Re: When do we update the homepage to a modern design? (was Re: Moving away from (unsupportable) FusionForge on Alioth)

[Игорь Пашев]

2017-05-15 13:12 GMT+03:00 lumin :
> Especially look at the homepage of Gentoo


It's ugly, seriously.

Re: Mass bug filing about non free lena image.

[Игорь Пашев]

2015-08-12 16:32 GMT+03:00 Andreas Tille andr...@an3as.eu:
 Any productive suggestion?


Take a picture of one of Debian women ? :-)

Re: Facilitating external repositories

[Игорь Пашев]

2015-06-05 19:10 GMT+03:00 Josh Triplett j...@joshtriplett.org:
 Given that the packages in question appear to be Free Software (at least
 from a quick check of a couple of them, as well as the repository being
 named main), is there a reason you don't maintain them in Debian
 (including backports or volatile if you need to provide the newest
 packages for older distributions)?


It takes ages to pass mentors.debian.net? :-)


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/CALL-Q8zvSaHgMHJUOPr49Ub3aUCQPYUCKUR32=dh1g6bzmn...@mail.gmail.com

Re: Please more fish (was: so long and thanks for all the fish)

[Игорь Пашев]

2014-11-10 0:38 GMT+03:00 Simon Richter s...@debian.org:
 automake

With autotools one can always use plain shell code in configure.ac and
plain make in Makefile.am ;-)


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/call-q8zejcqrokhyk0gxsczk+p6oz7ft8ro3jlawlkucmle...@mail.gmail.com

Fwd: open-axiom is marked for autoremoval from testing

[Игорь Пашев]

Is it really a good idea to remove packages which FTBFS because of
*internal compiler error*?

Shouldn't GCC be removed instead? :-)


-- Forwarded message --
From: Debian testing autoremoval watch nore...@release.debian.org
Date: 2014-09-29 8:39 GMT+04:00
Subject: open-axiom is marked for autoremoval from testing
To: open-ax...@packages.debian.org


open-axiom 1.5.0~svn3056+ds-1 is marked for autoremoval from testing
on 2014-10-13

It is affected by these RC bugs:
761549: open-axiom: FTBFS: internal compiler error: in
cp_perform_integral_promotions, at cp/typeck.c:2066


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/CALL-Q8z9xhoaHfM6T+fgTZmyRLr4vDc+dkVunrX7XrduNxA=j...@mail.gmail.com

Re: -- APT's New Version --

[Игорь Пашев]

As for me I'd like to have source and clean subcommands for apt ;-)

(For aptitude too)


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/CALL-Q8xK-8RoFPcTn8T1FM=oj9dols6m1i2bfu12sudvhhh...@mail.gmail.com

Re: OpenRC

[Игорь Пашев]

The discussions on init system have discovered much energy of
developers and users,
so I think they are able to use that energy to support multiple systems :-)


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8xWFX7X95jG8H+SRWOc-tFH6EDhpPP1Y8dmKSxb=v=_...@mail.gmail.com

Re: How can I make this work

[Игорь Пашев]

2014-01-29 Roelof Wobben rwob...@hotmail.com:
 Now it fails with this message :

 /usr/bin/ld: SDL_2dgl.o: undefined reference to symbol 'floor@@GLIBC_2.2.5'
 //lib/x86_64-linux-gnu/libm.so.6: error adding symbols: DSO missing from
 command line
 collect2: error: ld returned 1 exit status
 make[3]: *** [gnujump] Fout 1
 make[3]: Map '/home/roelof/Downloads/gnujump-1.0.8/src' wordt verlaten
 make[2]: *** [all-recursive] Fout 1

Hi, gnujump should be linked to libm for floor()

In Debian package (gnujump) there is a patch to add libm:
http://patch-tracker.debian.org/patch/series/view/gnujump/1.0.8-1/06_link_against_-lm_.patch


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8zhhmjupb+qx-cvke1vv9enbajmwbyosmkvfof4t12...@mail.gmail.com

Re: GnuTLS in Debian

[Игорь Пашев]

Do I understand correctly the following:

Application M under the MIT license linked to LGPL3 library L - ok
Application C under the CDDL license linked to LGPL3 library L - ok
Application G under the GPL3 license linked to LGPL3 library L - ok,
all under GPL3

Bang!

Application M is now under the GPL3 ?
Application C is now illegally linked to L ?


:-)


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8ztfuhcVUxmkqg7qWVBdVLfs-dcMeK=7nvkzz5nfnq...@mail.gmail.com

Re: can i have the main os files in zip file?

[Игорь Пашев]

2014/1/3 freddie simmonds frederick.simmo...@aspects.net:
 i want to make my own oprating system from debian system for a certain usb
 device.
 however, trying to accses the files from a disc, i cant find them.

Majority of files are in packages, see
http://www.debian.org/doc/manuals/debian-faq/ch-pkg_basics


 can you give me a copy of the files for the os? if so, in zip file as well?

To create your custom OS from Debian you can start learning Debootstrap:
https://wiki.debian.org/Debootstrap


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8y_vb4dbqjm-5vtuaorgcbof_44eg_xrxdebc+uc7d...@mail.gmail.com

Bug#733651: general: Any USB card reader works only after being replugged.

[Игорь Пашев]

I can confirm this.

# uname -a
Linux hope 3.10-2-amd64 #1 SMP Debian 3.10.7-1 (2013-08-17) x86_64 GNU/Linux

# cat /etc/debian_version
jessie/sid

udev/204-5


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8xmqr1tuu17bhbdk8suowxdrawze4rs4yr2porkwbx...@mail.gmail.com

Re: Bug#732159: Should this package be removed?

[Игорь Пашев]

2013/12/21 John Paul Adrian Glaubitz glaub...@physik.fu-berlin.de:
 mplayer2 is badly maintained as compared to mplayer, don't know
 anything about mpv though. Just learned about it as of today,
 in fact.

mplayer2 looks more portable to me


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8zwf+knqzQAq0CSCSn4z83texsmra18NoE9=hjynw7...@mail.gmail.com

Re: debconf as a registry

[Игорь Пашев]

2013/11/27 Joey Hess jo...@debian.org:
 I'd think it would be obvious why it's not good design to put parsers
 for every possible config file format in debconf.

There is libaugeas.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8ypsa2ljf0yud0cc7uzucvgb6yz4ohpfb9gr6jclzx...@mail.gmail.com

Re: Bug#727708: tech-ctte: Decide which init system to default to in Debian.

[Игорь Пашев]

2013/10/30 Helmut Grohne hel...@subdivi.de:
 What is going to happen with non-Linux ports?

Debian is not Debian without non-Linux ports.

As for me, I think it is not very hard to maintain diffrent init
systems for different kernels.
Especially if Debian GNU/Linux get rid of sysvinit: writing systemd or upstrart
services is simple (as well as SMF).

Just one example from OSDyson (lighttpd):

1. dh-smf: http://cgit.osdyson.org/dh-smf.git
2. lighttpd depends on dh-smf for illumos kernel:
http://cgit.osdyson.org/lighttpd.git/tree/debian/control#n10
3. No changes in d/rules (!):
http://cgit.osdyson.org/lighttpd.git/tree/debian/rules
4. DH automatically peeks up dh_smf:
http://cgit.osdyson.org/debhelper.git/tree/dh?id=3769023faf4758f944e710480c43cda220821690#n524
5. dh_smf looks into this directory:
http://cgit.osdyson.org/lighttpd.git/tree/debian/lighttpd.smf
6. dh_installinit is no-op on Dyson


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8zvzx9lr74v6erazu6dgj8ez6kvyigzswq+v77fbwr...@mail.gmail.com

Re: Anybody using quilt?

[Игорь Пашев]

https://wiki.debian.org/Projects/DebSrc3.0

Anything other looks bad :-)



2013/8/26 Svante Signell svante.sign...@gmail.com

 Hi,

 Is any of you Debian maintainers/developers using guilt (qit+quilt)for
 patch management/developement? Is it good or bad? If you are not, what
 do you use?

 Thanks!



 --
 To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
 with a subject of unsubscribe. Trouble? Contact
 listmas...@lists.debian.org
 Archive:
 http://lists.debian.org/1377523260.22719.190.ca...@s1499.it.kth.se

Re: pidof changing from sysvinit-devel to procps

[Игорь Пашев]

Since we are talking about pidof, I'd like to note that pgrep is more
portable ;-)

2013/8/9, Craig Small csm...@debian.org:
 Besides my Debian duties I am also upstream for procps. I have been in
 discussion with the sysvinit-tools upstream and they want to find a new
 home for pidof so it fits with similiar tools (pidof used to be in
 procps in the dark ages). This means shortly that pidof will disappear
 from sysvinit-tools and appear in procps.

 If your package uses pidof, we need to talk about it NOW so that this
 change doesn't put you in the lurch. I believe merely depending on procps
 will do what is needed, with the right version.

