[gentoo-user] hald won't start

[gentoo-user] hald won't start

[Mick]

[-- Attachment #1: Type: text/plain, Size: 341 bytes --]

Hi All,

I noticed on a box of mine that hald won't start:
==========================================
# /etc/init.d/hald restart
 * Starting Hardware Abstraction Layer daemon ...          [ !! ]
==========================================

Confusingly, the logs don't show anything.  How do I troubleshoot this?
-- 
Regards,
Mick

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

[gentoo-user] Re: hald won't start

[Mick]

[-- Attachment #1: Type: text/plain, Size: 527 bytes --]

On Saturday 14 July 2007 20:11, Mick wrote:
> Hi All,
>
> I noticed on a box of mine that hald won't start:
> ==========================================
> # /etc/init.d/hald restart
>  * Starting Hardware Abstraction Layer daemon ...          [ !! ]
> ==========================================
>
> Confusingly, the logs don't show anything.  How do I troubleshoot this?

I just noticed that I do not have CONFIG_INOTIFY_USER enabled in the kernel.  
Perhaps this has something to do with it?
-- 
Regards,
Mick

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-user] Re: hald won't start

[Dale]

[-- Attachment #1: Type: text/plain, Size: 895 bytes --]

Mick wrote:
> On Saturday 14 July 2007 20:11, Mick wrote:
>   
>> Hi All,
>>
>> I noticed on a box of mine that hald won't start:
>> ==========================================
>> # /etc/init.d/hald restart
>>  * Starting Hardware Abstraction Layer daemon ...          [ !! ]
>> ==========================================
>>
>> Confusingly, the logs don't show anything.  How do I troubleshoot this?
>>     
>
> I just noticed that I do not have CONFIG_INOTIFY_USER enabled in the kernel.  
> Perhaps this has something to do with it?
>   

I'm not positive but I had several errors because I was missing that
option in my kernel.  You may want to build that one in regardless, just
to prevent future problems.  I seem to recall that the error I got told
me I needed that in the kernel but that was a few weeks ago and I'm not
a elephant.  :/  It's a good place to start though.

Dale

:-)  :-)


[-- Attachment #2: Type: text/html, Size: 1354 bytes --]

Re: [gentoo-user] Re: hald won't start

[Mick]

[-- Attachment #1: Type: text/plain, Size: 683 bytes --]

On Saturday 14 July 2007 21:08, Dale wrote:
> Mick wrote:

> > I just noticed that I do not have CONFIG_INOTIFY_USER enabled in the
> > kernel. Perhaps this has something to do with it?
>
> I'm not positive but I had several errors because I was missing that
> option in my kernel.  You may want to build that one in regardless, just
> to prevent future problems.  I seem to recall that the error I got told
> me I needed that in the kernel but that was a few weeks ago and I'm not
> a elephant.  :/  It's a good place to start though.

Thanks Dale, I will recompile the kernel as soon as it finishes emerging OOo 
(in the next day or so . . . )

-- 
Regards,
Mick

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-user] hald won't start

[Alessandro del Gallo]

Mick ha scritto:
> Hi All,
> 
> I noticed on a box of mine that hald won't start:
> ==========================================
> # /etc/init.d/hald restart
>  * Starting Hardware Abstraction Layer daemon ...          [ !! ]
> ==========================================
> 
> Confusingly, the logs don't show anything.  How do I troubleshoot this?



I had the same problem:

hald was sucking 99% cpu, removables devices didn't work
and launching hald from terminal with the option "hald --daemon=no 
--verbose=yes" i discovered it was on a infinite loop echo-ing
something about rules so I got in /etc/udev/rules.d
and found (by probing one by one) 99-libgphoto2.rules that
was HALS's murder.

I hope it's useful for you

Alessandro
-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] hald won't start

[Mick]

[-- Attachment #1: Type: text/plain, Size: 901 bytes --]

On Monday 16 July 2007 10:26, Alessandro del Gallo wrote:
> Mick ha scritto:
> > Hi All,
> >
> > I noticed on a box of mine that hald won't start:
> > ==========================================
> > # /etc/init.d/hald restart
> >  * Starting Hardware Abstraction Layer daemon ...          [ !! ]
> > ==========================================
> >
> > Confusingly, the logs don't show anything.  How do I troubleshoot this?
>
> I had the same problem:
>
> hald was sucking 99% cpu, removables devices didn't work
> and launching hald from terminal with the option "hald --daemon=no
> --verbose=yes" i discovered it was on a infinite loop echo-ing
> something about rules so I got in /etc/udev/rules.d
> and found (by probing one by one) 99-libgphoto2.rules that
> was HALS's murder.

