Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.debian.kernel > #50314 > unrolled thread

Bug#793786: initramfs-tools: fix the broken netconsole feature in initramfs-tools

Back to article view | Back to linux.debian.kernel

Contents

  Bug#793786: initramfs-tools: fix the broken netconsole feature in initramfs-tools Roger Shimizu <rogershimizu@gmail.com> - 2015-07-27 16:50 +0200
    Bug#793786: initramfs-tools: fix the broken netconsole feature in initramfs-tools Roger Shimizu <rogershimizu@gmail.com> - 2015-07-29 12:00 +0200
      Bug#793786: initramfs-tools: fix the broken netconsole feature in initramfs-tools Roger Shimizu <rogershimizu@gmail.com> - 2015-07-31 19:10 +0200
      Bug#793786: initramfs-tools: fix the broken netconsole feature in initramfs-tools Roger Shimizu <rogershimizu@gmail.com> - 2015-08-02 12:20 +0200
        Bug#793786: initramfs-tools: fix the broken netconsole feature in initramfs-tools Ben Hutchings <ben@decadent.org.uk> - 2015-12-09 04:50 +0100
          Bug#793786: initramfs-tools: fix the broken netconsole feature in initramfs-tools Roger Shimizu <rogershimizu@gmail.com> - 2016-01-10 14:20 +0100
          Bug#793786: initramfs-tools: fix the broken netconsole feature in initramfs-tools Ben Hutchings <ben@decadent.org.uk> - 2016-01-11 00:20 +0100
            Bug#793786: initramfs-tools: fix the broken netconsole feature in initramfs-tools Roger Shimizu <rogershimizu@gmail.com> - 2016-01-11 02:00 +0100
              Bug#793786: initramfs-tools: fix the broken netconsole feature in initramfs-tools Ben Hutchings <ben@decadent.org.uk> - 2016-01-11 02:30 +0100
    Bug#793786: marked as done (initramfs-tools: fix the broken  netconsole feature in initramfs-tools) owner@bugs.debian.org (Debian Bug Tracking System) - 2015-12-28 03:50 +0100
    Processed: Re: Bug#793786: initramfs-tools: fix the broken  netconsole feature in initramfs-tools owner@bugs.debian.org (Debian Bug Tracking System) - 2016-01-10 08:30 +0100
    Bug#793786: marked as done (initramfs-tools: fix the broken  netconsole feature in initramfs-tools) owner@bugs.debian.org (Debian Bug Tracking System) - 2016-01-12 12:30 +0100

#50314 — Bug#793786: initramfs-tools: fix the broken netconsole feature in initramfs-tools

[Multipart message — attachments visible in raw view] — view raw

Package: initramfs-tools
Version: 0.120
Severity: normal
Tags: patch

Dear Maintainer,

Currently there're a few issues on netconsole support in initramfs-tools:

- netconsole setting (param to kernel module) can only be set in 
/etc/initramfs-tools/modules, but not from the expected kernel command line
- debug info of initramfs-tools can be outputed to local console or saved 
to local file, which is inconvient for embeded system when the boot fails 
and there's neither console or serial console

So here's the two patches to fix them.
Please help to merge. Thank you!

Cheers,
Roger

-- Package-specific info:
-- /proc/cmdline
console=ttyS0,115200 earlyprintk netconsole=@192.168.11.150/,6666@192.168.11.1/ root=/dev/md2 initrd=0x00800040,12M panic=5 BOOTVER=0.21 mtdparts=nand_mtd:0x1000000(boot),0xfe800000(rootfs),0x800000(reserve) debug

-- System Information:
Debian Release: 8.1
  APT prefers stable
  APT policy: (990, 'stable'), (500, 'stable-updates'), (500, 'unstable'), (1, 'experimental')
Architecture: armel (armv5tel)

Kernel: Linux 4.0.0-2-kirkwood
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: sysvinit (via /sbin/init)

Versions of packages initramfs-tools depends on:
ii  cpio         2.11+dfsg-4.1
ii  klibc-utils  2.0.4-2
ii  kmod         18-3
ii  udev         215-17+deb8u1

Versions of packages initramfs-tools recommends:
pn  busybox | busybox-initramfs | busybox-static  <none>

Versions of packages initramfs-tools suggests:
ii  bash-completion  1:2.1-4

-- no debconf information

[toc] | [next] | [standalone]

#50332

Dear Ian,

Thanks for your response!

