[osol-discuss] slow CIFS performance

[Bo Gao]

Hi,

my set up is SXCE 130, gigabyte EP45-UD3P, Intel E2180 overclocked to 3.6Ghz, 
2GB memory, one 60GB 7200rpm 2.5inch Hitachi for the operating system, 3 1.5TB 
drive for RaidZ storage. Network I have two onboard realtek gigabyte and one 
Intel 1000MT PCI-E card.

I have observed copying files over the network under CIFS has been around 
10MB/sec; I disabled the onboard realtek cards and used the Intel one, same 
result; I copied file from the raidz to the Hitachi drive (within the same 
machine), I have 35MB/s read and 25MB/s write.

Has anyone encountered similar issue before? Can anyone provide a bit insight 
of what to look for?

Thanks.
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Why are prices for Solaris 10 OS support not on Oracles Web pages?

[Go OpenSolaris]

First of all, thank you all for the response. Just hearing some response makes 
you feel better.

Even our sales contacts cannot yet answer clearly on non-Sun H/w support 
unfortunately. So we are in limbo until then.
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] So when are we gonna fork this sucker?

[Ian Collins]

On 04/16/10 03:32 PM, Elvis Presley wrote:

It is inevitable that many will always compare the 'open'ness of 'open'solaris 
with that of fedora/rhel.

To say that you fear management reprisals for 'open'ly discussing 'open'solaris 
signals almost complete death of opensolaris to me.  :(

   
Maybe, Mr Anonymous Coward, you should learn to read what was written 
before commenting on it.


--
Ian.

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] So when are we gonna fork this sucker?

[Elvis Presley]

It is inevitable that many will always compare the 'open'ness of 'open'solaris 
with that of fedora/rhel.

To say that you fear management reprisals for 'open'ly discussing 'open'solaris 
signals almost complete death of opensolaris to me.  :(

Message was edited by: elvisior
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Where to get iso of b 133 or b 134

[Edward Ned Harvey]

> From: opensolaris-discuss-boun...@opensolaris.org [mailto:opensolaris-
> discuss-boun...@opensolaris.org] On Behalf Of Harry Putnam
> 
> I think this may have been mentioned already here, but looking home
> page of opensolaris.org, I'm not seeing a pointer.

If you go to http://opensolaris.org
And go to Downloads
You'll see a section called "Developer builds"
It is hard to see because it's not a hyperlink.  http://www.genunix.org 
The website simply says:
or download the candidate builds on genunix.org.

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] MythTV

[weekleyj]

On 3/18/10 3:24 PM, James Lever wrote:


On 19/03/2010, at 12:20 AM, homerun wrote:


Just asking , is there interest to port MythTV to OpenSolaris.
http://www.mythtv.org/


There is definite interest.

Ken Mays reported building it from source on desktop-discuss quite some
time ago and when I queried him about it late last year he responded
saying,

“”"
You need a few dependencies which are now a part of OSOL.

Use this web page as a guide to what you need:
http://www.mythtv.org/wiki/Dependencies_(Building_from_source)
“””

I would be interested in having a regular binary build in that I can
pull from a repository mostly because I’ve still not found time to build
from source.

I think the main benefit of MythTV on OSOL at the moment is either as a
frontend or as a slave backend or backend with network tuners due to
capture card support in the operating system.

Me…. I would be using it as a primary dedicated backend with a pair of
HDHomeRun DVB-T tuners :)

cheers,
James




I can add the command line hdhomerun software "libhdhomerun" to the 
"ported" pile.  There was a minor bug in the makefile preventing a 
successful build.

If anyone is interested I can provide the Makefile off list.

John
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] So when are we gonna fork this sucker?

[Martin Bochnig]

Answer: TOO high.


%mab



On Fri, Apr 16, 2010 at 12:22 AM, Peter Jones  wrote:
> I am interested in very clear terms what the pros and cons of a FORK is from 
> a technical point of view?
>
> Support for a successful operating system is very roughly composed of the 
> following
> Technical development
> ISV development and partner input
> IHV development and partner input
> Marketing, branding,and strategy
> Support
> User community/customers
> Resources/working capital
>
> To differentiate with a distinctive business model is difficult but it must 
> be clear to the customer/user.Sun/oracle offers a great advantage over other 
> os's in that it provides access to active customers base.clearly to fork 
> technically it would be weighed up against the other issues.
>
> The third way would be to be proactive...present a business case to oracle as 
> a separate business enity.There are many exciting options
> --
> This message posted from opensolaris.org
> ___
> opensolaris-discuss mailing list
> opensolaris-discuss@opensolaris.org
>
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Created a FAT32 partition using Fdisk, but it can not be mounted..

[Sarah.kho]

Thank you very much. my problem resolved.
The steps are like this:

use fdisk to create the partition
use mkfs to create the file system
use mount to mount the partition.

note: both mkfs and mount need the -F pcfs parameter to know what is the
partition format.