Thank you Alessandro, what do you mean by probing - how did you probe them?
-- 
Regards,
Mick

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-user] hald won't start

[Alessandro del Gallo]

moving it to /tmp and restart hald,
becouse the error I got when launching
# hald --daemon=no --verbose=yes
is "Unhandled Rule (0)"
so deleting the rules (delete de rule, start hald, if it doesn't
works put the rule back , and move another) from etc/udev/rules.d
I debugged the problem, the bad rule was owned by gphoto2

sorry for my bad english

Mick ha scritto:
> On Monday 16 July 2007 10:26, Alessandro del Gallo wrote:
>> Mick ha scritto:
>>> Hi All,
>>>
>>> I noticed on a box of mine that hald won't start:
>>> ==========================================
>>> # /etc/init.d/hald restart
>>>  * Starting Hardware Abstraction Layer daemon ...          [ !! ]
>>> ==========================================
>>>
>>> Confusingly, the logs don't show anything.  How do I troubleshoot this?
>> I had the same problem:
>>
>> hald was sucking 99% cpu, removables devices didn't work
>> and launching hald from terminal with the option "hald --daemon=no
>> --verbose=yes" i discovered it was on a infinite loop echo-ing
>> something about rules so I got in /etc/udev/rules.d
>> and found (by probing one by one) 99-libgphoto2.rules that
>> was HALS's murder.
> 
> Thank you Alessandro, what do you mean by probing - how did you probe them?

-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] hald won't start

[Mick]

[-- Attachment #1: Type: text/plain, Size: 801 bytes --]

On Tuesday 17 July 2007 10:52, Alessandro del Gallo wrote:
> moving it to /tmp and restart hald,
> becouse the error I got when launching
> # hald --daemon=no --verbose=yes
> is "Unhandled Rule (0)"
> so deleting the rules (delete de rule, start hald, if it doesn't
> works put the rule back , and move another) from etc/udev/rules.d
> I debugged the problem, the bad rule was owned by gphoto2

The thing is I do not get any errors in /var/log/messages and I do not have 
gphoto2 (I do not have gnome on this box).

An interesting thing, when I emerge hal it is installed and started alright.  
If I try to restart it (or during boot) it fails.  No error messages other 
than [!!].  I am totally in the dark with this.

> sorry for my bad english

No problemo!
-- 
Regards,
Mick

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-user] hald won't start

[Mick]

[-- Attachment #1: Type: text/plain, Size: 2498 bytes --]


On Tuesday 17 July 2007 10:52, Alessandro del Gallo wrote:
> Mick ha scritto:
> > On Monday 16 July 2007 10:26, Alessandro del Gallo wrote:
> >> Mick ha scritto:
> >>> Hi All,
> >>>
> >>> I noticed on a box of mine that hald won't start:
> >>> ==========================================
> >>> # /etc/init.d/hald restart
> >>>  * Starting Hardware Abstraction Layer daemon ...          [ !! ]
> >>> ==========================================
> >>>
> >>> Confusingly, the logs don't show anything.  How do I troubleshoot this?
> >>
> >> I had the same problem:
> >>
> >> hald was sucking 99% cpu, removables devices didn't work
> >> and launching hald from terminal with the option "hald --daemon=no
> >> --verbose=yes" i discovered it was on a infinite loop echo-ing
> >> something about rules so I got in /etc/udev/rules.d
> >> and found (by probing one by one) 99-libgphoto2.rules that
> >> was HALS's murder.
> >
> > Thank you Alessandro, what do you mean by probing - how did you probe
> > them?

> moving it to /tmp and restart hald,
> becouse the error I got when launching
> # hald --daemon=no --verbose=yes
> is "Unhandled Rule (0)"
> so deleting the rules (delete de rule, start hald, if it doesn't
> works put the rule back , and move another) from etc/udev/rules.d
> I debugged the problem, the bad rule was owned by gphoto2

I've had a chance to get back to this machine and this is what it is showing:
==========================================
# hald --daemon=no --verbose=yes
11:16:08.930 [I] hald.c:533: hal 0.5.9
11:16:08.931 [I] hald.c:598: Will not daemonize
11:16:08.945 [I] hald_dbus.c:4807: local server is listening at 
unix:abstract=/var/run/hald/dbus-1LzyssbMzl,guid=faadd829b7aaff57dce0510046b451e8
11:16:09.174 [I] hald_runner.c:299: Runner has pid 31241
11:16:09.180 [W] ci-tracker.c:200: Could not get uid for connection: 
org.freedesktop.DBus.Error.NameHasNoOwner Could not get UID of 
name 'org.freedesktop.DBus': no such name
11:16:09.181 [E] hald_dbus.c:4462: Cannot get caller info for 
org.freedesktop.DBus
Runner started - allowed paths 
are '/usr/libexec:/usr/lib/hal/scripts:/usr/bin'
11:16:09.196 [I] hald_runner.c:180: runner connection is 0x8096988
11:16:09.330 [I] mmap_cache.c:251: cache mtime is 1184442336
*** [DIE] osspec.c:watch_fdi_files():349 : Unable to initialize inotify: 
Function not implemented
==========================================