 If your package uses, or you have a strong case for, non-LSB pidof flags
 then it is essential you speak up. The command line options that may be
 going are -c -n -m  This is not strictly a Debian thing so you can
 always speak up about the options at [1].

 For most people (hopefully) this change should be invisible; but for the
 minority that it's important, now is the time.

  - Craig
 [1] http://www.freelists.org/archive/procps

 --
 Craig Small VK2XLZ   http://enc.com.au/  csmall at : enc.com.au
 Debian GNU/Linux http://www.debian.org/  csmall at : debian.org
 GPG fingerprint: 5D2F B320 B825 D939 04D2  0519 3938 F96B DF50 FEA5


 --
 To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
 with a subject of unsubscribe. Trouble? Contact
 listmas...@lists.debian.org
 Archive: http://lists.debian.org/20130809111050.ga6...@enc.com.au




-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8zS_Vr=Eore=AU2TU3vXuBNVbPptqzBH=1pwza1r2w...@mail.gmail.com

Re: Survey answers part 3: systemd is not portable and what this means for our ports

[Игорь Пашев]

2013/7/22 Roger Leigh rle...@codelibre.net:
  We would be effectively locked in.

We are locked in sysvinit.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8xfw+ypmuiwagmn0fmvs1ovc+zc4obdbbserurudv-...@mail.gmail.com

Re: Survey answers part 3: systemd is not portable and what this means for our ports

[Игорь Пашев]

2013/7/18 William Giokas 1007...@gmail.com:
 Having not used OpenRC, I have no comment on the real world advantages
 or disadvantages of either init system


I'm a user of Gentoo and Debian.

I do not care of what to type: 'emerge -avuND world' or 'apt-get upgrade'
I do not care of which init system is used: both just work.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8yqeb3vwqo32uhs8dxqsovkakuytekzowkqkuzww_v...@mail.gmail.com

Re: SONAME migration: from liblambda0 to liblambda1

[Игорь Пашев]

2013/7/15 Jerome BENOIT g62993...@rezozer.net:
 the former binary package
 liblambda0 is not discarded by the new binary package liblambda1, neither in 
 Sid
 nor at the upgrading stage with aptitude.
 I guess that something is missing in `debian/control', any clue ?

I think this is intentional. And this is why package names follow SONAMES.
If old package is removed all dependent packages will be broken.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8xjhtnwuwgv3eaev+qk-pzzh9f5yv619ozexaegjno...@mail.gmail.com

Re: Survey answers part 3: systemd is not portable and what this means for our ports

[Игорь Пашев]

Why not to use different init systems on different kernels?
Debian already supports 3 (three) init systems *at once*, sysvinit,
upstart, systemd.
This is much harder that using single system.

FYI, on Dyson [1] I've made dh_installinit noop, and working on dh-smf [2]


[1] http://osdyson.org
[2] http://cgit.osdyson.org/dh-smf.git


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8zaRq=VOdjj9E7ED3eEP7R1CQ6=2gnrop8deb3jhgd...@mail.gmail.com

Re: Survey answers part 3: systemd is not portable and what this means for our ports

[Игорь Пашев]

2013/7/14 Marco d'Itri m...@linux.it:
 which is a waste of time and adds code which cannot
 be well tested.


Isn't Debian itself is a  waste of time, while we have RedHat? :-P

Let a hundred flowers bloom; let a hundred schools of thought contend.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8zxs_0q2uofjz-+uryhs_ism7fr8je446uceetm1dk...@mail.gmail.com

Re: 7.0- 7.1: any reasons for switching from {4,5,6}.0.x scheme?

[Игорь Пашев]

2013/6/17 Neil McGovern ne...@debian.org:
 #712586


 So... why is postgres relying on an undocumented numbering schema, and
 breaking when it doesn't match what it expects?


Why it checks for version at all? Isn't package dependencies should be used?

What about Debian derivatives?


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8zboJsyFzYck27ZUA-PT1OnVwHkiVeds5ZE=wtmqua...@mail.gmail.com

Re: DH way to set SONAME

[Игорь Пашев]

Changing package name is not enough. You should patch the library
build system to produce different soname and library filename

2013/6/9, Jerome BENOIT g62993...@rezozer.net:
 Hello List,

 because of backward incompatibility I have to add a suffix to
 my library package name. But lintian keep noticing the mismatch.
 I know the the soname of a library can be specify by passing
 appropriate options to the gcc tools: is a DH way to do so ?

 Thanks in advance,
 Jerome


 --
 To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
 with a subject of unsubscribe. Trouble? Contact
 listmas...@lists.debian.org
 Archive: http://lists.debian.org/51b41a07.5060...@rezozer.net




-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8yT+voGByvyLebx=FCsUcoLR=vweg_igta9-0get4g...@mail.gmail.com

Re: systemd .service file conversion

[Игорь Пашев]

2013/5/31 Jeff Epler jep...@unpythonic.net:
 Yes and yes.  Wheezy kfreebsd amd64 is dandy for server and OK for some
 minor graphical desktop stuff (opengl is not in a good state right now,
 at least with nvidia hardware: nouveau is no-go due to not having kernel
 support and proprietary won't install).  if you want zfs on debian
 (which is what I wanted) it's probably a better choice today than debian
 linux with zfs-on-linux. (at least, that's the call I made)


Debian GNU/kopensolaris is coming too ;-)

As for init system, my point is having different init systems for
different kernels is ok.
Having different init systems for the same kernel is pain in ass.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8wYp-8g_10snjquRMjqhw=9+5s2o6ixh6b7eb1_kn5...@mail.gmail.com

Re: optimizing PNGs

[Игорь Пашев]

2013/5/29 Paul Wise p...@debian.org:
 Does that mean switching away from autotools?
No.

 Or not shipping configure/Makefile.in in the tarballs?
No.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8xssprcxgasvfqribbp4_pz4sfhry+hkvzuwvliinw...@mail.gmail.com

Re: default MTA

[Игорь Пашев]

2013/5/28 Josselin Mouette j...@debian.org:
 Nobody,
 I repeat, nobody, ever reads local mail on most desktop systems


I read mail from my desktop :-)
It is forwarded to my gmail account


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8xL8vibh9w7eaUVf9kfGxhzzLqjiBiM9Xo=_w-jxea...@mail.gmail.com

Re: systemd .service file conversion

[Игорь Пашев]

2013/5/27 brian m. carlson sand...@crustytoothpaste.net:
 At the risk of adding another level of indirection, we could add a
 meta-init format that can generate an appropriate file for any of these.


http://xkcd.com/927/


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8xqutu4eznaxhg-amfzklwfn7xjcas3w33mv6wevbd...@mail.gmail.com

Re: systemd .service file conversion

[Игорь Пашев]

2013/5/23 Helmut Grohne hel...@subdivi.de:
 * stdout/stderr to syslog redirection
This is possibly implementable, but needs more than a line of shell.


In Solaris SMF each service has its own log file with SMF messages
*and* all stdout/stderr


pashev@bok:~$ find /var/log/svc/
/var/log/svc/
/var/log/svc/network-initial:default.log
/var/log/svc/svc.startd.log
/var/log/svc/milestone-single-user:default.log
/var/log/svc/system-svc-global:default.log
/var/log/svc/network-service:default.log
/var/log/svc/milestone-name-services:default.log
/var/log/svc/system-console-login:vt4.log
/var/log/svc/system-console-login:vt3.log
/var/log/svc/system-sysevent:default.log
...

pashev@bok:~$ tail /var/log/svc/system-cron:default.log
[ May 13 03:19:21 Method start exited with status 0. ]
[ May 14 10:12:38 Enabled. ]
[ May 14 10:12:49 Executing start method (/lib/svc/method/cron). ]
[ May 14 10:12:50 Method start exited with status 0. ]
[ May 15 15:11:06 Enabled. ]
[ May 15 15:11:18 Executing start method (/lib/svc/method/cron). ]
[ May 15 15:11:18 Method start exited with status 0. ]
[ May 15 15:15:22 Enabled. ]
[ May 15 15:15:33 Executing start method (/lib/svc/method/cron). ]
[ May 15 15:15:34 Method start exited with status 0. ]


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8y7q4s5bhrp1tbxouih1wd-oycdhr6tcogmtz8ardx...@mail.gmail.com

Re: Packaging releases without a tarball (sometimes)

[Игорь Пашев]

Put git updates as a patch, e. g. debian/patches/git-update.patch
Bump debian part of version as usual.

AFAIK python is packaged in this way


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8z5b_z-anu67jprgyme2v94q56xds8xmoh0ag1bcfo...@mail.gmail.com

Re: Temporary solution for changelog problem in binNMUs

[Игорь Пашев]

2013/5/14 Raphael Hertzog hert...@debian.org:
 But ansgar's objection about the duplication of the changelog in multiple
 .deb when it used to be shared via a symlink also makes sense. As does the
 fact that there's currently no way to not install some control files.