On Thu, Apr 15, 2010 at 10:12 AM, Dave Koelmeyer [via Opensolaris General
discuss mail-list]

> wrote:

> I have a wee how-to here if it helps at all:
>
>
> http://davekoelmeyer.wordpress.com/2010/04/09/format-a-usb-flash-drive-with-fat32-on-opensolaris-x64/
> --
> This message posted from opensolaris.org
> ___
> opensolaris-discuss mailing list
> [hidden 
> email]
>
>
> --
>  View message @
> http://n2.nabble.com/Created-a-FAT32-partition-using-Fdisk-but-it-can-not-be-mounted-tp4901842p4905752.html
> To unsubscribe from Created a FAT32 partition using Fdisk, but it can not
> be mounted.., click here< (link removed) ==>.
>
>
>

-- 
View this message in context: 
http://n2.nabble.com/Created-a-FAT32-partition-using-Fdisk-but-it-can-not-be-mounted-tp4901842p4910075.html
Sent from the Opensolaris General discuss mail-list mailing list archive at 
Nabble.com.
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] So when are we gonna fork this sucker?

[Peter Jones]

I am interested in very clear terms what the pros and cons of a FORK is from a 
technical point of view?

Support for a successful operating system is very roughly composed of the 
following
Technical development 
ISV development and partner input 
IHV development and partner input
Marketing, branding,and strategy
Support
User community/customers
Resources/working capital

To differentiate with a distinctive business model is difficult but it must be 
clear to the customer/user.Sun/oracle offers a great advantage over other os's 
in that it provides access to active customers base.clearly to fork technically 
it would be weighed up against the other issues.

The third way would be to be proactive...present a business case to oracle as a 
separate business enity.There are many exciting options
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] So when are we gonna fork this sucker?

[Peter Tribble]

On Wed, Apr 14, 2010 at 2:44 PM, Chad Welsh  wrote:
> Well we have heard from one of the OGB members about forking this project to 
> save it from the hands of [b]"HOracle"[/b] the multi-headed, voracious, open 
> source devouring monster. So, when and who might be the first step/person to 
> make this happen to protect what has been done so far?

Anyone can. And has been able to for almost 5 years.

And they may not necessarily be what you're thinking of in terms of a fork,
but Nexenta/OSUNIX/Belenix and the other distro efforts would all welcome
some help, I'm sure.

The question is why fork in the first place? Preventing the primary sponsor
of the project from being able to contribute because you don't understand
their business seems a poor choice. It's also hugely negative rather than
pushing forward in new areas. What can you manage to do better that is
worth the damage down by a fork?

(I can see different technical decisions as being valid drivers for going
along an alternative path. But I think you need to clearly formulate your
goals before deciding how to implement your strategy.)

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

Re: [osol-discuss] [website-discuss] Free OpenSolaris CD no longer offered

[Martin Bochnig]

Hi Alan,

thanks for your response.
Sorry that I took that personally, although I was not targeted. I just
have a few problems here.
Among other things - like every day - with a certain problem in the
SPARC-Xorg ws.

Also I see from the window, that today my car outside got the 3rd
police ticket in 2 days.

Also I type this via a small laptop keyboard, because all other kbds
are connected to SB100, SB1500, SB2000, T2000 and SB2500 ...



Let us please defer this. I respond later, some day. Too many things
are (literally) in the air right now




regards,
%martin


On Thu, Apr 15, 2010 at 8:52 PM, Alan Coopersmith
 wrote:
> Martin Bochnig wrote:
 For the benefit of everyone else who might want to use them? (Unless you
 believe they're less deserving than Sun, of course.)
>>> sure. But this is my position: If Sun doesnt want them (upstream into
>>> ON),
>>
>>
>> Err: ON for libpciaccess.
>> xwin for the rest.
>
> libpciaccess itself is in the X gate, though it calls libraries like 
> libdevinfo
> in ON.   I don't remember seeing any open requests from you to have code
> integrated into either gate, though I've been busy with many other things
> and may have missed or forgotten them.   Ideally though, changes to things 
> like
> libpciaccess would be submitted by you to the upstream projects, and we'd just
> pull them in automatically in the next resync, so we don't have to maintain a
> patch, especially not a patch we can't really test or maintain.
>
>> But we discussed this long ago, 1000 times.
>> Sun doesnt want. EOF
>
> Yes, as we've discussed many times before, the management of Sun's SPARC
> workstations & graphics groups, back when Sun sold such things and thus still
> had a group, decided that they were not going to invest in a lot of work to
> support hardware that would not be generating any revenue, since it was no
> longer sold and would be past the end of its support life by the time 
> customers
> started buying support contracts for the next enterprise release of Solaris.
> Now that those product lines and product teams are long gone, it will be even
> harder to convince new management to spend any resources there.
>
> I'm not doing anything to block you from being able to provide drivers, and
> the FOX project is still open to host code you wish to share with users or
> other distro builders, but we're not planning to pull any of those drivers
> into Oracle's releases.   Unfortunately, the SPARC support for source juicer
> has been delayed, so you can't yet submit them via there for hosting in the
> pkg.opensolaris.org/contrib repo, but you can provide them directly to users
> in a variety of other ways.
>
>> This was accepted from my side.
>> But blaming users for writing more emails, than they contribute code ... ?
>> At least in case of some individuals that is a bad joke!
>>
>> Shouldnt one be careful with too general allegations?
>
> I'm sorry you took offense to my joke - I really wasn't thinking about you at
> all, but the people who post many of the complaints, but have never written a
> line of code, have never helped another user on IRC or mailing lists, have
> never gone to their local user groups, have never filed a bug or helped track
> down an issue, have never participated in design or code review discussions,
> and have never contributed to the community in any other way.   Even if not
> much of your code has been contributed back to the main OpenSolaris project
> gates, you've contributed in many other ways.
>
> --
>        -Alan Coopersmith-        alan.coopersm...@oracle.com
>         Oracle Solaris Platform Engineering: X Window System
>
>
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] [website-discuss] Free OpenSolaris CD no longer offered