What do you make of this?
-- 
Regards,
Mick

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-user] hald won't start

[Mick]

[-- Attachment #1: Type: text/plain, Size: 3049 bytes --]

On Saturday 04 August 2007 11:21, Mick wrote:
> On Tuesday 17 July 2007 10:52, Alessandro del Gallo wrote:
> > Mick ha scritto:
> > > On Monday 16 July 2007 10:26, Alessandro del Gallo wrote:
> > >> Mick ha scritto:
> > >>> Hi All,
> > >>>
> > >>> I noticed on a box of mine that hald won't start:
> > >>> ==========================================
> > >>> # /etc/init.d/hald restart
> > >>>  * Starting Hardware Abstraction Layer daemon ...          [ !! ]
> > >>> ==========================================
> > >>>
> > >>> Confusingly, the logs don't show anything.  How do I troubleshoot
> > >>> this?
> > >>
> > >> I had the same problem:
> > >>
> > >> hald was sucking 99% cpu, removables devices didn't work
> > >> and launching hald from terminal with the option "hald --daemon=no
> > >> --verbose=yes" i discovered it was on a infinite loop echo-ing
> > >> something about rules so I got in /etc/udev/rules.d
> > >> and found (by probing one by one) 99-libgphoto2.rules that
> > >> was HALS's murder.
> > >
> > > Thank you Alessandro, what do you mean by probing - how did you probe
> > > them?
> >
> > moving it to /tmp and restart hald,
> > becouse the error I got when launching
> > # hald --daemon=no --verbose=yes
> > is "Unhandled Rule (0)"
> > so deleting the rules (delete de rule, start hald, if it doesn't
> > works put the rule back , and move another) from etc/udev/rules.d
> > I debugged the problem, the bad rule was owned by gphoto2
>
> I've had a chance to get back to this machine and this is what it is
> showing: ==========================================
> # hald --daemon=no --verbose=yes
> 11:16:08.930 [I] hald.c:533: hal 0.5.9
> 11:16:08.931 [I] hald.c:598: Will not daemonize
> 11:16:08.945 [I] hald_dbus.c:4807: local server is listening at
> unix:abstract=/var/run/hald/dbus-1LzyssbMzl,guid=faadd829b7aaff57dce0510046
>b451e8 11:16:09.174 [I] hald_runner.c:299: Runner has pid 31241
> 11:16:09.180 [W] ci-tracker.c:200: Could not get uid for connection:
> org.freedesktop.DBus.Error.NameHasNoOwner Could not get UID of
> name 'org.freedesktop.DBus': no such name
> 11:16:09.181 [E] hald_dbus.c:4462: Cannot get caller info for
> org.freedesktop.DBus
> Runner started - allowed paths
> are '/usr/libexec:/usr/lib/hal/scripts:/usr/bin'
> 11:16:09.196 [I] hald_runner.c:180: runner connection is 0x8096988
> 11:16:09.330 [I] mmap_cache.c:251: cache mtime is 1184442336
> *** [DIE] osspec.c:watch_fdi_files():349 : Unable to initialize inotify:
> Function not implemented
> ==========================================
>
> What do you make of this?

I removed /usr/share/hal and tried to remerge it.  It complained that I "MUST" 
compile pciutils without the zlib flag.  Not sure why pciutils has the zlib 
flag as a default, but if it is causing a problem with hal shouldn't it be 
removed?

Anyway, after masking zlib for pciutils and remerging both packages it now 
seems to be working again . . .

Thanks for your suggestions.
-- 
Regards,
Mick

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-user] hald won't start

[Alessandro del Gallo]

Hi,

Last 2 lines of your log:
  Unable to initialize inotify:
 > > Function not implemented

inotify comes from kernel, in my config:

  elwood linux # zcat /proc/config.gz |grep -i notify
  CONFIG_INOTIFY=y
  CONFIG_INOTIFY_USER=y
  CONFIG_DNOTIFY=y

in menuconfig:

  ┌─────── File systems

[...]