What about creating symlinks in /var/lib/dpkg/info/ ?

/var/lib/dpkg/info/pkg1.changelog - pkg2.changelog


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8ws96ugg_gv-f5nb9ihcq-70wxrtmsm4vq3fowyrpy...@mail.gmail.com

Re: systemd^wfoo on linux, bar on bsd,so what (Re: /bin/sh (was Re: jessie release goals)

[Игорь Пашев]

2013/5/13 Philipp Kern pk...@debian.org:
 On Mon, May 13, 2013 at 08:46:23AM +0100, Roger Leigh wrote:
 There is no need for udev to be dependent upon a specific init
 system, other than laziness.

 Except if you want to receive device plug events as triggers to start
 up / shut down services. The separation then gets quite blurry with
 whom should handle events and what packages should ship to handle
 events. Bonus points for dependency handling between services and
 hotplug events.


AFAIK in Solaris it is implemented with syseventd and does depend on
init system.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8znk9a7duqm+q0tts4_yh4+dgk756kf508cxavrocy...@mail.gmail.com

Re: systemd^wfoo on linux, bar on bsd,so what (Re: /bin/sh (was Re: jessie release goals)

[Игорь Пашев]

2013/5/14 Игорь Пашев pashev.i...@gmail.com:
 2013/5/13 Philipp Kern pk...@debian.org:
 On Mon, May 13, 2013 at 08:46:23AM +0100, Roger Leigh wrote:
 There is no need for udev to be dependent upon a specific init
 system, other than laziness.

 Except if you want to receive device plug events as triggers to start
 up / shut down services. The separation then gets quite blurry with
 whom should handle events and what packages should ship to handle
 events. Bonus points for dependency handling between services and
 hotplug events.


 AFAIK in Solaris it is implemented with syseventd and does depend on
 init system.

Please, ignore my previous message :-)


--
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8wwxDMMFG+iGi_9sgtoY0LFVYZeyLmJFNc9OqFxen=n...@mail.gmail.com

Re: Merging / and /usr (was: jessie release goals)

[Игорь Пашев]

2013/5/12 Stéphane Glondu glo...@debian.org:

 What about /etc ? /var ? both contain data that can mess up with a
 running system...

All go into a snapshot.

That's why I stand for moving /var/lib/mysql and similar things out of /var


--
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8yaNGi2Nv5g+7OX-GRHakMh68eUePrD+MvJc6=kfhz...@mail.gmail.com

Re: /bin/sh (was Re: jessie release goals)

[Игорь Пашев]

2013/5/12 John Paul Adrian Glaubitz glaub...@physik.fu-berlin.de:
 Honestly, you simply can't expect every single package in Debian to run on
 any of the supported kernels. If systemd profits from the use of
 Linux-specific kernel features, which is a good thing in my humble opinion
 because Linux has many very advanced and useful features, it should use
 them.


 I already think that it's a bit crazy (and geeky) that we have the capability 
 in Debian to choose between different kernels, so I guess it's natural that 
 we have discussions about being able to make choices for init as well :).

Yeah, that's quite easy ;-)
http://cgit.osdyson.org/rsyslog.git/tree/debian/rules#n55


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8wxh8qs7x63gxph55otexblt61nck_guxshf9ojmpp...@mail.gmail.com

Re: Merging / and /usr (was: jessie release goals)

[Игорь Пашев]

2013/5/11 Aron Xu a...@debian.org:
 An easy example is that, on Solaris, there is a something called boot
 environment (BE), which is essentially snapshots of the combination of
 /usr and /boot, users can switch between different BEs easily without
 affecting any user data. Without /usr merge, doing such work could be
 much more complicated because user data and system data is mixed in
 the file system's hierarchy, it's hard to make sure switching between
 different snapshots won't change user data.


Aron, merging / and /usr is not related to the ability of creating
BEs. The beadm utility
will create a snapshot of / and all underlying filesystems. Even
/var/tmp (if it is separated fs)
should be a part of BE (not shared between BEs), because of FHS..


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8xe3jwmqyw_9b1yeqqurcbbw5cggspbmqopaak+vb1...@mail.gmail.com

Re: /bin/sh (was Re: jessie release goals)

[Игорь Пашев]

2013/5/11 Josselin Mouette j...@debian.org:
 We have had two releases with kfreebsd, which failed to
 provide anything usable. Debian is only about Linux, and has always
 been.


I have some news about it ;-)


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8wC9UYa=3g-5vuphm1reen+qs0rournmxwzeuupmdu...@mail.gmail.com

Re: /bin/sh (was Re: jessie release goals)

[Игорь Пашев]

2013/5/12 Josselin Mouette j...@debian.org:
 GNOME depends on a working glibc, too. Does it dictate the C library?


Yes. Portability still makes sense. Portability is a part of the word
Free in Free Software.

Debian is about Free Software.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8yededi4ctvhrvt0qucke5te1jrnzza5jaaeaguewb...@mail.gmail.com

Re: /srv (was /export)

[Игорь Пашев]

2013/5/8 Thorsten Glaser t...@debian.org:
 Игорь Пашев pashev.igor at gmail.com writes:

 What about moving default location for applications to /export ?

 *NO*!

Maybe not /export, but /srv, which already exists, the idea is
separating system data and application data.
(Actually, I do not like /export, because of /etc and tab completion ;-)

And probably /export is wrong place for this :-/ [1]


 We do not need *more* such non-Unixy, FLOS, nōnsense!

 I seem to understand you’re coming from Solaris here, which has a
 history of doing weird things like that…? Please, no.
I know how weird Solaris is, no need to tell me :-)
But referring to true-UNIX also does not make any sense.

[1] http://lists.linux-foundation.org/pipermail/fhs-discuss/2011-May/000158.html


--
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8zknircpf1l1dvtmbge-rg1pdgt3slrbjfjx2tkm2k...@mail.gmail.com

Re: /export (was Re: jessie release goals)

[Игорь Пашев]

2013/5/8 Peter Samuelson pe...@p12n.org:
 Wrong list, please bring this up instead on fhs-discuss.  (If that
 still exists - it looks a bit stale - but anyway, debian-devel is not
 for FHS discussion.)


I'd like to have an open discussion.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8z8mjonp88vmvm1_xuqwtjawu53gkdeertuwefjv8e...@mail.gmail.com

Re: /export (was Re: jessie release goals)

[Игорь Пашев]

2013/5/8 Thomas Goirand z...@debian.org:
 How about moving /home
 into /usr?


I proposed exactly an opposite thing for databases :-)

If do not like /usr/home, you might not like to have your data under
/var/lib ;-)


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8ypHeC=aqnsvfwwqbya7ohk6zu5zh4ye0efsv4dyc8...@mail.gmail.com

Re: /export (was Re: jessie release goals)

[Игорь Пашев]

2013/5/8 Josselin Mouette j...@debian.org:
 The FHS has the perfect place for such data: /srv. I agree we should
 move the data there, but there is no reason to invent a new place.

Yeah. /export was my typo ;-)


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8wu-tv86ab_2i1gham3m+t1jjvmso3-+c4c37_fccx...@mail.gmail.com

Re: /export (was Re: jessie release goals)

[Игорь Пашев]

2013/5/8 Andrey Rahmatullin w...@wrar.name:
 On Wed, May 08, 2013 at 03:51:33PM +0800, Thomas Goirand wrote:
 How about moving /etc into /home? How about moving /home
 into /usr?
 Note that / was moved to /usr and /usr to /home just for HDD size reasons
 [1].

 [1]: http://lists.busybox.net/pipermail/busybox/2010-December/074114.html

Separating system and user data now make sense for other reasons. but
still makes sense:

compression, encryption, mirroring, easy system reinstall, etc.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8xjun1vxkdsmj3u4c9fowvjmbqvnzselr56sfhrzvr...@mail.gmail.com

Setting up package vs triggers

[Игорь Пашев]

=== BEFORE =

$ sudo apt-get install dbus
Reading package lists... Done
Building dependency tree
Reading state information... Done
Suggested packages:
  dbus-x11
The following NEW packages will be installed:
  dbus
0 upgraded, 1 newly installed, 0 to remove and 0 not upgraded.
Need to get 0 B/378 kB of archives.
After this operation, 796 kB of additional disk space will be used.
Selecting previously unselected package dbus.
(Reading database ... 73608 files and directories currently installed.)
Unpacking dbus (from .../dbus_1.6.2-2+dyson1_illumos-amd64.deb) ...
Processing triggers for manifest-import ...
svccfg: Loaded 1 smf(5) service descriptions
Processing triggers for man-db ...
Setting up dbus (1.6.2-2+dyson1) ...