[Alan Coopersmith]

Martin Bochnig wrote:
>>> For the benefit of everyone else who might want to use them? (Unless you
>>> believe they're less deserving than Sun, of course.)
>> sure. But this is my position: If Sun doesnt want them (upstream into
>> ON),
> 
> 
> Err: ON for libpciaccess.
> xwin for the rest.

libpciaccess itself is in the X gate, though it calls libraries like libdevinfo
in ON.   I don't remember seeing any open requests from you to have code
integrated into either gate, though I've been busy with many other things
and may have missed or forgotten them.   Ideally though, changes to things like
libpciaccess would be submitted by you to the upstream projects, and we'd just
pull them in automatically in the next resync, so we don't have to maintain a
patch, especially not a patch we can't really test or maintain.

> But we discussed this long ago, 1000 times.
> Sun doesnt want. EOF

Yes, as we've discussed many times before, the management of Sun's SPARC
workstations & graphics groups, back when Sun sold such things and thus still
had a group, decided that they were not going to invest in a lot of work to
support hardware that would not be generating any revenue, since it was no
longer sold and would be past the end of its support life by the time customers
started buying support contracts for the next enterprise release of Solaris.
Now that those product lines and product teams are long gone, it will be even
harder to convince new management to spend any resources there.

I'm not doing anything to block you from being able to provide drivers, and
the FOX project is still open to host code you wish to share with users or
other distro builders, but we're not planning to pull any of those drivers
into Oracle's releases.   Unfortunately, the SPARC support for source juicer
has been delayed, so you can't yet submit them via there for hosting in the
pkg.opensolaris.org/contrib repo, but you can provide them directly to users
in a variety of other ways.

> This was accepted from my side.
> But blaming users for writing more emails, than they contribute code ... ?
> At least in case of some individuals that is a bad joke!
> 
> Shouldnt one be careful with too general allegations?

I'm sorry you took offense to my joke - I really wasn't thinking about you at
all, but the people who post many of the complaints, but have never written a
line of code, have never helped another user on IRC or mailing lists, have
never gone to their local user groups, have never filed a bug or helped track
down an issue, have never participated in design or code review discussions,
and have never contributed to the community in any other way.   Even if not
much of your code has been contributed back to the main OpenSolaris project
gates, you've contributed in many other ways.

-- 
-Alan Coopersmith-alan.coopersm...@oracle.com
 Oracle Solaris Platform Engineering: X Window System

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] So when are we gonna fork this sucker?

[Erlan Sergaziev]

>> LOL. If builds were regular before and suddenly stopped it's a good sign
that Oracle eh ... is about to deliver :)
>Do you actually read, what's written here?

I understand the heated debate here, but, come on, let's not loose some
sense of humor.
This was my attempt to make a joke (cp. pregnancy).

Anyhow, my point was that Oracle's silence caused such a situation that
close community members (perhaps even OGB) have to resort to these funny
clues to guess what's coming.
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] [indiana-discuss] [laptop-discuss] So when are we gonna fork this sucker?

[C. Bergström]

Dennis Clarke wrote:




Regarding the i18n code, do we have any sort of detailed specs on it such that 
a motivated programmer could sit down with a stack of POSIX books and just 
begin to hammer it out from first principals? I mean, without doing something 
obscene like reverse engineering or simple disassembly.

Dennis 


ps: sorry about the lack of message indenting but Sun Convergence isn't doing 
that for me this morning. Thankfully it does not break maillist threads.
  

Dennis

The task is very clear..

#1 Take the closed bin.. extract.. run objdump on it to determine the 
symbol(s) (which will probably match the name of the file) This is an 
interface to a POSIX compliant function as you realize..
#2 Take Citrus from NetBSD and see how they implemented it.  (In many 
cases it will be trivial)
#3 QA/fix bugs.. (Simply build a static archive like they do and see if 
it works as a suitable drop-in replacement)


Steps 1-2 shouldn't take long tbh and the POSIX books are probably not 
even required if the NetBSD compatibility can be believed.