      [*] Inotify file change notification support

      [*]   Inotify support for userspace

bye bye


Mick ha scritto:
> On Saturday 04 August 2007 11:21, Mick wrote:
>> On Tuesday 17 July 2007 10:52, Alessandro del Gallo wrote:
>>> Mick ha scritto:
>>>> On Monday 16 July 2007 10:26, Alessandro del Gallo wrote:
>>>>> Mick ha scritto:
>>>>>> Hi All,
>>>>>>
>>>>>> I noticed on a box of mine that hald won't start:
>>>>>> ==========================================
>>>>>> # /etc/init.d/hald restart
>>>>>>  * Starting Hardware Abstraction Layer daemon ...          [ !! ]
>>>>>> ==========================================
>>>>>>
>>>>>> Confusingly, the logs don't show anything.  How do I troubleshoot
>>>>>> this?
>>>>> I had the same problem:
>>>>>
>>>>> hald was sucking 99% cpu, removables devices didn't work
>>>>> and launching hald from terminal with the option "hald --daemon=no
>>>>> --verbose=yes" i discovered it was on a infinite loop echo-ing
>>>>> something about rules so I got in /etc/udev/rules.d
>>>>> and found (by probing one by one) 99-libgphoto2.rules that
>>>>> was HALS's murder.
>>>> Thank you Alessandro, what do you mean by probing - how did you probe
>>>> them?
>>> moving it to /tmp and restart hald,
>>> becouse the error I got when launching
>>> # hald --daemon=no --verbose=yes
>>> is "Unhandled Rule (0)"
>>> so deleting the rules (delete de rule, start hald, if it doesn't
>>> works put the rule back , and move another) from etc/udev/rules.d
>>> I debugged the problem, the bad rule was owned by gphoto2
>> I've had a chance to get back to this machine and this is what it is
>> showing: ==========================================
>> # hald --daemon=no --verbose=yes
>> 11:16:08.930 [I] hald.c:533: hal 0.5.9
>> 11:16:08.931 [I] hald.c:598: Will not daemonize
>> 11:16:08.945 [I] hald_dbus.c:4807: local server is listening at
>> unix:abstract=/var/run/hald/dbus-1LzyssbMzl,guid=faadd829b7aaff57dce0510046
>> b451e8 11:16:09.174 [I] hald_runner.c:299: Runner has pid 31241
>> 11:16:09.180 [W] ci-tracker.c:200: Could not get uid for connection:
>> org.freedesktop.DBus.Error.NameHasNoOwner Could not get UID of
>> name 'org.freedesktop.DBus': no such name
>> 11:16:09.181 [E] hald_dbus.c:4462: Cannot get caller info for
>> org.freedesktop.DBus
>> Runner started - allowed paths
>> are '/usr/libexec:/usr/lib/hal/scripts:/usr/bin'
>> 11:16:09.196 [I] hald_runner.c:180: runner connection is 0x8096988
>> 11:16:09.330 [I] mmap_cache.c:251: cache mtime is 1184442336
>> *** [DIE] osspec.c:watch_fdi_files():349 : Unable to initialize inotify:
>> Function not implemented
>> ==========================================
>>
>> What do you make of this?
> 
> I removed /usr/share/hal and tried to remerge it.  It complained that I "MUST" 
> compile pciutils without the zlib flag.  Not sure why pciutils has the zlib 
> flag as a default, but if it is causing a problem with hal shouldn't it be 
> removed?
> 
> Anyway, after masking zlib for pciutils and remerging both packages it now 
> seems to be working again . . .
> 
> Thanks for your suggestions.

-- 
gentoo-user@gentoo.org mailing list

[gentoo-user] hald won't start

[Alexander Skwar]

Hello!

Since this morning, I'm unable to start hald. This is in so
far a pretty big problem, as this means, that I cannot boot :|

After setting HALD_VERBOSE=yes in /etc/rc.conf, I found that
it filled the syslog with a lot of lines like

  device_info.c:984: Unhandled rule (0)!

Searching the archive of this list, I found a thread started
by Mick on July 14; see eg. 
http://www.nabble.com/forum/ViewPost.jtp?post=11596495&framed=y

In this thread, Alessandro del Gallo suggested to move files
from /etc/udev/rules.d out of the way, to figure out what's
exactly causing the issue. I did that, and even when I have
*no* files at all in rules.d anymore, I cannot start hald by
running "/usr/sbin/hald --daemon=no --verbose=yes".

>From what other sources is hald fetching rules? Ie. what
other file/directory might I need to empty?

Thanks,
Alexander

-- 
gentoo-user@gentoo.org mailing list