DBUS service faled to start:
maintenance17:24:17 svc:/system/dbus:default

Because it is started just after manifest import, before dbus is configured
(/etc/dbus-1/system.conf does not exist I guess)


== AFTER ==
after adding /etc/apt/apt.conf.d/triggers [1]:
DPkg::NoTriggers true;
PackageManager::Configure smart;
DPkg::ConfigurePending true;
DPkg::TriggersPending true;

$ sudo apt-get install dbus
Reading package lists... Done
Building dependency tree
Reading state information... Done
Suggested packages:
  dbus-x11
The following NEW packages will be installed:
  dbus
0 upgraded, 1 newly installed, 0 to remove and 0 not upgraded.
Need to get 0 B/378 kB of archives.
After this operation, 796 kB of additional disk space will be used.
Selecting previously unselected package dbus.
(Reading database ... 73608 files and directories currently installed.)
Unpacking dbus (from .../dbus_1.6.2-2+dyson1_illumos-amd64.deb) ...
Processing triggers for man-db ...
Setting up dbus (1.6.2-2+dyson1) ...
Processing triggers for manifest-import ...
svccfg: Loaded 1 smf(5) service descriptions


manifest-import is triggerred after dbus is configured, and dbus
service succesfully starts


Could anyone explain what is going on? :-)
Why man-db is still triggerred before configureing dbus?

[1] 
http://raphaelhertzog.com/2011/05/30/trying-to-make-dpkg-triggers-more-useful-and-less-painful/


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8zCsKkiuSq=e5bp4iku5wcb6mrhsonqrcaoe78ganz...@mail.gmail.com

Re: New version of libselinux makes libpcre3 pseudo-essential

[Игорь Пашев]

2013/5/8 Bastian Blank wa...@debian.org:
 On Tue, May 07, 2013 at 07:49:37PM +0400, Игорь Пашев wrote:
 Does it matter? If libselinux depends on libpcre3, libpcre3 will be pulled.
 I believed all libraries are optional.

 Please re-read the policy, especially 2.5:
 | Packages must not depend on packages with lower priority values
 | (excluding build-time dependencies).


Will it change anything except non-linux systems will get libpcre3 by
default, while do not need it?

Maybe make libselinux optional? ;-)


--
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8zf56jjpuqqxt8qvqueccrj-8dp7bofi-dyjohrgur...@mail.gmail.com

Merging / and /usr (was: jessie release goals)

[Игорь Пашев]

What about merging / and /usr ?


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8zwv5QtT05VA+cQg+OF=paijrt606x2shdte+b4qv2...@mail.gmail.com

Re: Merging / and /usr (was: jessie release goals)

[Игорь Пашев]

2013/5/7 Paul Tagliamonte paul...@debian.org:
 Not when you have a 500 meg internal storage that the firmware boots off
 of, and using a multi-gig CF card to store the mega-awesome-app you're
 using it for.

Similar to Live CDs where /usr used to be compressed on a CD?

http://cgit.osdyson.org/live.git/tree/protostar#n162


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8xeaqmxyl4y-fezmjjqp7ivobxoxzwnfzg2qffq021...@mail.gmail.com

Re: New version of libselinux makes libpcre3 pseudo-essential

[Игорь Пашев]

2013/5/7 Laurent Bigonville bi...@debian.org:
 Hello,

 As Guillem asked me in a bug report (#695620), I'm bringing this to
 debian-devel.

 The latest version of libselinux is now using libpcre3 instead of the
 regex functions from the glibc. This is improving the performances
 quite substantially when relabeling the filesystem.

 The version of libselinux in experimental is already using libpcre3.

 This would probably requires to bump the priority of libpcre3 from
 standard to required.

 What are your thoughts about this?


Does it matter? If libselinux depends on libpcre3, libpcre3 will be pulled.
I believed all libraries are optional.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8ywQ=eh+abm2_SCXNp=sgglpkvu+zdytscq4ipvpkh...@mail.gmail.com

/export (was Re: jessie release goals)

[Игорь Пашев]

Currently /var/lib contains data for system utilities (dpkg, apt,
aptitute) and for user application like databases.

What about moving default location for applications to /export ?

Think about modern fs like btrfs and zfs:

One may want to create a snapshot of the entire system, user data
should not get into such a snapshot,
but, of course, /usr along with dpkg database - should.

Again, the idea is to move all user data away from /var/lib


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8xg-ZpWDrpPa==buxsh+f-qg2pc5sqfnoifaaxtxsj...@mail.gmail.com

Re: Switching packages to non-awaiting triggers

[Игорь Пашев]

Hi, I think I know one more issue related to this, correct me if it is not.

This might be a problem for systemd, but I will talk about SMF [1] and
concrete example - dbus.

1. dbus package includes SMF manifest
2. for dbus service to appear, its manifest should be imported
3. importing manifests if done via triggers (AFAIK like in systemd package)
4. dbus service is enabled by default (will start automatically after
importing the manifest)

so if installing dbus package, dbus service failed to start because of
missing /etc/dbus-1/system.conf.

This looks like trigger is activated before dbus package is configured.

Is it correct?


[1] http://ru.wikipedia.org/wiki/Service_Management_Facility


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8xhdiSq=iX__CGXJcXrQo92wTdsST1NU=zjzswydjv...@mail.gmail.com

Re: Switching packages to non-awaiting triggers

[Игорь Пашев]

2013/5/7 Michael Biebl bi...@debian.org:
 We don't use dpkg triggers to start (systemd) services.

Hm?

http://anonscm.debian.org/gitweb/?p=pkg-systemd/systemd.git;a=blob;f=debian/systemd.triggers;h=f6c3c0ce0da08815f366868d3b3a3f9e5907875f;hb=refs/heads/debian


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8z83YR7c_AnPVFX=0huqwlusxmna6qbvlc3-xmzuky...@mail.gmail.com

Re: Switching packages to non-awaiting triggers

[Игорь Пашев]

2013/5/7 Michael Biebl bi...@debian.org:
 If you actually looked at the package, you would have noticed that it has
 nothing to do with starting services.

Maybe, systemctl enable does not start the service?

Main question is: can triggers be activated before package is
configured (namely conffiles are in place)?


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8xiouh-oqzmpdcdyahu6zc9wjw62xmncahv_hnz0yo...@mail.gmail.com

Re: jessie release goals

[Игорь Пашев]

2013/5/6 Andrey Rahmatullin w...@wrar.name:
 There is also a related thing that was discussed in the past: stop
 disabling services via /etc/default.


Modern service management systems might be better choice.
E. g. systemd or SMF :-P


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8y4a7h7wijzq8opj2bcyxass5qj9kex83q9kjbbunx...@mail.gmail.com

Re: Debian on illumos (Dyson) LiveCD

[Игорь Пашев]

Hi, all!

In a shadow of Debian 7 release, Dyson - a Debian port to illumos kernel -
became self-hosted and now has an installer :-)

http://osdyson.org/projects/dyson/wiki/Dyson_Installer


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8x7FR0BJ_eHQgX-B5rAES=pjdivvfb-almnmo6frkp...@mail.gmail.com

Re: Debian on illumos (Dyson) LiveCD

[Игорь Пашев]

2013/5/6 Adam Borowski kilob...@angband.pl:

 I just tried it and ported some stuff to it, it seems that its libc is
 pretty bizarre.  For example math functions in C++ require quite a bunch
 of manual work (you need to explicitely cast all arguments unless they
 happen to be float, double or long double).

 Do you intend to switch to glibc?  If no, there'll be quite a lot of porting
 work (as most upstreams don't bother about Solaris).  If yes, most of that
 effort will become wasted.

Yes, I intend to switch to glibc.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8y57894A==dw4wdnbwlbetfpnf7wnyf9w0zzs7cz3s...@mail.gmail.com

Add multiarch library paths to DEFAULT_LIBRARY_PATH in Dpkg::Shlibs?

[Игорь Пашев]

Hi, all.

Now, /usr/share/perl5/Dpkg/Shlibs.pm defines default library search path as:
use constant DEFAULT_LIBRARY_PATH =
qw(/lib /usr/lib /lib32 /usr/lib32 /lib64 /usr/lib64
   /emul/ia32-linux/lib /emul/ia32-linux/usr/lib);

and adds multiarch dirs only for crosscompiling:

if ($crossprefix) {
push @crosslibrarypaths, /lib/$multiarch, /usr/lib/$multiarch,
/$crossprefix/lib, /usr/$crossprefix/lib,
/$crossprefix/lib32, /usr/$crossprefix/lib32,
/$crossprefix/lib64, /usr/$crossprefix/lib64;
}

Multiarch dirs are added below by parsing /etc/ld.so.conf:
parse_ldso_conf(/etc/ld.so.conf) if -e /etc/ld.so.conf;

I think it would be better to add multiarch dirs to DEFAULT_LIBRARY_PATH,
and put them in first positions.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8wzrdbs2xuajbn+zh8_bwmpmtfqe+tgzo1muicpxr0...@mail.gmail.com

Re: Add multiarch library paths to DEFAULT_LIBRARY_PATH in Dpkg::Shlibs?

