Re: [packages/ipmitool] do not suggest some crafted script when there are contemporary and generic module loaders

[Jacek Konieczny]

On 2016-07-29 13:10, Elan Ruusamäe wrote:

On 28.07.2016 23:39, Tomasz Pala wrote:

who knows what device would become available at /dev/ipmi0 at next
reboot... (unless cleaned in rc.sysinit). This package should be banned,


+1


I'm considering C: ipmi-init somewhere, but I'm not sure where to put
it...

you could as well fix the script to always replace /dev/ipmi0 if /dev is
static


There is no point in keeping every legacy stuff just because we had it 
in PLD at some time. Some pieces of software were just mistakes, that is 
it. Maintaining broken stuff 'for compatibility' is pointless, it still 
requires work bringing, but only brings problems.


If some system relies on old, broken behavior, then the system needs an 
update. Much bigger distibutions won't support indefinite backward 
compatibility. We cannot afford that even more.


It is the same case for 'Require: mingetty' in rc-scripts.

Jacek
___
pld-devel-en mailing list
pld-devel-en@lists.pld-linux.org
http://lists.pld-linux.org/mailman/listinfo/pld-devel-en

Re: [packages/ipmitool] do not suggest some crafted script when there are contemporary and generic module loaders

[Elan Ruusamäe]

On 28.07.2016 23:39, Tomasz Pala wrote:


And take a look at this ipmi.init - it creates static device node (in
non-devfs systems) with RANDOM (not permanently assigned in kernel) major!

if [ ! -c /dev/ipmi0 ] && [ $RETVAL -eq 0 ]; then
major=$(awk '/ ipmidev$/{print $1}' /proc/devices)
/bin/mknod -m 0600 /dev/ipmi0 c $major 0 || RETVAL=1

who knows what device would become available at /dev/ipmi0 at next
reboot... (unless cleaned in rc.sysinit). This package should be banned,

I'm considering C: ipmi-init somewhere, but I'm not sure where to put it...
you could as well fix the script to always replace /dev/ipmi0 if /dev is 
static


--
glen

___
pld-devel-en mailing list
pld-devel-en@lists.pld-linux.org
http://lists.pld-linux.org/mailman/listinfo/pld-devel-en

Re: [packages/ipmitool] do not suggest some crafted script when there are contemporary and generic module loaders

[Tomasz Pala]

On Thu, Jul 28, 2016 at 14:08:35 +0200, Arkadiusz Miśkiewicz wrote:

>> Local IPMI handling is *required* for watchdog only,
> 
> ipmievd is another local user.

Can be remote as well: -H. As it's more reliable to handle BMC from
separate machine than the monitored one. OTOH /dev/watchdog must be
handled from within the same system, and it is by PID=1 (systemd).


And take a look at this ipmi.init - it creates static device node (in
non-devfs systems) with RANDOM (not permanently assigned in kernel) major!

if [ ! -c /dev/ipmi0 ] && [ $RETVAL -eq 0 ]; then
major=$(awk '/ ipmidev$/{print $1}' /proc/devices)
/bin/mknod -m 0600 /dev/ipmi0 c $major 0 || RETVAL=1

who knows what device would become available at /dev/ipmi0 at next
reboot... (unless cleaned in rc.sysinit). This package should be banned,

I'm considering C: ipmi-init somewhere, but I'm not sure where to put it...

-- 
Tomasz Pala 
___
pld-devel-en mailing list
pld-devel-en@lists.pld-linux.org
http://lists.pld-linux.org/mailman/listinfo/pld-devel-en

Re: [packages/ipmitool] do not suggest some crafted script when there are contemporary and generic module loaders

[Arkadiusz Miśkiewicz]

On Thursday 28 of July 2016, Tomasz Pala wrote:

> Local IPMI handling is *required* for watchdog only,

ipmievd is another local user.