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Non Fud official speaking

[Reynaldo]

I've decided to adopt the 'wait and see' approach. Oracle may not be good at 
community relations, but there is no need for the panic all over the net.
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Where to get iso of b 133 or b 134

[Sven Marcel Buchholz]

take a look at:
http://www.genunix.org/distributions/indiana/
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Where to get iso of b 133 or b 134

[Viktor Cemasko]

Greetings,

On Thu, 15 Apr 2010 18:04:45 +0200, Harry Putnam wrote:


I think this may have been mentioned already here, but looking home
page of opensolaris.org, I'm not seeing a pointer.



Are there available isos of b 133/134?


http://genunix.org/distributions/indiana/

http://dlc.sun.com/torrents/info/

Have  a nice day!
--
Üdvözlettel,http://opensolaris.org/os/project/indiana/
Cemasko Viktor. http://www.opera.com/browser/

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Where to get iso of b 133 or b 134

[eXeC001er]

genunix.org

2010/4/15 Harry Putnam 

> I think this may have been mentioned already here, but looking home
> page of opensolaris.org, I'm not seeing a pointer.
>
> Are there available isos of b 133/134?
>
> ___
> opensolaris-discuss mailing list
> opensolaris-discuss@opensolaris.org
>
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

[osol-discuss] Where to get iso of b 133 or b 134

[Harry Putnam]

I think this may have been mentioned already here, but looking home
page of opensolaris.org, I'm not seeing a pointer.

Are there available isos of b 133/134?

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] [website-discuss] Free OpenSolaris CD no longer offered

[Martin Bochnig]

On Thu, Apr 15, 2010 at 5:44 PM, Alan Coopersmith
 wrote:
> Matthias Pfützner wrote:
>> You (Chris Pickett) wrote:
>>> Next on this channel: Source code access on a paid subscription level
>>> and you have to pay $$$ per incident if you want to contribute to ON
>>
>> HOW do you think, LEGALLY you can LIMIT access to Open Source?
>>
>> Start thinking, please!
>
> Besides if we really wanted to make money, there's not much revenue in
> charging for commits to ON at the current contribution rate - much better
> to charge per post to the mailing lists / web forums.
>
> Of course, we'd offer free posting credits to those who did contribute code,
> since such contributions require some discussion for review and such - so
> at this point, Roland & Olga would have a decent balance of free posts built
> up for them to use, but certain other people would be deeply in debt to 
> Oracle.
>
> Fortunately, they keep people like me far away from the business folks, so
> we don't try to sell insane ideas like this to them...   8-)
>
> --
>        -Alan Coopersmith-        alan.coopersm...@oracle.com
>         Oracle Solaris Platform Engineering: X Window System



Sadly Sun does not WANT my contributions.
So why shall I publish them?
I only do so, if you guys intend to use my libpciaccess fixes plus
SPARC gfx Xorg patches.
Then alone and only.

Therefore not.




%martin bochnig
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] [website-discuss] Free OpenSolaris CD no longer offered

[Alan Coopersmith]

Matthias Pfützner wrote:
> You (Chris Pickett) wrote:
>> Next on this channel: Source code access on a paid subscription level
>> and you have to pay $$$ per incident if you want to contribute to ON
> 
> HOW do you think, LEGALLY you can LIMIT access to Open Source?
> 
> Start thinking, please!

Besides if we really wanted to make money, there's not much revenue in
charging for commits to ON at the current contribution rate - much better
to charge per post to the mailing lists / web forums.

Of course, we'd offer free posting credits to those who did contribute code,
since such contributions require some discussion for review and such - so
at this point, Roland & Olga would have a decent balance of free posts built
up for them to use, but certain other people would be deeply in debt to Oracle.

Fortunately, they keep people like me far away from the business folks, so
we don't try to sell insane ideas like this to them...   8-)

-- 
-Alan Coopersmith-alan.coopersm...@oracle.com
 Oracle Solaris Platform Engineering: X Window System

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Crash analysis help - reboot after panic: Freeing a free IOMMU page: paddr=0xccca2000

[Bruno Sousa]

Hi Brian,

Indeed it seems to be a duplicated bug, so let's see if we don't hit any
bug regression in snv_134.
Please allow me to get the system running for a month or so without
hitting , at least, this specific issue, and after that i can close the
bug report , confirming that the bug was seen in snv_132 but fixed in
snv_134 .