I'd like to ask whether you could offer the "Reviewed-by" for my first
patch of Bug#793786?
(0001-advance-the-timing-of-insmod-netconsole.patch)
You must be a netconsole user that understand the reasoning why to
load netconsole module with param should be earlier than calling
"load_modules" routine.
Look forward to your feedback!

Cheers,
Roger

On Wed, Jul 29, 2015 at 4:17 AM, Ian Campbell <ijc@debian.org> wrote:
> On Wed, 2015-07-29 at 01:16 +0900, Roger Shimizu wrote:
>> Hope you don't mind that I polished your code and post in BTS.
>
> Absolutely not, thanks for doing so!
>
> In case it is necessary any code from http://www.hellion.org.uk/blog/po
> sts/debugging-initramfs-over-netconsole/
> should be considered:
>
> Signed-off-by: Ian Campbell <ijc@debian.org>
>
> for the purposes of inclusion in initramfs-tools.
>
> Ian.


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

[toc] | [prev] | [next] | [standalone]

#50344

[Multipart message — attachments visible in raw view] — view raw

Dear Maintainer,

Per Ian's confirmation, enclosed the updated patchset.
1st patch is not touched, only some commit log change in 2nd patch.
Thank you!

Cheers,
Roger

[toc] | [prev] | [next] | [standalone]

#50353

[Multipart message — attachments visible in raw view] — view raw

Dear Ian,

Thanks for your feedback!

After reading your comment, I reviewed the whole procedure of loading
netconsole module during initramfs-tools booting, previously and after
my patch, I start to consider that original "modprobe netconsole
netconsole=$param" line in init file is merely a dirty hack. The real
fix is to load the netconsole module with param inside load_modules()
routine.

So I created v3 patchset, which looks more clean to me.
Of course patchset v2 & v3 both works well as I tested. So I let the
package maintainer to choose which is better to merge.


Dear Maintainer,

Enclosed is the updated patchset v3.

Changelog:
v1 => v2: for 0002 patch: commit log change
v2 => v3: for 0001 patch: change loading netconsole module timing from
separately in init script to load_modules() routine
v2 pros: limit the affected range only when netconsole is set; cons:
netconsole module would be load twice, so seems a dirty hack
v3 pros: clean code; cons: add a condition judging on each module
loading (overhead?)

Both v2 and v3 are fine to me.
Please simply pick one when to merge. Thank you!

Cheers,
Roger

[toc] | [prev] | [next] | [standalone]

#51605

[Multipart message — attachments visible in raw view] — view raw

On Sun, 2 Aug 2015 19:08:30 +0900 Roger Shimizu <rogershimizu@gmail.com> wrote:
> Dear Ian,
> 
> Thanks for your feedback!
> 
> After reading your comment, I reviewed the whole procedure of loading
> netconsole module during initramfs-tools booting, previously and after
> my patch, I start to consider that original "modprobe netconsole
> netconsole=$param" line in init file is merely a dirty hack. The real
> fix is to load the netconsole module with param inside load_modules()
> routine.
> 
> So I created v3 patchset, which looks more clean to me.
> Of course patchset v2 & v3 both works well as I tested. So I let the
> package maintainer to choose which is better to merge.
[...]

An initramfs built with MODULES=most (default) always includes
netconsole.ko (assuming it was enabled in the kernel configuration).
So
the netconsole parameter needs to work even if netconsole doesn't
appear in /conf/modules, and your v3 breaks that.  So I applied your
v2.  Thanks.

Ben.

-- 
Ben Hutchings
I'm always amazed by the number of people who take up solipsism because
they heard someone else explain it. - E*Borg on alt.fan.pratchett

[toc] | [prev] | [next] | [standalone]

#52126

[Multipart message — attachments visible in raw view] — view raw

> CaseA (MODULES=list, netconsole param from bootloader, tested on an
> armel/orion5x device):
> - /etc/initramfs-tools/initramfs.conf:
> MODULES=list
>
> - /etc/initramfs-tools/modules:
> mv643xx_eth
> netconsole
> mvmdio
> sata_mv
> sd_mod
> raid1
> jfs
>
> CaseB (MODULES=dep, netconsole param from bootloader, tested on an
> armel/orion5x device):
> - /etc/initramfs-tools/initramfs.conf:
> MODULES=dep
>
> - /etc/initramfs-tools/modules:
> mv643xx_eth
> netconsole
> mvmdio
>
> CaseC (MODULES=most, netconsole param from bootloader, tested on x86-64 device):
> -/etc/initramfs-tools/initramfs.conf:
> MODULES=most
>
> - /etc/initramfs-tools/modules:
> # empty

all previous cases were using netconsole param from bootloader, so I
created Case A2 to use netconsole param from conf/modules.