-- 
Arkadiusz Miśkiewicz, arekm / ( maven.pl | pld-linux.org )
___
pld-devel-en mailing list
pld-devel-en@lists.pld-linux.org
http://lists.pld-linux.org/mailman/listinfo/pld-devel-en

Re: [packages/ipmitool] do not suggest some crafted script when there are contemporary and generic module loaders

[Tomasz Pala]

On Thu, Jul 28, 2016 at 13:54:05 +0200, Tomasz Pala wrote:

>>> as you did not replaced existing "-init" solution with described one, 
> 
> One more thing: since apparently it's beyond your comprehension, that

And, please don't add S: apache to firefox. You know, local clients,
remote servers thing... Ever heard of that? Ever used IPMI over IP?
Local IPMI handling is *required* for watchdog only, so this ipmi-init
fuckup should have been properly implemented somewhere else. Wait a
minute... it was!

http://git.pld-linux.org/gitweb.cgi?p=packages/watchdog.git;a=blob;f=watchdog.init

-- 
Tomasz Pala 
___
pld-devel-en mailing list
pld-devel-en@lists.pld-linux.org
http://lists.pld-linux.org/mailman/listinfo/pld-devel-en

Re: [packages/ipmitool] do not suggest some crafted script when there are contemporary and generic module loaders

[Tomasz Pala]

On Thu, Jul 28, 2016 at 13:46:23 +0200, Tomasz Pala wrote:

>> as you did not replaced existing "-init" solution with described one, 

One more thing: since apparently it's beyond your comprehension, that
Suggests matters only during install time, thus removing it doesn't change
anything in already working machines, so it doesn't require any special
attention (except maybe for Suggest: 
do_NOT_install-ipmi-init_do_it_the_proper_way_instead),
and this is your another bullshitting recently ("[packages/openvpn]
more bolder xz url note"), please do not bother replying me. *PLONK*
[google that, pl specific]

-- 
Tomasz Pala 
___
pld-devel-en mailing list
pld-devel-en@lists.pld-linux.org
http://lists.pld-linux.org/mailman/listinfo/pld-devel-en

Re: [packages/ipmitool] do not suggest some crafted script when there are contemporary and generic module loaders

[Tomasz Pala]

On Thu, Jul 28, 2016 at 13:08:56 +0300, Elan Ruusamäe wrote:

> as you did not replaced existing "-init" solution with described one, 

This was not a solution, but some nasty unnecessary scripting,
duplicating existing function of loading specified modules at system
start.

> you should not just go and remove the suggest.

This MUST NOT be suggested to anyone (sane).

> revert the change or finish the transition!

"Finishing" it equals Suggests: udev. Or removing the package entirely.
I don't support the former, and won't do the last in case someone that
was missuggested before still uses it.

If you can't understand, that is is totally obsoleted now, requires no
replacement and was bogus since the day it was born, I think you should
go get laid and stop posting this bullshit. EOT.

-- 
Tomasz Pala 
___
pld-devel-en mailing list
pld-devel-en@lists.pld-linux.org
http://lists.pld-linux.org/mailman/listinfo/pld-devel-en

Re: [packages/ipmitool] do not suggest some crafted script when there are contemporary and generic module loaders

[Elan Ruusamäe]

On 28.07.2016 00:50, Tomasz Pala wrote:

On Wed, Jul 27, 2016 at 23:31:36 +0200, Tomasz Pala wrote:


>>it's soft dep. soft deps are to inform user of suggested packages

>
>And this one should not be suggested. This package should be entirely
>removed.

Similar cleanup was done years ago in lm_sensors:

http://git.pld-linux.org/gitweb.cgi?p=packages/lm_sensors.git;a=commitdiff;h=ed9691db2aa168387fb4ff788d934b5b3127e9e7

and there is no reason for creating init scripts loading every single
module out there (joystick-init, alsa-init, floppy-init, usb-init?!) If
you want modules loaded, that ARE NOT handled by udev/systemd, just put
them in /etc/modules.d or provide /etc/modules-load.d/ipmi.conf.
Suggesting script which tries to mknod something is a NO GO. Not in XXI
century.
as you did not replaced existing "-init" solution with described one, 
you should not just go and remove the suggest.


revert the change or finish the transition!