[Игорь Пашев]

2013/4/16 Paul Wise p...@debian.org:
 On Tue, Apr 16, 2013 at 7:34 PM, Игорь Пашев wrote:

 Now, /usr/share/perl5/Dpkg/Shlibs.pm defines default library search path as:

 Sounds like a topic for the debian-dpkg list?

No.


--
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8xjWfE7_F0O�QEess_f-opFBEupW+zSR=ptekurk...@mail.gmail.com

Re: Add multiarch library paths to DEFAULT_LIBRARY_PATH in Dpkg::Shlibs?

[Игорь Пашев]

2013/4/16 Raphael Hertzog hert...@debian.org:
 On Tue, 16 Apr 2013, Игорь Пашев wrote:
 I think it would be better to add multiarch dirs to DEFAULT_LIBRARY_PATH,
 and put them in first positions.

 Why?

 This modules tries to mimick ld.so's logic to find libraries as closely
 as possible.

/lib:/usr/lib was the default path, now it is
/lib/multiarch:/usr/lib/multiarch, isn't it?


--
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8zaupp+sgpramfyg7+ak7jmmu-1h2kf-udo2nx0jy0...@mail.gmail.com

Re: Add multiarch library paths to DEFAULT_LIBRARY_PATH in Dpkg::Shlibs?

[Игорь Пашев]

2013/4/16 Neil Williams codeh...@debian.org:
 On Tue, 16 Apr 2013 16:11:24 +0400

 No - there's confusion here between the runtime link path and the
 build time link path. Dpkg::Shlibs at the point quoted is concerned
 with the build time paths. Where to find the stuff which has been
 linked against the built binaries for the current architecture of the
 current build.

 The shared library libfoo.so.1.2.3 lives in /usr/lib/$triplet for
 runtime usage but the -dev package which provides the .so symlink which
 is used to find the actual .so.N.N.N file is not necessarily in a
 MultiArch path. The implementation of MultiArch for -dev packages is
 not complete and most -dev packages are not co-installable with a
 foreign architecture of the same -dev package. (Any one package build is
 only building for a single architecture at any one time, discounting
 the issues with CC_FOR_BUILD which needs to be handled differently.
 Therefore, building bar for i386 and then armhf means removing some
 of the i386 build-deps and installing the armhf build-deps instead.
 This typically includes removing libfoo-dev:i386 and installing
 libfoo2:armhf  libfoo-dev:armhf. libfoo2:armhf can, of course, stay
 installed for later.)

 MultiArch in Debian is principally concerned with runtime paths, the
 build-time paths and consequent cross-compilation support still has a
 few wrinkles to resolve. (or dpkg-cross could have been removed from
 Wheezy.)


So, is it important, that multiarch dirs go after crossdirs:
our @librarypaths = (DEFAULT_LIBRARY_PATH, @crosslibrarypaths);
?


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8zz=a5bwt1w6rsnrbs7c1munfxsxjazrpghrr+k1y7...@mail.gmail.com

Debootstrap progress does not look monotonic

[Игорь Пашев]

Progress, reported by debootstrap does not look monotonic.

Do I miss something?

 .
 254
 255 P: 2243294 76407470
 256
 257 P: 2289076 76407470 - HERE
 258
 259 I: VALIDATING
 260
 261 IA: aptitude
 262
 263 IF: Validating %s
 264
 265 I: RETRIEVING
 266
 267 IA: aptitude-common
 268
 269 IF: Retrieving %s
 270
 271 P: 113904 76407470 AND HERE



-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8wywekx4npfgenfawc0i7ncqy4y9neyq1rcbh-vd+b...@mail.gmail.com

alternative debian/rules

[Игорь Пашев]

I've just realized that debian/rules might not be a makefile, but can
be a script in any language.

Is there any package using debian/rules whihc is not a makefile?
Are there any packages that can get advantages by using debian/rules
written in bash/perl/python/etc?


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8zdme0i22qwgjxjrbstqxehk03zgcr+xb+6d5z59n9...@mail.gmail.com

Re: alternative debian/rules

[Игорь Пашев]

2013/4/5 Adam Borowski kilob...@angband.pl:
 The policy says


Indeed.

So, in any case one can use its own tool just like dh:

%:
debian/megatool $@


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8z603EAhb86khzjAfOY1XvO42A+Li_V5mrjUx=w+uq...@mail.gmail.com

objcopy vs absolute addresses

[Игорь Пашев]

Hi, all.

While trying to build GRUB on illumos [*-pc-solaris2.11], I've found that
for different absolute addresses (-Wl,-Ttext -Wl,ADDR) objcopy -O binary
produces different objects files, but it is expected to be the same.

Test script see at the bottom.

I've checked it with binutils 2.22 on Debian and on Gentoo with
different compilers.

Debian's binutils have this problem, Gentoo:

Gentoo:
# ./objcopy-test
+ CC='gcc -m32'
+ OBJCOPY=objcopy
+ cat
+ gcc -m32 -c conftest.c -o conftest.o
+ for link_addr in 2000 8000 7C00
+ gcc -m32 -nostdlib -Wl,-N -Wl,-Ttext -Wl,2000 conftest.o -o conftest.exec.2000
/usr/lib/gcc/x86_64-pc-linux-gnu/4.6.3/../../../../x86_64-pc-linux-gnu/bin/ld:
warning: cannot find entry symbol _start; defaulting to
2000
+ objcopy -O binary conftest.exec.2000 conftest.2000
+ for link_addr in 2000 8000 7C00
+ gcc -m32 -nostdlib -Wl,-N -Wl,-Ttext -Wl,8000 conftest.o -o conftest.exec.8000
/usr/lib/gcc/x86_64-pc-linux-gnu/4.6.3/../../../../x86_64-pc-linux-gnu/bin/ld:
warning: cannot find entry symbol _start; defaulting to
8000
+ objcopy -O binary conftest.exec.8000 conftest.8000
+ for link_addr in 2000 8000 7C00
+ gcc -m32 -nostdlib -Wl,-N -Wl,-Ttext -Wl,7C00 conftest.o -o conftest.exec.7C00
/usr/lib/gcc/x86_64-pc-linux-gnu/4.6.3/../../../../x86_64-pc-linux-gnu/bin/ld:
warning: cannot find entry symbol _start; defaulting to
7c00
+ objcopy -O binary conftest.exec.7C00 conftest.7C00
+ md5sum conftest.2000 conftest.7C00 conftest.8000
05349871a5ab8bb0c860d3109d65f58d  conftest.2000
05349871a5ab8bb0c860d3109d65f58d  conftest.7C00
05349871a5ab8bb0c860d3109d65f58d  conftest.8000
+ exit 0


Debian (testing):
# ./objcopy-test
+ CC=gcc -m32
+ OBJCOPY=objcopy
+ cat
+ gcc -m32 -c conftest.c -o conftest.o
+ gcc -m32 -nostdlib -Wl,-N -Wl,-Ttext -Wl,2000 conftest.o -o conftest.exec.2000
/usr/bin/ld.bfd.real: warning: cannot find entry symbol _start;
defaulting to 2000
+ objcopy -O binary conftest.exec.2000 conftest.2000
+ gcc -m32 -nostdlib -Wl,-N -Wl,-Ttext -Wl,8000 conftest.o -o conftest.exec.8000
/usr/bin/ld.bfd.real: warning: cannot find entry symbol _start;
defaulting to 8000
+ objcopy -O binary conftest.exec.8000 conftest.8000
+ gcc -m32 -nostdlib -Wl,-N -Wl,-Ttext -Wl,7C00 conftest.o -o conftest.exec.7C00
/usr/bin/ld.bfd.real: warning: cannot find entry symbol _start;
defaulting to 7c00
+ objcopy -O binary conftest.exec.7C00 conftest.7C00
+ md5sum conftest.2000 conftest.7C00 conftest.8000
c2d829df4c6a7e335ab0e6a944ec87a3  conftest.2000
a6a514773337be95da81db75cdd011ba  conftest.7C00
5883efd0da029145374269a65f8d689b  conftest.8000
+ exit 0


I've event tested it on OpenIndiana (binutils 2.19, GCC 4.4):
# LD_ALTEXEC=/usr/bin/gld ./objcopy-test
+ CC='gcc -m32'
+ OBJCOPY=objcopy
+ cat
+ 1 conftest.c 0 \EOF
void
cmain (void)
{
   *((int *) 0x1000) = 2;
}
EOF
+ gcc -m32 -c conftest.c -o conftest.o
+ gcc -m32 -nostdlib -Wl,-N -Wl,-Ttext -Wl,2000 conftest.o -o conftest.exec.2000
/usr/bin/gld: warning: cannot find entry symbol _start; defaulting to
2000
+ objcopy -O binary conftest.exec.2000 conftest.2000
+ gcc -m32 -nostdlib -Wl,-N -Wl,-Ttext -Wl,8000 conftest.o -o conftest.exec.8000
/usr/bin/gld: warning: cannot find entry symbol _start; defaulting to
8000
+ objcopy -O binary conftest.exec.8000 conftest.8000
+ gcc -m32 -nostdlib -Wl,-N -Wl,-Ttext -Wl,7C00 conftest.o -o conftest.exec.7C00
/usr/bin/gld: warning: cannot find entry symbol _start; defaulting to
7c00
+ objcopy -O binary conftest.exec.7C00 conftest.7C00
+ md5sum conftest.2000 conftest.7C00 conftest.8000
42a9a3664e7dc9327f62aa111493b7b0  conftest.2000
42a9a3664e7dc9327f62aa111493b7b0  conftest.7C00
42a9a3664e7dc9327f62aa111493b7b0  conftest.8000
+ exit 0




Could you check this out (run the script)?
I'm thinking it could be cause by one of Debian patches.