CaseA2 (MODULES=list, netconsole param from conf/modules, tested on an
armel/kirkwood device):
- /etc/initramfs-tools/initramfs.conf:
MODULES=list

- /etc/initramfs-tools/modules:
mv643xx_eth
netconsole
mvmdio
sata_mv
sd_mod
raid1
xfs

This time I ran case A/A2/B on an armel/kirkwood device Linkstation LS-WXL.
Enclosed the netconsole boot log.

Cheers,
Roger

[toc] | [prev] | [next] | [standalone]

#52136

[Multipart message — attachments visible in raw view] — view raw

On Sun, 2016-01-10 at 16:20 +0900, Roger Shimizu wrote:
> Control: reopen -1
> 
> On Wed, Dec 9, 2015 at 12:39 PM, Ben Hutchings <ben@decadent.org.uk> wrote:
> > An initramfs built with MODULES=most (default) always includes
> > netconsole.ko (assuming it was enabled in the kernel configuration).
> > So
> > the netconsole parameter needs to work even if netconsole doesn't
> > appear in /conf/modules, and your v3 breaks that.  So I applied your
> > v2.  Thanks.
> 
> I find current v2 patch will sometimes fail because it load netconsole.ko before
> loading network driver.

So maybe we should run 'udevadm settle' before loading modules?

> I made some changes based on patch v3, can confirmed various scenarios,
> including "MODULES=most" you mentioned. So here's patch v4.
> (also in my github repo's netconsole branch:
> https://github.com/rogers0/initramfs-tools/tree/netconsole)
> 
> the fix already takes various cases into account:
> - MODULES=most (default) or MODULES={dep,list} (popular for embedded devices)
> - netconsole param may be inside conf/modules or come with bootloader's config
> 
> for "MODULES=most" case , netconsole.ko is included by default, so there's
> no need to list it in conf/modules, and netconsole param should be put into
> bootloader's config
> 
> for "MODULES={dep,list}" case, netconsole param can be either from
> conf/modules or bootloader, and the previous one is preferred if both are
> specified.
[...]

The kernel parameter should be preferred if both are specified.  (In
general, the most specific configuration source should take
precedence.)

Ben.

-- 
Ben Hutchings
Power corrupts.  Absolute power is kind of neat.
                           - John Lehman, Secretary of the US Navy 1981-1987

[toc] | [prev] | [next] | [standalone]

#52139

On Mon, Jan 11, 2016 at 8:13 AM, Ben Hutchings <ben@decadent.org.uk> wrote:
> On Sun, 2016-01-10 at 16:20 +0900, Roger Shimizu wrote:
>> I find current v2 patch will sometimes fail because it load netconsole.ko before
>> loading network driver.
>
> So maybe we should run 'udevadm settle' before loading modules?

I don't understand this part.
If you have patch, I can run it in my netconsole testable environment,
then feedback the log.

>> I made some changes based on patch v3, can confirmed various scenarios,
>> including "MODULES=most" you mentioned. So here's patch v4.
>> (also in my github repo's netconsole branch:
>> https://github.com/rogers0/initramfs-tools/tree/netconsole)
>>
>> the fix already takes various cases into account:
>> - MODULES=most (default) or MODULES={dep,list} (popular for embedded devices)
>> - netconsole param may be inside conf/modules or come with bootloader's config
>>
>> for "MODULES=most" case , netconsole.ko is included by default, so there's
>> no need to list it in conf/modules, and netconsole param should be put into
>> bootloader's config
>>
>> for "MODULES={dep,list}" case, netconsole param can be either from
>> conf/modules or bootloader, and the previous one is preferred if both are
>> specified.
> [...]
>
> The kernel parameter should be preferred if both are specified.  (In
> general, the most specific configuration source should take
> precedence.)

Yes, I agree.
I meant if both are *same* specific level, param from conf/modules
will be preferred.
It's what's the test case A2 for.