--
glen

___
pld-devel-en mailing list
pld-devel-en@lists.pld-linux.org
http://lists.pld-linux.org/mailman/listinfo/pld-devel-en

Re: [packages/ipmitool] do not suggest some crafted script when there are contemporary and generic module loaders

[Tomasz Pala]

On Wed, Jul 27, 2016 at 23:31:36 +0200, Tomasz Pala wrote:

>> it's soft dep. soft deps are to inform user of suggested packages
> 
> And this one should not be suggested. This package should be entirely
> removed.

Similar cleanup was done years ago in lm_sensors:

http://git.pld-linux.org/gitweb.cgi?p=packages/lm_sensors.git;a=commitdiff;h=ed9691db2aa168387fb4ff788d934b5b3127e9e7

and there is no reason for creating init scripts loading every single
module out there (joystick-init, alsa-init, floppy-init, usb-init?!) If
you want modules loaded, that ARE NOT handled by udev/systemd, just put
them in /etc/modules.d or provide /etc/modules-load.d/ipmi.conf.
Suggesting script which tries to mknod something is a NO GO. Not in XXI
century.

-- 
Tomasz Pala 
___
pld-devel-en mailing list
pld-devel-en@lists.pld-linux.org
http://lists.pld-linux.org/mailman/listinfo/pld-devel-en

Re: [packages/ipmitool] do not suggest some crafted script when there are contemporary and generic module loaders

[Tomasz Pala]

On Thu, Jul 28, 2016 at 00:26:13 +0300, Elan Ruusamäe wrote:

> it's soft dep. soft deps are to inform user of suggested packages

And this one should not be suggested. This package should be entirely
removed.

> revert your removal.

Forget it.

-- 
Tomasz Pala 
___
pld-devel-en mailing list
pld-devel-en@lists.pld-linux.org
http://lists.pld-linux.org/mailman/listinfo/pld-devel-en

Re: [packages/ipmitool] do not suggest some crafted script when there are contemporary and generic module loaders

[Elan Ruusamäe]

it's soft dep. soft deps are to inform user of suggested packages

revert your removal.

On 27.07.2016 22:06, gotar wrote:

commit 30c556b20dde02d4520b85755006804f1dcb1aec
Author: Tomasz Pala 
Date:   Wed Jul 27 21:06:04 2016 +0200

 do not suggest some crafted script when there are contemporary and generic 
module loaders

  ipmitool.spec | 1 -
  1 file changed, 1 deletion(-)
---
diff --git a/ipmitool.spec b/ipmitool.spec
index b3a74c4..61189ed 100644
--- a/ipmitool.spec
+++ b/ipmitool.spec
@@ -15,7 +15,6 @@ BuildRequires:automake
  BuildRequires:libltdl-devel
  BuildRequires:libtool
  BuildRequires:rpmbuild(macros) >= 1.268
-Suggests:  ipmi-init
  Obsoletes:ipmitool-devel
  BuildRoot:%{tmpdir}/%{name}-%{version}-root-%(id -u -n)
  



 gitweb:

http://git.pld-linux.org/gitweb.cgi/packages/ipmitool.git/commitdiff/30c556b20dde02d4520b85755006804f1dcb1aec

___
pld-cvs-commit mailing list
pld-cvs-com...@lists.pld-linux.org
http://lists.pld-linux.org/mailman/listinfo/pld-cvs-commit



--
glen

___
pld-devel-en mailing list
pld-devel-en@lists.pld-linux.org
http://lists.pld-linux.org/mailman/listinfo/pld-devel-en