Here the test script (from GRUB's configure):
 8 ==
#!/bin/sh

set -u
set -e
set -x

CC=${CC-gcc -m32}
OBJCOPY=${OBJCOPY-objcopy}

cat  conftest.c \EOF
void
cmain (void)
{
   *((int *) 0x1000) = 2;
}
EOF

$CC -c conftest.c -o conftest.o

for link_addr in 2000 8000 7C00; do
$CC -nostdlib -Wl,-N -Wl,-Ttext -Wl,$link_addr conftest.o -o
conftest.exec.$link_addr
$OBJCOPY -O binary conftest.exec.$link_addr conftest.$link_addr
done

md5sum conftest.[0-9]*

exit 0
 8 ==


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8w0Nyx9XDK=NW0oo+buRkSmt+cm=p7rk0y52ppmurv...@mail.gmail.com

Re: objcopy vs absolute addresses

[Игорь Пашев]

2013/3/30 Игорь Пашев pashev.i...@gmail.com:
 Hi, all.

 While trying to build GRUB on illumos [*-pc-solaris2.11], I've found that
 for different absolute addresses (-Wl,-Ttext -Wl,ADDR) objcopy -O binary
 produces different objects files, but it is expected to be the same.

Building grub-legacy on Debian/amd64 fails too:

checking for objcopy... objcopy
checking if C symbols get an underscore after compilation... no
checking whether objcopy works for absolute addresses... no
configure: error: GRUB requires a working absolute objcopy; upgrade
your binutils


--
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8zvzsf76-2vmsymwmyjsz1kmnwakbfwes_q4n4acx1...@mail.gmail.com

Re: objcopy vs absolute addresses

[Игорь Пашев]

Actually, grub-legacy is buildable... Probably, some patches make it possible.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8zgtnsbp1rfsj_bkklwkkm2msjncvhi3vxn+ecff9-...@mail.gmail.com

Re: objcopy vs absolute addresses

[Игорь Пашев]

2013/3/30 Jeff Epler jep...@unpythonic.net:
 This is something to do with the .note.gnu.build-id.

 Both of these alternatives give three identical object files:

 Removing this section in objcopy:
 $OBJCOPY -R .note.gnu.build-id ...

 Or requesting that the linker not generate it in the first place
 $CC -nostdlib -Wl,--build-id=none ...



Holy cow! Thank you!


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8w6kjqklejd29pjtc8w_vb4+kroomzn+vhyv9_szu-...@mail.gmail.com

Debian on illumos (Dyson) LiveCD

[Игорь Пашев]

Hi, all!

It was a long and hard way, but here it is. The new LiveCD.
http://osdyson.org/news/20

This is the last stage before creating installable port of Debian on illumos.


What is Dyson - http://osdyson.org/projects/dyson/wiki

It uses 64-bit (amd64) illumos kernel, libc (not glibc yet), and SMF
init system (not sysv, systemd, upstart or openrc ;-).
The primary filesystem is ZFS (not on LiveCD, of course), main network
configuration tools are dladm and ipadm.
In all other aspects it is Debian in nature (multiarch, GNU userland, etc.).


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8zhtEURg=4a2nF1D-rvMyMqU=4Sgg4Y-sCjSmabtxk=a...@mail.gmail.com

grub2 on zfs root

[Игорь Пашев]

Suddenly, I've found that grub2 knows about kopensolaris [1]

Could anybody say how it happened?
Any experience on using Grub2 with ZFS root?
And even more: on OpenSolaris?

[1] http://packages.debian.org/source/sid/grub2


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8x4X=ug3eotw5cmxizheqtzydwbs0tvgmo0zffjoux...@mail.gmail.com

Re: Creating symlinks to manpages

[Игорь Пашев]

2013/2/12 Colin Watson cjwat...@debian.org:
 There is, as far as I can see, no code in dh_installman that forces the
 section to upper-case.


I guess this is due to .TH


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8xndjw-m3sekemaxzdrnd1jcco16jjt1yi2x0fstmo...@mail.gmail.com

Re: Creating symlinks to manpages

[Игорь Пашев]

2013/2/12 Colin Watson cjwat...@debian.org:
 IMO it's rational to patch manual pages to lower-case the section in
 such cases, and forward that patch upstream so you don't need to
 continue maintaining it.  You'd have to do the same thing if they
 specified an entirely wrong section number, which does happen sometimes.

I like this idea.

But not in this life :-D [1]


[1] https://github.com/illumos/illumos-gate/tree/master/usr/src/man


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8zow6wg+hk-co3ekldqwh5phw+a6-b2lm7ves-srle...@mail.gmail.com

Creating symlinks to manpages

[Игорь Пашев]

Hi all.

Some manpages describe several utilities in one page,
so I have to create symlink for manpages, e. g. dcopy.1.gz - clri.1.gz
(while clri.1.gz is installed with dh_installman).

The thing I concern it that I need to know exact file name for link target.
If I write foo.1m in d/XXX.manpages, it results in foo.1M.gz (with
capital M) in a package.

Is there any handy way to link manpages? That is I could just write
`bar` is described in `foo`. I know about .so, but looks urgly because
debhelper finally replaces it with symlink :-)


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8xWmUNXxb+jjt+FvTSNQy=wR-YNr=24epcddbckxub...@mail.gmail.com

Re: Creating symlinks to manpages

[Игорь Пашев]

2013/2/11 Kartik Mistry kartik.mis...@gmail.com:
 On Mon, Feb 11, 2013 at 4:48 PM, Игорь Пашев pashev.i...@gmail.com wrote:
 Is there any handy way to link manpages? That is I could just write
 `bar` is described in `foo`. I know about .so, but looks urgly because
 debhelper finally replaces it with symlink :-)

 dh_link will do.

 For example: dh_link usr/share/man/man1/ayttm.1.gz
 usr/share/man/man1/ayttm_streamer_wrapper.1.gz

Yeah, that's what I do, but my problem is that I have to know final file name,
e. i. take care of section and compression.


--
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8zL5ueO3EbN04YH_yMG5wo9gmeXLA3c0HP=jzOLD7=k...@mail.gmail.com

Re: Creating symlinks to manpages

[Игорь Пашев]

2013/2/11 Jakub Wilk jw...@debian.org:
 The thing I concern it that I need to know exact file name for link
 target.


 Why is that a problem?


Because final file name is out of my control :-)
It is controlled by dh_installman and dh_compress

I have foo.1 in d/xxx.manpages, and only want to have
bar.1 - foo.1, no matter where foo.1 get its location
or how it is compressed.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8yembw_fe6z540ch_na+-mhjbdsjwtnv6tt_0d5ofu...@mail.gmail.com

Re: Creating symlinks to manpages

[Игорь Пашев]

Does it make sense?

1. Extend d/xxx.manpages to form:

bar.1 foo.1

2. dh_installman will create symlink bar.1 - foo.1

3. dh_compress will take care of symlinks, e. i.:

a) compress foo.1 to foo.1.gz (or whatever)
b) rm bar.1
c) create symlink bar.1.gz - foo.1.gz

I remember such a thing in Slackware :-)


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8w7bcjpgL9TQRrc1oGpcsZ2E3mQSTjvPMwweB==KU=c...@mail.gmail.com

Re: Creating symlinks to manpages

[Игорь Пашев]

And http://www.debian.org/doc/debian-policy/ch-docs.html


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8x707OpshOhd6LzAf3ow_NT5dxz-UmK0YN_CiwR=j1...@mail.gmail.com

Re: Creating symlinks to manpages

[Игорь Пашев]

2013/2/11 Игорь Пашев pashev.i...@gmail.com:


 a) compress foo.1 to foo.1.gz (or whatever)
 b) rm bar.1
 c) create symlink bar.1.gz - foo.1.gz


hm.. dh_compress does it already.


