Aurélien Larcher | 16 Oct 19:26 2014


I was wondering if the generated manifest should prepend "org.opensolaris.category.2008:" to the actual classification in transforms/manifest-metadata-template:

Best regards,

oi-dev mailing list
oi-dev <at>
Aurélien Larcher | 15 Oct 21:34 2014

NUMPY update

sorry for prioritizing what I use for work but I am now considering updating numpy to 1.8.2 (previous stable release) in /hipster, which I use now on my workstation.

I see that

alarcher <at> phainos> pkg contents -r -o fmri,type -t depend numpy
FMRI                                            TYPE
library/python-2/numpy-26 <at> 1.4.1,5.11-2014.1.3.0 conditional
library/python-2/numpy-27 <at> 1.4.1,5.11-2014.1.3.0 conditional


alarcher <at> phainos> pkg search -l -o 'depend:require:library/python-2/numpy-26'
alarcher <at> phainos> pkg search -l -o 'depend:require:library/python-2/numpy-27'

which seems to indicate that there are no consumers of this module.

I would like to make sure that it is the case and know whether it would potentially break something.
Best regards,


LARCHER Aurélien          | KTH, School of Computer Science and Communication
Work: +46 (0) 8 790 71 42 | Lindstedtsvägen 5, Plan 4, 100 44 Stockholm, SWEDEN
oi-dev mailing list
oi-dev <at>
Alexander Pyhalov | 13 Oct 20:46 2014

Re: xmessage, ts, notion-3, and openttd


Josef 'Jeff' Sipek писал 07.10.2014 06:38:
> I have 4 new components to contribute.  A window manager, a command 
> batching
> utility, a recipe for xmessage (as far as I can tell, our current 
> xmessage
> package comes from oi-build from a couple of years ago), and finally 
> OpenTTD
> - because there's no reason OI can't be used for gaming. ;)
> You can get the changes via:
> $ git pull git:// hipster
> You can see the diffs at:

I have the following issue while compiling

warning: redundant redeclaration of ‘char* strcasestr(const char*, const 
char*)’ in same scope [-Wredundant-decls]
  char *strcasestr(const char *haystack, const char *needle);
In file included from /usr/gcc/4.8/include/c++/4.8.3/cstring:42:0,
/usr/include/string.h:121:14: warning: previous declaration of ‘char* 
strcasestr(const char*, const char*)’ [-Wredundant-decls]
  extern char *strcasestr(const char *, const char *);
[SRC] Linking openttd
Undefined                       first referenced
  symbol                             in file
icu_4_0::ParagraphLayout::Line::getVisualRun(int) const gfx_layout.o
icu_4_0::ParagraphLayout::Line::getWidth() const gfx_layout.o
icu_4_0::Collator::createInstance(icu_4_0::Locale const&, UErrorCode&) 
icu_4_0::UnicodeString::~UnicodeString() string.o
icu_4_0::LEFontInstance::getSubFont(unsigned short const*, int*, int, 
int, LEErrorCode&) const gfx_layout.o
icu_4_0::ParagraphLayout::ParagraphLayout(unsigned short const*, int, 
icu_4_0::FontRuns const*, icu_4_0::ValueRuns const*, icu_4_0::ValueRuns 
const*, icu_4_0::LocaleRuns const*, unsigned char, char, LEErrorCode&) 
icu_4_0::LEFontInstance::mapCharToGlyph(int, icu_4_0::LECharMapper 
const*) const gfx_layout.o
icu_4_0::ParagraphLayout::nextLine(float) gfx_layout.o
icu_4_0::LEFontInstance::~LEFontInstance() gfx_layout.o
icu_4_0::UnicodeString::UnicodeString(unsigned short const*) string.o
icu_4_0::LEFontInstance::canDisplay(int) const gfx_layout.o
icu_4_0::LEFontInstance::pixelsToUnits(LEPoint&, LEPoint&) const 
icu_4_0::LEFontInstance::getLineHeight() const gfx_layout.o
icu_4_0::LEFontInstance::mapCharToGlyph(int, icu_4_0::LECharMapper 
const*, char) const gfx_layout.o
icu_4_0::LEFontInstance::transformFunits(float, float, LEPoint&) const 
icu_4_0::UMemory::operator delete(void*) gfx_layout.o
icu_4_0::ParagraphLayout::Line::getLeading() const gfx_layout.o
icu_4_0::Locale::~Locale()          string.o
icu_4_0::LEFontInstance::xUnitsToPoints(float) const gfx_layout.o
icu_4_0::LEFontInstance::yUnitsToPoints(float) const gfx_layout.o
icu_4_0::LEFontInstance::yPixelsToUnits(float) const gfx_layout.o
icu_4_0::LEFontInstance::xPixelsToUnits(float) const gfx_layout.o
icu_4_0::BreakIterator::createCharacterInstance(icu_4_0::Locale const&, 
UErrorCode&) string.o
icu_4_0::UMemory::operator new(unsigned int) gfx_layout.o
icu_4_0::FontRuns::FontRuns(int)    gfx_layout.o
icu_4_0::FontRuns::~FontRuns()      gfx_layout.o
typeinfo for icu_4_0::LEFontInstance gfx_layout.o
icu_4_0::LEFontInstance::getDynamicClassID() const gfx_layout.o
icu_4_0::LEFontInstance::mapCharsToGlyphs(unsigned short const*, int, 
int, char, icu_4_0::LECharMapper const*, char, icu_4_0::LEGlyphStorage&) 
const gfx_layout.o
icu_4_0::BreakIterator::createWordInstance(icu_4_0::Locale const&, 
UErrorCode&) string.o
icu_4_0::Locale::Locale(char const*, char const*, char const*, char 
const*) string.o
icu_4_0::FontRuns::add(icu_4_0::LEFontInstance const*, int) gfx_layout.o
icu_4_0::LEFontInstance::unitsToPoints(LEPoint&, LEPoint&) const 
ld: fatal: symbol referencing errors. No output written to openttd
collect2: error: ld returned 1 exit status
make[2]: *** [openttd] Error 1
make[2]: Leaving directory 
make[1]: *** [all] Error 1
make[1]: Leaving directory 
gmake: *** 
Error 2