Bruno

On 15-4-2010 14:01, Brian Ruthven - Solaris Network Sustaining - Oracle
UK wrote:
>
> Ah, I had missed the reference to snv_132 in your original mail. In
> that case, it's possible this is simply a repeat of 6897282, which was
> fixed in snv_133, and you may now already have the fix. I suspect
> http://defect.opensolaris.org/bz/show_bug.cgi?id=15601 can be closed
> as a dup of
> http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6875273
> unless  you see it again post snv_133.
>
> Regards,
> Brian
>
>
> Bruno Sousa wrote:
>> Hi Brian,
>>
>> First off all let me thank you for your time spent on reading this issue.
>> Regarding your comments , yes the defect 15601 has been logged by me
>> .I will keep all those informations stored somewhere, and meantime if
>> required i can upload the core file , and i haven't done so because
>> i'm enable to create an official support request my supplier says
>> that OpenSolaris is not supported anymore.
>> After the crash i noticed that i had done in , a recent past, and pkg
>> image-update to snv_134 while running snv_132 . So the crash happened
>> in snv_132 and now i'm running snv_134 without the workaround
>> mentioned in
>> http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6875273 .
>>
>> I sure hope this issue can be fixed, otherwise the project may suffer
>> a big turn, and i want to avoid this at all cost.
>>
>> Once again , thanks for all your support .
>>
>> Regards,
>> Bruno
>>
>>
>>
>>
>> On 14-4-2010 12:16, Brian Ruthven - Solaris Network Sustaining -
>> Oracle UK wrote:
>>>
>>> Hi Bruno,
>>>
>>> Googling for this panic, I see defect 15601
>>> (http://defect.opensolaris.org/bz/show_bug.cgi?id=15601) has been
>>> logged, I suspect by you?
>>>
>>> If so, then I think you've already done just about everything I
>>> would expect of a bug finder - you've logged a bug including the
>>> panic message, the stack trace, and you have the dump available if
>>> necessary (keep the vmdump.X file handy somewhere). I'd also keep a
>>> copy of the messages file(s) which cover before and after the panic,
>>> just in case.
>>>
>>> It might be worth expanding on the "crash quite often" to try and
>>> put a figure on it (like "average 3 times a day" or something) and
>>> give a rough idea of the workload, along with an idea of how long
>>> the server has been used (under what load) before the problem
>>> happened and include any image-update dates. If you don't add this
>>> to the bug, keep a record in case anybody asks later!
>>>
>>> Searching for existing bugs based on the stack trace, it looks
>>> similar to CR 6897282 (reported in snv_125) but closed as a dup of
>>> CR 6875273 which integrated in snv_133. As you seem to be running
>>> snv_134, it would seem this bug has re-surfaced. Can you verify your
>>> snv_134 installation with "pkg verify" to make sure nothing is
>>> reported as corrupt.
>>>
>>> Try the workaround mentioned in
>>> http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6875273
>>> and see if that gives you some stability for your migration project
>>> efforts.
>>>
>>> Regards,
>>> Brian
>>>
>>>
>>>
>>>
>>> Bruno Sousa wrote:
 Hi all,

 Recently one of the servers that my company uses , a Dell R710, attached 
 to 2 Sun JBOD J4400 started to crash quite often.
 Finally i got a message in /var/adm/messages that might point to something 
 usefull, but i don't have the expertise to start to
 troubleshooting this problem, so any help would be highly valuable.

 Thanks in advance,

 Bruno Sousa



 The significant messages are :

 Apr 13 11:12:04 san01 savecore: [ID 570001 auth.error] reboot after panic: 
 Freeing a free IOMMU page: paddr=0xccca2000
 Apr 13 11:12:04 san01 savecore: [ID 385089 auth.error] Saving compressed 
 system crash dump in /var/crash/san01/vmdump.0

 I also noticed other "interesting" messages like :

 Apr 13 11:11:10 san01 unix: [ID 378719 kern.info] NOTICE: cpu_acpi: _PSS 
 package evaluation failed for with status 5 for CPU 0.
 Apr 13 11:11:10 san01 unix: [ID 388705 kern.info] NOTICE: cpu_acpi: error 
 parsing _PSS for CPU 0
 Apr 13 11:11:10 san01 unix: [ID 928200 kern.info] NOTICE: SpeedStep 
 support is being disabled due to errors parsing ACPI P-state objects 
 exported by BIOS

 Apr 13 11:10:50 san01 scsi: [ID 243001 kern.info] 
 /p...@0,0/pci8086,3...@4/pci1028,1...@0 (mpt0):
 Apr 13 11:10:50 san01   DMA restricted below 4GB boundary due to errata
 Apr 13 11:11:32 san01 scsi: [ID 243001 kern.info] 
 /p...@0,0/pci8086,3...@9/pci1000,3

Re: [osol-discuss] [indiana-discuss] [laptop-discuss] So when are we gonna fork this sucker?

[Dennis Clarke]

On 04-15-10, "C. Bergström"  wrote: 
Dennis Clarke wrote:
>
>
>
>Regarding the i18n code, do we have any sort of detailed specs on it such that 
>a motivated programmer could sit down with a stack of POSIX books and just 
>begin to hammer it out from first principals? I mean, without doing something 
>obscene like reverse engineering or simple disassembly.
>
>Dennis 
>ps: sorry about the lack of message indenting but Sun Convergence isn't doing 
>that for me this morning. Thankfully it does not break maillist threads.
>  
Dennis

The task is very clear..

#1 Take the closed bin.. extract.. run objdump on it to determine the symbol(s) 
(which will probably match the name of the file) This is an interface to a 
POSIX compliant function as you realize..
#2 Take Citrus from NetBSD and see how they implemented it.  (In many cases it 
will be trivial)
#3 QA/fix bugs.. (Simply build a static archive like they do and see if it 
works as a suitable drop-in replacement)

Steps 1-2 shouldn't take long tbh and the POSIX books are probably not even 
required if the NetBSD compatibility can be believed.

---

We think alike. 

