETURN for normal Inquiry] 83
>>> B___T___L  Page
>>> 0  20   0   83
>>> 16 bytes of Inquiry Data returned
>>>  : 00 83 00 0c 01 03 00 08 50 01 4e e2 b1 65 f5 d7P N  e  
>>> 
>>> VPD Page:  [00-FF or RETURN for normal Inquiry] 80
>>> B___T___L  Page
>>> 0  20   0   80
>>> 24 bytes of Inquiry Data returned
>>>  : 00 80 00 14 20 20 20 20 20 57 44 2d 57 4d 41 59 WD-WMAY
>>> 0010 : 30 34 33 32 38 34 38 3904328489
>>> 
>>> The second half of the hex string from 83h is your WWN (50014ee2b165f5d7 
>>> above), and ASCII from 80h (WMAY04328489 above) the serial number.  With 
>>> that info one can check the cabling path or repl

Re: [OpenIndiana-discuss] What to do with 4k sector SATA drives... All fixes are for SCSI drives only!!??

2013-03-17 Thread Jan Owoc
On Sun, Mar 17, 2013 at 9:06 AM, Hans J. Albertsson
 wrote:
>> On Sun, Mar 17, 2013 at 8:30 AM, Hans J. Albertsson
>>  wrote:
>>>
>>> When I look around in all the various places where the 4k blocksize issue
>>> is
>>> discussed, it turns out all the advice only ever deals with SCSI or SAS
>>> drives.
>>
>> After reading this page [1], what specific further questions do you
>> have? The goal is to have that page a catch-all for your kind of
>> question.
>>
>> [1] http://wiki.illumos.org/display/illumos/ZFS+and+Advanced+Format+disks
>
> Your counter-question baffles me a bit... but:
>
> Can you point at any part of that wiki page that actually deals with how to
> produce a 4kblocksize pool on a SATA, not a SCSI, drive that is actually 4k
> physical blocksize but reports having 512 bytes blocks.
>
> The trick of editing sd.conf shouldn't work on SATA drives, since they're
> not scsi, and sd.conf deals with SCSI harddisks plus some ATAPI CD/DVDs and
> such..

>From what I understood, both SATA and SCSI drives can be modified
using sd.conf, but I don't have any drives I can test at the moment,
so either of us could be wrong ;-).

However, there is a sample listing of drives that can be entered into
sd.conf [2], and many (all?) of them are SATA, so that is where I
presumed it would work with SATA. Let me know if it doesn't, as I will
may to work with such drives in the near future.

[2] 
http://wiki.illumos.org/display/illumos/List+of+sd-config-list+entries+for+Advanced-Format+drives

Cheers,
Jan

___
OpenIndiana-discuss mailing list
OpenIndiana-discuss@openindiana.org
http://openindiana.org/mailman/listinfo/openindiana-discuss


Re: [OpenIndiana-discuss] What to do with 4k sector SATA drives... All fixes are for SCSI drives only!!??

2013-03-17 Thread Hans J. Albertsson

Your counter-question baffles me a bit... but:

Can you point at any part of that wiki page that actually deals with how 
to produce a 4kblocksize pool on a SATA, not a SCSI, drive that is 
actually 4k physical blocksize but reports having 512 bytes blocks.


The trick of editing sd.conf shouldn't work on SATA drives, since 
they're not scsi, and sd.conf deals with SCSI harddisks plus some ATAPI 
CD/DVDs and such..


Or am I missing something?

P.S. I can't just test right now, since the machines are miles away, 
otherwise I would have just plugged the drives in and tested. I'd like 
to know more before spending a couple of hours on a possibly meaningless 
test.




On 2013-03-17 15:48, Jan Owoc wrote:

Hi Hans,

On Sun, Mar 17, 2013 at 8:30 AM, Hans J. Albertsson
 wrote:

When I look around in all the various places where the 4k blocksize issue is
discussed, it turns out all the advice only ever deals with SCSI or SAS
drives.

After reading this page [1], what specific further questions do you
have? The goal is to have that page a catch-all for your kind of
question.

[1] http://wiki.illumos.org/display/illumos/ZFS+and+Advanced+Format+disks

Cheers,
Jan

___
OpenIndiana-discuss mailing list
OpenIndiana-discuss@openindiana.org
http://openindiana.org/mailman/listinfo/openindiana-discuss



___
OpenIndiana-discuss mailing list
OpenIndiana-discuss@openindiana.org
http://openindiana.org/mailman/listinfo/openindiana-discuss


Re: [OpenIndiana-discuss] What to do with 4k sector SATA drives... All fixes are for SCSI drives only!!??

2013-03-17 Thread Jan Owoc
Hi Hans,

On Sun, Mar 17, 2013 at 8:30 AM, Hans J. Albertsson
 wrote:
> When I look around in all the various places where the 4k blocksize issue is
> discussed, it turns out all the advice only ever deals with SCSI or SAS
> drives.