It seems it needs some LDFLAGS/CFLAGS  (-R/usr/g++/lib -L/usr/g++/lib) 
to link to correct icu (we have both library/icu and library/g++/icu 
installed on the build server).
Perhaps, it finds incorrect one?

Another strange one is notion-3. It installs config files in 
usr/etc/notion. Perhaps, you should specify something like

Merged xmessage update and ts addition.

System Administrator of Southern Federal University Computer Center

oi-dev mailing list
oi-dev <at>
Aurélien Larcher | 13 Oct 20:40 2014

Device driver utility repository ?

on a related topic, is there a repository with DDU source or should I set one up on GitHub from the files delivered with OI ?

oi-dev mailing list
oi-dev <at>
Alexander Pyhalov | 13 Oct 07:58 2014

New Hipster ISOs are available


We are ready to announce new OpenIndiana Hipster 2014.10.10 ISOs.


SHA256 Checksums:


I remind you, that you can pkg freeze entire to stay with this 
particular snapshot.

Snapshot notes:


System Administrator of Southern Federal University Computer Center
Alasdair Lumsden | 8 Oct 20:19 2014

Re: Resignation

> A related issue however is the apparent lack of ownership over the wiki.

In terms of ownership, EveryCity is providing free hosting of various bits of OpenIndiana physical
infrastructure, but it's down to the OpenIndiana project to determine who has ownership. There is a gulf
here that nobody has stepped up to fill after my resignation.

Keith Wesolowski quipped a joke about OI, referring to it as the Bernie Lomax distribution, which I think is
quite apt:'s

I don't think the project is going to succeed unless the various interested parties come together and
figure out who is responsible for what. People are going to have to step up and take responsibility,
otherwise it's just a lot of complaining and hot air about how nothing is happening.

Regarding the wiki directly, various people, myself included, have admin accounts and can create more. If
you're volunteering, I'm happy to set you up with one. If you want access to the zone confluence is running
in, I can provide that also.

Not that I'm involved any more and I largely just lurk, but I think the disconnect between /dev and /hipster
needs to end. It's confusing.

I have proposed for years now that:

/hipster = rolling release
/dev = snapshots of /hipster
/release = /periodic snapshots of /dev that are considered more stable

For example you could do automatic /dev releases every 2 weeks. /release can come out once a year, and in the
month running up to a /release, you can focus on fixes rather than new features.

Easy, simple.

It does mean /dev and Jon Tibble's effort making way for Andrzej/ALP/etc's hipster effort. The first
/release could be based on /dev as is now, but after that, my personal opinion is that Jon Tibble should help
with the hipster effort. Perhaps in particular with ensuring quality /release releases and managing
that bug fixing process.