By the way, I did get your other message and have it printed and stuck to my 
office wall here. I'll try to get back to you on that later but today I am in 
meetings. In fact, I have to go into the datacenter of a major two letter car 
manufacturer and calm the nerves of the IT director there. He is pretty 
concerned about Solaris right now and he runs on piles of Sun gear all setup by 
me years and years ago. So, I can tell you that the zillion dollar datacenters 
are nervous.  No kidding. 

This whole transition has made waves everywhere and I really don't like it when 
it hits my revenue streams. Pun intended.

I hope to catch up with you later.

Dennis 

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] [indiana-discuss] [laptop-discuss] So when are we gonna fork this sucker?

[Dennis Clarke]

> On 04-15-10, joerg.schill...@fokus.fraunhofer.de (Joerg Schilling) wrote: 
> Alan Coopersmith  wrote:

>> It would be nice to finally get that i18n code replaced, if for no other
>> reason than to stop seeing articles about how the IBM System/Z port can't
>> make progress because they can't build libc.

> Isn't this funny while knowing that it is IBM that brevents this as IBM owns 
> the code that cannot be made OSS.

> Jörg



Regarding the i18n code, do we have any sort of detailed specs on it such that 
a motivated programmer could sit down with a stack of POSIX books and just 
begin to hammer it out from first principals? I mean, without doing something 
obscene like reverse engineering or simple disassembly.

Dennis 

ps: sorry about the lack of message indenting but Sun Convergence isn't doing 
that for me this morning. Thankfully it does not break maillist threads.
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Crash analysis help - reboot after panic: Freeing a free IOMMU page: paddr=0xccca2000

[Brian Ruthven - Solaris Network Sustaining - Oracle UK]

Ah, I had missed the reference to snv_132 in your original mail. In that 
case, it's possible this is simply a repeat of 6897282, which was fixed 
in snv_133, and you may now already have the fix. I suspect 
http://defect.opensolaris.org/bz/show_bug.cgi?id=15601 can be closed as 
a dup of 
http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6875273 
unless  you see it again post snv_133.


Regards,
Brian


Bruno Sousa wrote:

Hi Brian,

First off all let me thank you for your time spent on reading this issue.
Regarding your comments , yes the defect 15601 has been logged by me 
.I will keep all those informations stored somewhere, and meantime if 
required i can upload the core file , and i haven't done so because 
i'm enable to create an official support request my supplier says that 
OpenSolaris is not supported anymore.
After the crash i noticed that i had done in , a recent past, and pkg 
image-update to snv_134 while running snv_132 . So the crash happened 
in snv_132 and now i'm running snv_134 without the workaround
mentioned in 
http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6875273 .


I sure hope this issue can be fixed, otherwise the project may suffer 
a big turn, and i want to avoid this at all cost.


Once again , thanks for all your support .

Regards,
Bruno




On 14-4-2010 12:16, Brian Ruthven - Solaris Network Sustaining - 
Oracle UK wrote:


Hi Bruno,

Googling for this panic, I see defect 15601 
(http://defect.opensolaris.org/bz/show_bug.cgi?id=15601) has been 
logged, I suspect by you?


If so, then I think you've already done just about everything I would 
expect of a bug finder - you've logged a bug including the panic 
message, the stack trace, and you have the dump available if 
necessary (keep the vmdump.X file handy somewhere). I'd also keep a 
copy of the messages file(s) which cover before and after the panic, 
just in case.


It might be worth expanding on the "crash quite often" to try and put 
a figure on it (like "average 3 times a day" or something) and give a 
rough idea of the workload, along with an idea of how long the server 
has been used (under what load) before the problem happened and 
include any image-update dates. If you don't add this to the bug, 
keep a record in case anybody asks later!


Searching for existing bugs based on the stack trace, it looks 
similar to CR 6897282 (reported in snv_125) but closed as a dup of CR 
6875273 which integrated in snv_133. As you seem to be running 
snv_134, it would seem this bug has re-surfaced. Can you verify your 
snv_134 installation with "pkg verify" to make sure nothing is 
reported as corrupt.


Try the workaround mentioned in 
http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6875273 
and see if that gives you some stability for your migration project 
efforts.


Regards,
Brian




Bruno Sousa wrote:

Hi all,

Recently one of the servers that my company uses , a Dell R710, attached to 2 
Sun JBOD J4400 started to crash quite often.
Finally i got a message in /var/adm/messages that might point to something 
usefull, but i don't have the expertise to start to
troubleshooting this problem, so any help would be highly valuable.