After reading this page [1], what specific further questions do you
have? The goal is to have that page a catch-all for your kind of
question.

[1] http://wiki.illumos.org/display/illumos/ZFS+and+Advanced+Format+disks

Cheers,
Jan

___
OpenIndiana-discuss mailing list
OpenIndiana-discuss@openindiana.org
http://openindiana.org/mailman/listinfo/openindiana-discuss


[OpenIndiana-discuss] What to do with 4k sector SATA drives... All fixes are for SCSI drives only!!??

2013-03-17 Thread Hans J. Albertsson
When I look around in all the various places where the 4k blocksize 
issue is discussed, it turns out all the advice only ever deals with 
SCSI or SAS drives.


So many of the easy, straightforward recipies won't work for SATA 
drives. What should I do??


I have the WD EARS drives which misrepresent their physical blocksize.

The zpool-12 binary I have found on the net won't work: it gets ld 
errors when starting.


HELP!!!





___
OpenIndiana-discuss mailing list
OpenIndiana-discuss@openindiana.org
http://openindiana.org/mailman/listinfo/openindiana-discuss


[OpenIndiana-discuss] pkgrepo error

2013-03-17 Thread Serge Fonville
Hi,

I have been running an IPS Server for a while now, without serious issues
Since some time I am getting errors when using it.
When I run *pkgrepo refresh -s /export/repo/ *I get a stack trace

> Initiating repository refresh.
> Traceback (most recent call last):
>   File "/usr/bin/pkgrepo", line 1226, in handle_errors
> __ret = func(*args, **kwargs)
>   File "/usr/bin/pkgrepo", line 1202, in main_func
> return func(conf, pargs)
>   File "/usr/bin/pkgrepo", line 984, in subcmd_refresh
> do_refresh(xport, xpub)
>   File "/usr/bin/pkgrepo", line 969, in do_refresh
> xport.publish_refresh(xpub)
>   File
> "/usr/lib/python2.6/vendor-packages/pkg/client/transport/transport.py",
> line 432, in wrapper
> return f(instance, *fargs, **f_kwargs)
>   File
> "/usr/lib/python2.6/vendor-packages/pkg/client/transport/transport.py",
> line 2568, in publish_refresh
> d.publish_refresh(header=header, pub=pub)
>   File "/usr/lib/python2.6/vendor-packages/pkg/client/transport/repo.py",
> line 1584, in publish_refresh
> refresh_index=True)
>   File "/usr/lib/python2.6/vendor-packages/pkg/server/repository.py", line
> 2703, in add_content
> rstore.add_content(refresh_index=refresh_index)
>   File "/usr/lib/python2.6/vendor-packages/pkg/server/repository.py", line
> 1067, in add_content
> build_index=refresh_index, incremental=True)
>   File "/usr/lib/python2.6/vendor-packages/pkg/server/repository.py", line
> 676, in __rebuild
> self.__refresh_index()
>   File "/usr/lib/python2.6/vendor-packages/pkg/server/repository.py", line
> 695, in __refresh_index
> return self.__run_update_index()
>   File "/usr/lib/python2.6/vendor-packages/pkg/server/repository.py", line
> 1655, in __run_update_index
> self.__update_searchdb_unlocked(fmris_to_index)
>   File "/usr/lib/python2.6/vendor-packages/pkg/server/repository.py", line
> 1011, in __update_searchdb_unlocked
> index_inst.server_update_index(fmris)
>   File "/usr/lib/python2.6/vendor-packages/pkg/indexer.py", line 774, in
> server_update_index
> tmp_index_dir)
>   File "/usr/lib/python2.6/vendor-packages/pkg/indexer.py", line 726, in
> _generic_update_index
> self._update_index(dicts, tmp_index_dir)
>   File "/usr/lib/python2.6/vendor-packages/pkg/indexer.py", line 601, in
> _update_index
> assert len(next_new_tok) > 0
> AssertionError
>
> pkgrepo: This is an internal error in pkg(5) version 35e1caa63de2.  Please
> log a
> Service Request about this issue including the information above and this
> message.

When I run pkg search I get an error

> pkg: Some repositories failed to respond appropriately:
> sergefonville.nl:
> http protocol error: code: 503 reason: Service Unavailable
> URL: '
> http://192.168.1.11/sergefonville.nl/search/1/False_2_None_None_%3A%3A%3Aphp
> '.

What do I need to do to fix this?

Thanks in advance.

Kind regards/met vriendelijke groet,

Serge Fonville

http://www.sergefonville.nl

Convince Microsoft!
They need to add TRUNCATE PARTITION in SQL Server
https://connect.microsoft.com/SQLServer/feedback/details/417926/truncate-partition-of-partitioned-table
___
OpenIndiana-discuss mailing list
OpenIndiana-discuss@openindiana.org
http://openindiana.org/mailman/listinfo/openindiana-discuss