Also some of the peanut gallery posts on this mailing list make me want to throw up. I don't think anyone
should be allowed to attend an OI meeting unless they have contributed at least X months worth of commits to
the OI github account. Talk is cheap, and people should have to earn the right to have an opinion on how the
project is run.

Back when I was project lead, I made the mistake of soliciting input from all interested parties, which
resulted in enormous weekly meetings with lots of talk and no action. It killed the project, as it became
mired in indecision and a total lack of focus. What is needed is a single minded lazer sharp focus.

The project is on life support. Commit or GTFO. 		 	   		  
Nikola M. | 6 Oct 11:24 2014

Re: Resignation

On 10/ 6/14 09:54 AM, Alexander Pyhalov wrote:
> Hello.
> On 10/06/2014 11:33, Nikola M. wrote:
>> But not creating Teams is I think biggest problem of all.
> Whom do you suppose these teams will consist of? Honestly, now we have 
> two informal teams -  RE - Adrzej Szeszo and Ken Mays and dev team 
> (which now consists from Jon Tibble (which supports /dev) and me 
> supporting Hipster). Plus several developers who are interested in 
> several packages and support mostly only these packages (as Aurelian 
> Larcher and Marcel Telka) This is the current state. We don't have 
> problem how to organize people. We just have no people to organize.
Well great, first information people need to activate themselves is what 
structure already exist.
You provided information on current structure and that is great.

We (meaning me shooting ideas all the time and you for starters) should 
talk about what to do.

I propose reviving OI weekly meetings that are always scheduled on 
Thursdays, 7PM UTC
at Freenode IRC channel #oi-meeting (

As topics I propose:
-Creating Hipster mailing list for pre-dev discussions and changes on 
packages, before releasing /dev.
-Making Hipster snapshots in-line with /dev releases, both from upgrade 
to Hipster form a8 as first phase
- Releasing a10 /dev and update to it from a9 as second phase.
- Not abandoning SPARC support contribution
- Mailing list inclusions of exlcluded people
- Plans for including more people in a structure and contribution docs
Everything else that I currently don't have time to write about.

Alexander Pyhalov | 30 Sep 13:52 2014

sic_team integration

Please, review this work:

This moves sic_team components 
(mozilla-nss,mozilla-nspr,java-jss,java-commons-logging) from oi-build 
to oi-userland, modifies them so that they use GCC and updates to
commons-logging 1.2
jss 4.3.1
nspr 4.10.7
nss 3.16.5

I've run testsuite for nss, it successfully passed.
Rebuilt illumos-gate (it has dependencies on nss/nspr).
Verified that basic pktool commands work with nss backend.

Best regards,
Alexander Pyhalov,
system administrator of Southern Federal University IT department
Aurélien Larcher | 29 Sep 14:31 2014

GDB in hipster

is anyone able to use gdb from the latest hipster package ?
Whenever I tried to load and run an executable I get:

procfs:3961 -- process not stopped.
procfs: up...
(gdb) bt
Target is executing.

But this is actually not the case.
I had compiled 7.6.x myself before and had the same issue, thus avoided it as I did not have the time to investigate.
Thank you,


LARCHER Aurélien          | KTH, School of Computer Science and Communication
Work: +46 (0) 8 790 71 42 | Lindstedtsvägen 5, Plan 4, 100 44 Stockholm, SWEDEN
oi-dev mailing list
oi-dev <at>
Alexander Pyhalov | 26 Sep 21:33 2014

Re: [OpenIndiana-discuss] Installing OpenOffice pkg://

russell писал 26.09.2014 21:47:
> Good Evening Alexander
> Can you advise a way to replace the
> consolidation/sunpro/sunpro-incorporation so that I can install
> OpenOffice.
> Regards

Brief answer: it's not possible.
Longer one... You have updated from 151a9, which is currently an 
unsupported path.
So you have, for example, developer/build/make <at> 0.5.11- 
installed (and other).
And sunpro-incorporation in Hipster has incorporate dependency on 
make <at> 0.5.11-
It means, only this version can be installed. Perhaps, if you downgraded 
it, the dependency
would be satisfied. But I bet on OI /dev a9 there are packages which has 
require dependency on make <at> 0.5.11-
(or what else). It means they can be satisfied by installing only 
make <at> 0.5.11- or higher.
But Hipster version is make <at> 0.5.11- So, IPS can't find 

Guys, what do you think about this? First of all, why can't we just make 
all incorporations
empty packages? What harm can it cause?