Thanks in advance,

Bruno Sousa



The significant messages are :

Apr 13 11:12:04 san01 savecore: [ID 570001 auth.error] reboot after panic: 
Freeing a free IOMMU page: paddr=0xccca2000
Apr 13 11:12:04 san01 savecore: [ID 385089 auth.error] Saving compressed system 
crash dump in /var/crash/san01/vmdump.0

I also noticed other "interesting" messages like :

Apr 13 11:11:10 san01 unix: [ID 378719 kern.info] NOTICE: cpu_acpi: _PSS 
package evaluation failed for with status 5 for CPU 0.
Apr 13 11:11:10 san01 unix: [ID 388705 kern.info] NOTICE: cpu_acpi: error 
parsing _PSS for CPU 0
Apr 13 11:11:10 san01 unix: [ID 928200 kern.info] NOTICE: SpeedStep support is 
being disabled due to errors parsing ACPI P-state objects exported by BIOS

Apr 13 11:10:50 san01 scsi: [ID 243001 kern.info] 
/p...@0,0/pci8086,3...@4/pci1028,1...@0 (mpt0):
Apr 13 11:10:50 san01   DMA restricted below 4GB boundary due to errata
Apr 13 11:11:32 san01 scsi: [ID 243001 kern.info] 
/p...@0,0/pci8086,3...@9/pci1000,3...@0 (mpt2):
Apr 13 11:11:32 san01   DMA restricted below 4GB boundary due to errata



Relevant specs of the machine :

SunOS san01 5.11 snv_134 i86pc i386 i86pc Solaris

rpool boot drives attached to a Dell SAS6/iR Integrated RAID Controller
(mpt0 Firmware version v0.25.47.0 (IR) )
2 HBA LSI 1068E, each connect to a J4400 jbod (mpt1 Firmware version
v1.26.0.0 (IT) )

multipath enabled and working

2 Quad-Cores, 16Gb ram

Detailed info :
  


mdb -k unix.0 vmcore.0

mdb: warning: dump is from SunOS 5.11 snv_132; dcmds and macros may 
not match kernel implementation
Loading modules: [ unix genunix specfs dtrace mac cpu.generic uppc 
pcplusmp scsi_vhci zfs mpt sd sockfs ip hook neti sctp arp usba uhci 
fctl stmf md lofs idm nfs random sppp fcip cpc crypto logindmux ptm 
nsctl 

Re: [osol-discuss] [indiana-discuss] [laptop-discuss] So when are we gonna fork this sucker?

[Joerg Schilling]

Alan Coopersmith  wrote:

> It would be nice to finally get that i18n code replaced, if for no other
> reason than to stop seeing articles about how the IBM System/Z port can't
> make progress because they can't build libc.

Isn't this funny while knowing that it is IBM that brevents this as IBM owns 
the code that cannot be made OSS.

Jörg

-- 
 EMail:jo...@schily.isdn.cs.tu-berlin.de (home) Jörg Schilling D-13353 Berlin
   j...@cs.tu-berlin.de(uni)  
   joerg.schill...@fokus.fraunhofer.de (work) Blog: 
http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] So when are we gonna fork this sucker?

[Matthias Pfützner]

You (Erlan Sergaziev) wrote:
> "I must say though, hearing that the dev builds have stopped being made 
> public sounds encouraging, it could be that 2010.04 is on it's way!"
> 
> LOL. If builds were regular before and suddenly stopped it's a good sign that 
> Oracle eh ... is about to deliver :)

Do you actually read, what's written here?

Same happened before 2009.06 was delivered! That's, why he calls it a good
sign!

> I hope Oracle people do understand what kind of hassle they're creating by 
> keeping silent about osol future.

As I stated: It's not Oracle that needs to adopt. It's us here, that need to
UNDERSTAND the difference in outbound messaging between Sun and Oracle!

Matthias
-- 
Matthias Pfützner | Tel.: +49 700 PFUETZNER  | "Du warst doch mal roman-
Lichtenbergstr.73 | mailto:matth...@pfuetzner.de | tisch." "Dafür bin ich
D-64289 Darmstadt | AIM: pfuetz, ICQ: 300967487  | jetzt rheumatisch." K.
Germany  | http://www.pfuetzner.de/matthias/ | Branagh, Peter's Friends
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] So when are we gonna fork this sucker?

[Erlan Sergaziev]

"I must say though, hearing that the dev builds have stopped being made 
public sounds encouraging, it could be that 2010.04 is on it's way!"

LOL. If builds were regular before and suddenly stopped it's a good sign that 
Oracle eh ... is about to deliver :)

I hope Oracle people do understand what kind of hassle they're creating by 
keeping silent about osol future.
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Free OpenSolaris CD

[Svein Skogen]

On 15.04.2010 12:58, Peter Jones wrote:
> My personal thoughts are that we are told CD's are no longer in great 
> demand.The new media for Opensolaris on hard copy should be branded flash 
> drives where the consumers can see the flash drive as a value add to the 
> Brand/OS.Selling them via amazon or oracle online shop is an interesting idea 
> to reach the target audience with additional download facilities of course.