--
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8zp9hmtbbyac60_kog6kg8aukz+fsn5ostqwnmo_vt...@mail.gmail.com

Re: Creating symlinks to manpages

[Игорь Пашев]

2013/2/11 Arno Töll a...@debian.org:
 Hi,

 On 11.02.2013 13:46, Игорь Пашев wrote:
 I have foo.1 in d/xxx.manpages, and only want to have
 bar.1 - foo.1, no matter where foo.1 get its location
 or how it is compressed.

 You should not use symlinks, but source pages such tasks [1].

 Install a manpage like:

 bar.1:
 .so man1/foo.1

If you look at dh_installman, you will see that it replaces such dummy pages
with symlinks.


--
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8wz5m8hs+ymcrmpszakv7kjhghaqqbxmqr2smso1gi...@mail.gmail.com

Re: socket-based activation has unmaintainable security?

[Игорь Пашев]

2013/2/6 Chow Loong Jin hyper...@debian.org:
 On 07/02/2013 01:35, Thomas Goirand wrote:
 
 Which would be the wrong way of doing things / wrong reason
 for using root as running user, since you can set the
 CAP_NET_BIND_SERVICE capability... (man capabilities ...)

 Yeah, I figured as much, but isn't that a Linuxism?

illumos (OpenSolaris) kernel has privileges [1]
So some services (can't remember which) are not started as root at all.

[1] http://illumos.org/man/5/privileges


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8y0ui+tv7fzsyjyuo1vsyydd2xaehvmhywu4oy6tov...@mail.gmail.com

Should llvm source package include *patched* sources?

[Игорь Пашев]

The package llvm-3.0-source includes tarball /usr/src/llvm-3.0/llvm-3.0.tar.xz.

This tarball seems to include original llvm sources without Debian patches.

clang-3.0 uses that tarball to build itself, obviously without Debian patches.

Should llvm source package include *patched* sources?
Or, at least, include debian-dir just like gcc-X.Y-source does?


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8wpqjjvtfgqjao5c9xbb7ng35xg1s6ld8nm-4perdq...@mail.gmail.com

Re: Go (golang) packaging, part 2

[Игорь Пашев]

2013/1/30 Marc Haber mh+debian-de...@zugschlus.de:
 On Wed, 30 Jan 2013 16:25:12 +, Steve McIntyre st...@einval.com
 wrote:
To be fair, it's similar in reverse. Some Debian developers prefer to
_not_ install Ruby *at all*. Given how utterly awful the internals of
the language implementation are, I'd happily support dropping Ruby
from Debian altogether. Maybe that's just me...

 Ruby is actually a pretty nice language, and it is needed by the two
 major Configuration Management tools that are both widely used in
 business, puppet and chef.

Don't forget vim-addon-manager!


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8z3oV07uzY8TO9fwspykacihOTmspB79V3yV0JOYQ=m...@mail.gmail.com

Re: screen says Bad tty if /dev/console is a symlink

[Игорь Пашев]

So, is it ok?

Index: screen/tty.sh
===
--- screen.orig/tty.sh  2013-01-27 02:16:57.916935245 +
+++ screen/tty.sh   2013-01-27 02:33:12.831241123 +
@@ -1506,11 +1506,21 @@
 char *tty;
 {
   struct stat st;
+  char * real;
+  int rc;

-  if (lstat(tty, st) || !S_ISCHR(st.st_mode) ||
- (st.st_nlink  1  strncmp(tty, /dev/, 5)))
+  real = realpath(tty, NULL);
+  if (!real)
 return -1;
-  return 0;
+
+  if (lstat(real, st) || !S_ISCHR(st.st_mode) ||
+(st.st_nlink  1  strncmp(real, /dev/, 5)))
+rc = -1;
+  else
+rc = 0;
+
+  free(real);
+  return rc;
 }


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8wfS8o+o3PjK=esYwi-7AN3pmN_=lsdxv3yfselb2i...@mail.gmail.com

Re: screen says Bad tty if /dev/console is a symlink

[Игорь Пашев]

 I guess you meant PATH_MAX here, in any case POSIX does not guarantee
 MAX variables to be defined, it would be better to use the POSIX.1-2008
 variant of realpath(3) that allocates when passed a NULL (by checking
 if it's available at configure time).

I thought my libc did not support it, but now I see even
canonicalize_file_name() o_O


 I don't understand why do you a realpath() here depending on the
 number of hard links.

Me too, but original code has (st.st_nlink  1  strncmp(tty, /dev/, 5)),
so if number of links == 1, then there is no pathname check.
No path check - No realpath mess.

 Otherwise if this function
 is called on something like ttyname(3)'s return value, then changing
 this function is the correct thing to do.

Yes, CheckTtyname() is used with the value returned by ttyname() :
= screen.cc ===
 978 #define SET_TTYNAME(fatal) do \
 979   { \
 980 if (!(attach_tty = ttyname(0))) \
 981   { \
 982 if (fatal) \
 983   Panic(0, Must be connected to a terminal.); \
 984 else \
 985   attach_tty = ; \
 986   } \
 987 else \
 988   { \
 989 if (stat(attach_tty, st)) \
 990   Panic(errno, Cannot access '%s', attach_tty); \
 991 if (CheckTtyname(attach_tty)) \
 992   Panic(0, Bad tty '%s', attach_tty); \
 993   } \
 994 if (strlen(attach_tty) = MAXPATHLEN) \
 995   Panic(0, TtyName too long - sorry.); \
 996   } while (0)


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8yk9PKwO5d=LusLVBEVp1z8=axnDMV4ic=hz_rncte...@mail.gmail.com

Re: screen says Bad tty if /dev/console is a symlink

[Игорь Пашев]

So, do you mean it is good to do realpath() and only then stat() etc.?

CheckTtyname (tty)
{
  realpath;
  check real path;
}

?


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8yg12W4wO0=yqg+qfpra5-yacpxtnqo8f-c2jaems9...@mail.gmail.com

screen says Bad tty if /dev/console is a symlink

[Игорь Пашев]

Before filling a bug, I need a review :-)

The patch fo screen.cc [1] introduces the function CheckTtyname():

 int
 CheckTtyname (tty)
 char *tty;
 {
   struct stat st;

   if (lstat(tty, st) || !S_ISCHR(st.st_mode) ||
  (st.st_nlink  1  strncmp(tty, /dev/, 5)))
 return -1;
   return 0;
 }

When I run screen on a real console, screen says Bad tty '/dev/console'
I guess this is due to:

1.  /dev/console is a symlink (don't ask ;-):
# ls -lh /dev/console
lrwxrwxrwx 1 root root 30 May 13  2012 /dev/console -
../devices/pseudo/cn@0:console

2. lstat() stats a symlink itself, and it is obviosly not a character device.

Does it make sense to replace lstat() with stat()?


[1] 
http://anonscm.debian.org/gitweb/?p=collab-maint/screen.git;a=blob;f=debian/patches/47screen-cc.patch;h=a8f58e6c0c61fab6250e952e2707af0ff98b8452;hb=HEAD


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8wdgX83y2dvT2PxPkb9uqt_4EnJFgGBYX0ip5=petl...@mail.gmail.com

Re: screen says Bad tty if /dev/console is a symlink

[Игорь Пашев]

Did I break anything?
Too paranoid?


Index: screen/tty.sh
===
--- screen.orig/tty.sh  2013-01-26 07:55:33.092171499 +
+++ screen/tty.sh   2013-01-26 08:56:34.254462991 +
@@ -1505,12 +1505,29 @@
 CheckTtyname (tty)
 char *tty;
 {
+  char real[MAX_PATH];
   struct stat st;
-
-  if (lstat(tty, st) || !S_ISCHR(st.st_mode) ||
- (st.st_nlink  1  strncmp(tty, /dev/, 5)))
-return -1;
-  return 0;
+  if (0 == lstat(tty, st)) {
+if (S_ISCHR(st.st_mode)) {
+  if ((1 == st.st_nlink) || (0 == strncmp(tty, /dev/, 5)))
+  return 0;
+} else if (S_ISLNK(st.st_mode)) {
+  if (0 == stat(tty, st)  S_ISCHR(st.st_mode)) {
+if (1 == st.st_nlink) {
+   return 0;
+} else if (realpath(tty, real)) {
+#if defined(__sun__)
+if (!strncmp(real, /devices/, 9))
+  return 0;
+#else
+if (!strncmp(real, /dev/, 5))
+  return 0;
+#endif
+}
+  }
+}
+  }
+  return -1;
 }


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8y8W8dPX45cXFEhLHrNVtNvuTv=v1ka2zjzyhlytfn...@mail.gmail.com

Re: screen says Bad tty if /dev/console is a symlink

[Игорь Пашев]

I'd rather drop checking for path at all.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8zSrj=fgggfqeva4k4iuqngbb6u12to8_fz1vp4uqo...@mail.gmail.com

Re: Time to merge back ubuntu improvements!