Currently we have the following incorporations:

pkg:/consolidation/X/X-incorporation <at> 0.5.11-2014.0.1.0
I think can be relaxed

pkg:/consolidation/admin/admin-incorporation <at> 0.5.11-2014.0.1.0
I think can be relaxed

pkg:/consolidation/cacao/cacao-incorporation <at> 0.5.11-
What is it?

pkg:/consolidation/cde/cde-incorporation <at> 0.5.11-
Perhaps, we could relax it, it seems to be something which is
going to stuck here, we can't do anything with it.
Or can we drop it one time?

pkg:/consolidation/cns/cns-incorporation <at> 0.5.11-
What is it?

pkg:/consolidation/dbtg/dbtg-incorporation <at> 0.5.11-
What is it? Could it be relaxed?

pkg:/consolidation/gfx/gfx-incorporation <at> 0.5.11-
What is it?

pkg:/consolidation/hcts/hcts-incorporation <at> 0.5.11-
It's ddu. Perhaps, we can relax it?

pkg:/consolidation/install/install-incorporation <at> 0.5.11-2014.0.1.942
There should be nothing wrong with this one, it's from slim_source.

Some java libraries.. Is it closed? I think we can relax it.

Perhaps we can just obsolete it?  I see nothing depending on it.

driver/network/ce is unlikely to be updated. I think, we can just relax 


I think, I'll relax it in the nearest future. Or is it better to 
obsolete it?

I think, we can relax it.

I think we can relax it.

We should do something with it. Do I understand correctly, that we can't 
java now? Perhaps, we can relax it?

I see that we have vpanels in oi-userland, but they are disabled.
Perhaps we should enable them? Who knows what is it? Some GUI admin 
tool? Will it work
on OI and do we need them?

> Russell
> On 14/09/2014 20:23, russell wrote:
>> Hi Alexander,
>> # pkg list | grep osnet
>> consolidation/osnet/osnet-incorporation           
>> 0.5.11-2014.1.0.14528
>> # pkg install
>> consolidation/sunpro/sunpro-incorporation <at> 0.5.11,5.11-2014.1.0.0
>> Creating Plan (Solver setup): |
>> pkg install: The installed package developer/gnu is not permissible.
>>    Reject:
>> pkg:// <at> 0.5.11,5.11-
>>    Reason:  All versions matching 'require' dependency
>> pkg:/developer/build/make <at> 0.5.11,5.11- are rejected
>>      Reject:
>> pkg:// <at> 0.5.11,5.11-

oi-dev mailing list
oi-dev <at>
Andrew M. Hettinger | 26 Sep 00:46 2014

SSD-based pools

I'm presently running tests on a pool using 3x Samsung 850 SSDs on a LSI-9211-8i (IT) contoller. I thought I'd try seperating the intent log to see if lowering the write amplification on the pool-drives would help, so I added another matching SSD for that, but under load I still seem to get extensive checksum errors. Does anyone have any ideas as to what would be causing this?

  pool: test-array
 state: DEGRADED
status: One or more devices has experienced an unrecoverable error.  An
        attempt was made to correct the error.  Applications are unaffected.
action: Determine if the device needs to be replaced, and clear the errors
        using 'zpool clear' or replace the device with 'zpool replace'.
  scan: scrub repaired 0 in 0h0m with 0 errors on Wed Sep 24 18:20:56 2014

        NAME                       STATE     READ WRITE CKSUM
        test-array                 DEGRADED     0     0     0
          mirror-0                 DEGRADED     0     0     0
            c0t50025388700060D4d0  DEGRADED     0     0   155  too many errors
            c0t50025388700060AEd0  DEGRADED     0     0   149  too many errors
            c0t50025388700060C2d0  DEGRADED     0     0   174  too many errors
          c0t50025388A067DBE9d0    ONLINE       0     0     0

errors: No known data errors
  ---- errors ---
  s/w h/w trn tot device
    0   2   6   8 c0t50025388700060D4d0
    0   0   0   0 c0t50025388700060AEd0
    0   0   0   0 c0t50025388700060C2d0
    0   0   0   0 c0t50025388A067DBE9d0

Andrew Hettinger
http://Prominic.NET | Skype: AndrewProminic
Tel: 866.339.3169 (toll free) -or- 1.217.356.2888 x. 110 (int'l)
Fax: 866.372.3356 (toll free) -or- 1.217.356.3356            (int'l)

oi-dev mailing list
oi-dev <at>