If Oracle play their cards right on this one, and get a few ISVs to
supply applications (even for sale) for (Open)Solaris, they may have a
winner on their hand. Properly done, (Open)Solaris might manage what
Linux never did as a suitable replacement desktop/laptop OS for Windows,
especially if they get ISVs like Adobe aboard. And given the current
ice-age climate between Adobe and Apple, this might be the right time
for it. ;)

//Svein

-- 
+---+---
  /"\   |Svein Skogen   | sv...@d80.iso100.no
  \ /   |Solberg Østli 9| PGP Key:  0xE5E76831
   X|2020 Skedsmokorset | sv...@jernhuset.no
  / \   |Norway | PGP Key:  0xCE96CE13
|   | sv...@stillbilde.net
 ascii  |   | PGP Key:  0x58CD33B6
 ribbon |System Admin   | svein-listm...@stillbilde.net
Campaign|stillbilde.net | PGP Key:  0x22D494A4
+---+---
|msn messenger: | Mobile Phone: +47 907 03 575
|sv...@jernhuset.no | RIPE handle:SS16503-RIPE
+---+---
 If you really are in a hurry, mail me at
   svein-mob...@stillbilde.net
 This mailbox goes directly to my cellphone and is checked
even when I'm not in front of my computer.

 Picture Gallery:
  https://gallery.stillbilde.net/v/svein/




signature.asc
Description: OpenPGP digital signature
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

[osol-discuss] Free OpenSolaris CD

[Peter Jones]

My personal thoughts are that we are told CD's are no longer in great 
demand.The new media for Opensolaris on hard copy should be branded flash 
drives where the consumers can see the flash drive as a value add to the 
Brand/OS.Selling them via amazon or oracle online shop is an interesting idea 
to reach the target audience with additional download facilities of course.
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] [website-discuss] Free OpenSolaris CD no longer offered

[Matthias Pfützner]

Folks,

You (Giovanni Tirloni) wrote:
> If they are integrating this program under another division, why wouldn't
> they mention that right now ?
> 
> That would save on all the FUD that's already spreading like wildfire. Am I
> to believe they are that naive ? I think they aren't.
> 
> Only time will tell how everything plays out but I think we are all waiting
> for a train that isn't coming to this station.
> 
> -- 
> Giovanni

It might be as simple as:

Open Source doesn't immediately generate money, but COSTS money, if you are a
sponsor.

So, as I also already stated: Oracle publishes things, when they are
decided. No infos about ONGOING INTERNAL DISCUSSIONS AND SPECULATIONS. Not
even internally!

>From that, it's as easy as that:

Try to learn to live with a NEW CULTURE of COMMUNICATION (not necessarily a
new way of handling the community! Oracle has other things to take care of in
the beginning, remember: Oracle promised, that Sun WILL DELIVER ADDITIONAL
MARGIN/PROFIT within the FIRST QUARTER (at least year), and that coming from a
position of LOOSING MONEY. So, they currently have much bigger fish to fry
(than how to spend money for Open Source engagements), that's for
sure!). Meanwhile, when you have pressing issues, you still have the
commercial way of dealing with that, which is: Your Sales Rep! Ask him/her,
complain to him/her! And don't EXPECT anyone from Oracle upper management
(those that decide!) to READ or LISTEN here in such email forums and
threads... And don't assume, that, BECAUSE you don't hear anything NOW, it
implies, you never will. Once Sun will be profitable again inside Oracle,
Oracle can THEN decide on how to spend the money. Also remember: Oracle is
very Larry-centric, many, if not all money-related decisions pass his desk
literally. So, as he's a SINGLE person, give him time to decide the bigger
things first.

That's all we're all are asking for for days, weeks, even months. Complaining
HERE doesn't help anybody!

So, again, also: Please STOP this discussions NOW, we're now doing that stuff
over a month now, it's getting really boring. No-one adds any real new infos,
it's seemingly only complaints, FUD and other some-such things...

Rest assured. Once things will be decided, we all will get to know them. Until
then, let's simply assume, that Oracle will be going forward with OpenSolaris
as it had been promising way and long before the acquisition.

OK, 'nuff now of this rant...

Matthias
-- 
Matthias Pfützner | Tel.: +49 700 PFUETZNER  | "Da muß ich mich wohl
Lichtenbergstr.73 | mailto:matth...@pfuetzner.de | irgendwie vertan haben!"
D-64289 Darmstadt | AIM: pfuetz, ICQ: 300967487  | Miracolix über einen neuen
Germany  | http://www.pfuetzner.de/matthias/ | Trank im Asterix Band 30.
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] [website-discuss] Free OpenSolaris CD no longer offered

[James Grisanzio]

hi ... when you guys respond to this thread please take 
opensolaris-announce off the cc list. Thanks.


Jim
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] [website-discuss] Free OpenSolaris CD no longer offered

[Matthias Pfützner]

You (Chris Pickett) wrote:
> Next on this channel: Source code access on a paid subscription level
> and you have to pay $$$ per incident if you want to contribute to ON
> >;-(
> 
> Chris

HOW do you think, LEGALLY you can LIMIT access to Open Source?

Start thinking, please!

Matthias
-- 
Matthias Pfützner | Tel.: +49 700 PFUETZNER  | "Da muß ich mich wohl
Lichtenbergstr.73 | mailto:matth...@pfuetzner.de | irgendwie vertan haben!"
D-64289 Darmstadt | AIM: pfuetz, ICQ: 300967487  | Miracolix über einen neuen
Germany  | http://www.pfuetzner.de/matthias/ | Trank im Asterix Band 30.
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org