[Игорь Пашев]

What is the defference:

1. Insert a new stage between stable and testing

and

2. double the period of automatic migration from unstable to testing?

m? :-)


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8x8+H4jD_WTvgby7BqFdRJB46rJMgXiOjd=wdaa_1e...@mail.gmail.com

Re: Packaging upstream tarballs with mixed C and Python sources

[Игорь Пашев]

2012/12/22 John Paul Adrian Glaubitz glaub...@physik.fu-berlin.de:
 Hi,

 I am currently trying to figure out how to get the new UAE fork
 fs-uae [1] ready for Debian and I haven't yet found a satisfiable
 way to build the package.

 fs-uae itself is written in C, but it ships together with a launcher
 GUI and a server for multi-player games, both written in Python. While
 the C source code for the emulator itself is located at the top
 directory of the source tarball, both the launcher and the server are
 located in sub-directories.

 If I just create the usual debhelper 9 rules file, the package will be
 built from the C sources only, ignoring the Python code for launcher
 and server.

 Does anyone have an idea how to create the three binary packages for
 the emulator, launcher and server from one source package. I am asking
 since I am not such an expert on Python packaging yet and hope that
 someone here knows better :).

 I am CC'ing upstream since I have been discussing the issue with him
 as well. Also, there might be the possibility to just split the
 upstream tarball into emulator, launcher and server to simplify
 packaging.

 The upstream tarball can be found here [2], if anyone wants to have a
 look.

 Cheers,

 Adrian

 [1] http://fengestad.no/fs-uae/
 [2] http://fengestad.no/fs-uae/stable/2.0.1/fs-uae-2.0.1.tar.gz


I think override_dh_auto_{build,install,clean} in debian/rules could help.
For example:

override_dh_auto_build:
cd server  ... # build server
$(MAKE) ... # build in top dir


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8wmo7xsjm4ua5budnyovyentnk7ehlm8vxmvxw0xke...@mail.gmail.com

Re: Mplayer not running with user privileges

[Игорь Пашев]

2012/12/7 Andrei POPESCU andreimpope...@gmail.com
 This is list is for development of Debian, your question should be
 posted on debian-user instead (or one of the debian-user-language
 lists).


This makes me think that the right list have been choosen: :-)

  I have ported debian-squeeze on my target x86


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8wdxgynzzppuqsbd24wjgrs2b4jhubj2e8rfdavy_+...@mail.gmail.com

Re: [OT] config file formats

[Игорь Пашев]

Guys, it looks like you are looking for The Silver Bullet.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8ygasil9fzoyhxxaipa-qt527tf+s9su0i03+3qhek...@mail.gmail.com

Re: Canonical pushes upstart into user session - systemd developer complains

[Игорь Пашев]

2012/12/2 Vincent Lefevre vinc...@vinc17.net:
 No, that's not sufficient. You may want relations between key-value
 pair. For instance, if you have a line with a key foo, then a line
 with a key bar must also exist. Or a line with a key number must
 have a value that is a number (more generally matching some regexp).

For such configs general programming languages are good.

E. g. perl:

$foo = wtf;

if ($foo  !$bar) {
ohshit(...);
}

...


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8zGhKjGBgp=i9BX5KVdQGrUce4Y0UAZzYuo3pYWb=u...@mail.gmail.com

Debian package browser

[Игорь Пашев]

Hi, all.

For obvious reasons I cannot find [in the Net] a way how to setup a
package browser
similar to one used at http://packages.debian.org or at
http://packages.ubuntu.com

Could anybody point me in the right direction?

Once again: I'd like to have debian package browser on my site.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8zo+4fy8yq4cx8zxo5wt6bwuwdqxeqvhbx93hq_ecb...@mail.gmail.com

Re: Debian package browser

[Игорь Пашев]

 Hi!

 The source code for packages.debian.org is available in this git
 repository:
  http://anonscm.debian.org/gitweb/?p=webwml/packages.git;a=summary

Thanks!


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8ywBH_+8PoVCcSxF6_7DdOfLETptne2Uu=txrv-m4l...@mail.gmail.com

Re: Really, about udev, not init sytsems

[Игорь Пашев]

2012/11/29 Wouter Verhelst wou...@debian.org:
 glibc and the kernel is developed by the same group of companies. Both
 interact very closely and integration of the sources was the natural
 consequence.


Please, *DONT* :-)

I've tired of this crap on illumos


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CALL-Q8xMeDSUHaLw=dtkub3p4cy3k-afzpza16ztblfl7kb...@mail.gmail.com

Do not CC me

[Игорь Пашев]

Hi there!

I see many note in this list like:
I'm registered to the list. So please *do not* Cc: me.

So I'd like to note:

1. Some e-mail cleints make it hard not to CC. For example GMail has only
two options: reply and reply to all. Reply will send email to the author,
not to the list

2. Some email cleints are smart enough to guess that CC and list email is
the same and will not duplicate it

Re: A common configuration format, anyone?

[Игорь Пашев]

2012/11/14 Philip Ashmore cont...@philipashmore.com

 simple format which, like xml, is human-readable


XML is not human-readable :-)

Re: History of Debian bootstrapping/porting efforts

[Игорь Пашев]

I took OpenIndiana and started to rebuild Debian packages ignoring formal
dependencies.
Each new package was install via dpkg-deb -x :-)

Re: major linux problems summary 2012

[Игорь Пашев]

These are not *linux* problems.


2012/11/3 Andrey Rahmatullin w...@wrar.name

 On Sat, Nov 03, 2012 at 07:30:00PM +, Ben Hutchings wrote:
  Don't feed the troll.
 It's not trolling.
 It's a honest list of problems which a lot of developers don't view as
 problems.

 --
 WBR, wRAR

Re: Mandatory -dbg packages

[Игорь Пашев]

Do not kick me, but in Solaris, there are :

1) special ELF section .SUNW_ldynsym extending .dynsym [1]
2) .SUNW_ctf with CTF data [2]
3) Function args are saved in stack for amd64

But from my point all it exists because Solaris lacks robust
package manager :-)

[1] https://blogs.oracle.com/ali/entry/what_is_sunw_ldynsym
[2] http://hub.opensolaris.org/bin/view/Project+ppc-dev/ctf

2012/10/27 Vincent Bernat ber...@debian.org

  ❦ 27 octobre 2012 12:42 CEST, Andrey Rahmatullin w...@wrar.name :

  Libraries with `-dbg` package are a pain to deal with when debugging
  some problem.
  You probably mean without.

 Yes.
 --
 printk(KERN_WARNING Multi-volume CD somehow got mounted.\n);
 2.2.16 /usr/src/linux/fs/isofs/inode.c

Re: [SUMMARY/PROPOSAL] Orphaning another maintainer's packages

[Игорь Пашев]

2012/10/25 Steve Langasek vor...@debian.org

  It may just mean you've managed to send
 your request to the wrong place


As I see, almost all debian guys are so courteous that they point to the
right place.

Re: (seemingly) declinging bug report numbers

[Игорь Пашев]

I will fill your bugreports for $1.99 per bug :-)

2012/10/19 Peter Samuelson pe...@p12n.org:

 [Kelly Clowers]
 But I basically never report bugs. I have used Sid for years, and in
 fact I often don't notice bugs in my personal workflow (maybe if I
 can think of myself as a user? I notice end-user-impacting bugs in
 other areas). If someone comes over and sees me working the might
 say, wow that is an annoying bug and I say what bug? Oh that.  I
 didn't notice, I just worked around it. Even with bugs I do notice,
 I usually just ignore and work around until it is fixed.

 Don't feel bad about that.  Reporting a bug is a _burden_, especially
 if you care enough to produce a high-quality report.  Even if the
 actual reporting part is pretty easy, you have to gather a lot of
 information: is it reproduceable and if so, how?  How sure am I that it
 isn't user error or local configuration?  How sure am I that it hasn't
 already been fixed by a newer upload?  Is there anything strange in my
 environment that I am forgetting to mention, that would make the bug
 hard for anyone else to reproduce?  And of course that's not even
 counting the time investment of working with the maintainer after the
 initial report.

 I don't fault anyone for deciding that the return on investment for
 producing a high-quality bug report is higher than for just working
 around it.  I often do the same.  We of course appreciate when users
 are willing to contribute a good bug report, but we don't require or
 expect everybody to do it.  Mostly we produce Debian so you can _use_
 it, not so you can spend your time helping us make it better.

 Peter


 --
 To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
 with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
 Archive: http://lists.debian.org/20121019164112.gf4...@p12n.org



-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/call-q8wn+y3kadso5dykw7yw-rjw6entrusk1+aaax0rz5j...@mail.gmail.com