Cheers,
Roger

[toc] | [prev] | [next] | [standalone]

#52140

[Multipart message — attachments visible in raw view] — view raw

On Mon, 2016-01-11 at 09:54 +0900, Roger Shimizu wrote:
> On Mon, Jan 11, 2016 at 8:13 AM, Ben Hutchings <ben@decadent.org.uk> wrote:
> > On Sun, 2016-01-10 at 16:20 +0900, Roger Shimizu wrote:
> > > I find current v2 patch will sometimes fail because it load netconsole.ko before
> > > loading network driver.
> > 
> > So maybe we should run 'udevadm settle' before loading modules?
> 
> I don't understand this part.
> If you have patch, I can run it in my netconsole testable environment,
> then feedback the log.

--- a/init
+++ b/init
@@ -207,6 +207,7 @@ run_scripts /scripts/init-top
 
 maybe_break modules
 [ "$quiet" != "y" ] && log_begin_msg "Loading essential drivers"
+udevadm settle  # make sure network drivers are loaded
 [ -n "${netconsole}" ] && modprobe netconsole netconsole="${netconsole}"
 load_modules
 [ "$quiet" != "y" ] && log_end_msg
--- END ---

It won't be reliable for USB networking, but I think that's inevitable.

> > > I made some changes based on patch v3, can confirmed various scenarios,
> > > including "MODULES=most" you mentioned. So here's patch v4.
> > > (also in my github repo's netconsole branch:
> > > https://github.com/rogers0/initramfs-tools/tree/netconsole)
> > > 
> > > the fix already takes various cases into account:
> > > - MODULES=most (default) or MODULES={dep,list} (popular for embedded devices)
> > > - netconsole param may be inside conf/modules or come with bootloader's config
> > > 
> > > for "MODULES=most" case , netconsole.ko is included by default, so there's
> > > no need to list it in conf/modules, and netconsole param should be put into
> > > bootloader's config
> > > 
> > > for "MODULES={dep,list}" case, netconsole param can be either from
> > > conf/modules or bootloader, and the previous one is preferred if both are
> > > specified.
> > [...]
> > 
> > The kernel parameter should be preferred if both are specified.  (In
> > general, the most specific configuration source should take
> > precedence.)
> 
> Yes, I agree.
> I meant if both are *same* specific level, param from conf/modules
> will be preferred.
> It's what's the test case A2 for.

The kernel paramater is more specific, because that is something the
user can (usually) edit at boot time.

Ben.

-- 
Ben Hutchings
Q.  Which is the greater problem in the world today, ignorance or apathy?
A.  I don't know and I couldn't care less.

[toc] | [prev] | [next] | [standalone]

#51910 — Bug#793786: marked as done (initramfs-tools: fix the broken netconsole feature in initramfs-tools)

[Multipart message — attachments visible in raw view] — view raw

Your message dated Sun, 27 Dec 2015 23:00:16 +0000
with message-id <E1aDKIS-0001mb-EK@franck.debian.org>
and subject line Bug#793786: fixed in initramfs-tools 0.121~rc2
has caused the Debian Bug report #793786,
regarding initramfs-tools: fix the broken netconsole feature in initramfs-tools
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)


-- 
793786: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=793786
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems

[toc] | [prev] | [next] | [standalone]

#52124 — Processed: Re: Bug#793786: initramfs-tools: fix the broken netconsole feature in initramfs-tools

Processing control commands:

> reopen -1
Bug #793786 {Done: Ben Hutchings <ben@decadent.org.uk>} [initramfs-tools] initramfs-tools: fix the broken netconsole feature in initramfs-tools
'reopen' may be inappropriate when a bug has been closed with a version;
all fixed versions will be cleared, and you may need to re-add them.
Bug reopened
No longer marked as fixed in versions initramfs-tools/0.121~rc2.

-- 
793786: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=793786
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems

[toc] | [prev] | [next] | [standalone]

#52158 — Bug#793786: marked as done (initramfs-tools: fix the broken netconsole feature in initramfs-tools)

[Multipart message — attachments visible in raw view] — view raw

Your message dated Tue, 12 Jan 2016 20:20:19 +0900
with message-id <CAEQ9gEneKu-o4Vtt2TnSrH+2yFLDcg=sw2OG7yiDd+LQwsGzPg@mail.gmail.com>
and subject line Re: Bug#793786: initramfs-tools: fix the broken netconsole feature in initramfs-tools
has caused the Debian Bug report #793786,
regarding initramfs-tools: fix the broken netconsole feature in initramfs-tools
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)


-- 
793786: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=793786
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems

[toc] | [prev] | [standalone]

Back to top | Article view | linux.debian.kernel

csiph-web