Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 198670 - net-misc/vpnc-0.5.1 on amd64 says 'dbus-send: Unknown type " uint32"'
Summary: net-misc/vpnc-0.5.1 on amd64 says 'dbus-send: Unknown type " uint32"'
Status: RESOLVED WORKSFORME
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: Current packages (show other bugs)
Hardware: AMD64 Linux
: High trivial (vote)
Assignee: AMD64 Project
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2007-11-10 13:00 UTC by Erik Quaeghebeur
Modified: 2007-11-12 13:14 UTC (History)
2 users (show)

See Also:
Package list:
Runtime testing required: ---


Attachments
emerge -info (emerge--info.txt,3.24 KB, text/plain)
2007-11-10 14:12 UTC, Erik Quaeghebeur
Details
emerge -pv vpnc (emerge-pv_vpnc.txt,239 bytes, text/plain)
2007-11-10 14:12 UTC, Erik Quaeghebeur
Details
/usr/sbin/vpnc --debug 3 myconffile 2> usrsbinvpnc--debug_3_myconffile.stderr (usrsbinvpnc--debug_3_myconffile.stderr,34 bytes, text/plain)
2007-11-10 14:14 UTC, Erik Quaeghebeur
Details
/usr/sbin/vpnc --debug 3 myconffile 1> usrsbinvpnc--debug_3_myconffile.stdout (usrsbinvpnc--debug_3_myconffile.stdout,63.43 KB, text/plain)
2007-11-10 14:14 UTC, Erik Quaeghebeur
Details
/usr/sbin/vpnc --debug 3 myconffile 1> usrsbinvpnc--debug_3_myconffile.stdout+stderr 2>&1 (usrsbinvpnc--debug_3_myconffile.stdout+stderr,63.47 KB, text/plain)
2007-11-10 14:17 UTC, Erik Quaeghebeur
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Erik Quaeghebeur 2007-11-10 13:00:53 UTC
vpnc-0.5.1 on amd64 says 'dbus-send: Unknown type " uint32"' when started.
This did not occur with previous versions.
The program still seems to work well.

Perhaps this is for upstream? (Please advise, I will then do the reporting.)

Reproducible: Always
Comment 1 Christian Faulhammer (RETIRED) gentoo-dev 2007-11-10 13:16:33 UTC
Please elaborate a bit...I wonder where vpnc is using dbus, so I assume it is elsewhere.  Provide emerge --info, emerge -pv vpnc, where does the message occur (init script, manual call), as user or root, what is the output with a higher debug level?
Comment 2 Erik Quaeghebeur 2007-11-10 14:12:18 UTC
Created attachment 135646 [details]
emerge -info
Comment 3 Erik Quaeghebeur 2007-11-10 14:12:53 UTC
Created attachment 135648 [details]
emerge -pv vpnc
Comment 4 Erik Quaeghebeur 2007-11-10 14:14:03 UTC
Created attachment 135650 [details]
/usr/sbin/vpnc --debug 3 myconffile 2> usrsbinvpnc--debug_3_myconffile.stderr
Comment 5 Erik Quaeghebeur 2007-11-10 14:14:54 UTC
Created attachment 135652 [details]
/usr/sbin/vpnc --debug 3 myconffile 1> usrsbinvpnc--debug_3_myconffile.stdout
Comment 6 Erik Quaeghebeur 2007-11-10 14:17:19 UTC
Created attachment 135653 [details]
/usr/sbin/vpnc --debug 3 myconffile 1> usrsbinvpnc--debug_3_myconffile.stdout+stderr 2>&1
Comment 7 Erik Quaeghebeur 2007-11-10 14:22:06 UTC
(In reply to comment #1)
> Provide emerge --info, emerge -pv vpnc, 

attached

> where does the message occur (init script, manual call), 

Both, but this is not entirely clear to me: I thought vpnc always uses a script and conf (even if it is the default ones); I've used the one for my university's vpn (which creates a split route as described in vpnc's README).

> as user or root, 

root

> what is the output with a higher debug level?

attached; I've noticed the location of the dbus(?) stderr output can change with in the vpnc stdout output 
Comment 8 Christian Faulhammer (RETIRED) gentoo-dev 2007-11-10 14:37:14 UTC
Hmm, vpnc is not using dbus itself, so it must be in a deeper layer.  Do you have any other programs that manipulate the routing table?  If so, please check if they also have a similar message.
 You can also debug it by starting $(dbus-monitor --system) who calls dbus in that moment.
 I asked for the init-script (from /etc/init.d/vpnc) because a call there could be different to the one you use manually.
Comment 9 Erik Quaeghebeur 2007-11-10 15:57:05 UTC
(In reply to comment #8)
> Hmm, vpnc is not using dbus itself, so it must be in a deeper layer.  Do you
> have any other programs that manipulate the routing table?  If so, please check
> if they also have a similar message.

# resolvconf -u
dbus-send: Unknown type " uint32"

I should've know, vpnc-0.5.1 pulled in resolvconf-gentoo...
This bug should now probably be renamed, you can do that Christian.

>  You can also debug it by starting $(dbus-monitor --system) who calls dbus in
> that moment.

Mmm, I'm not sure I've done what you meant:
start dbus-monitor --system in one konsole, ran resolvconf -u in another:
I got 

# dbus-monitor --system
signal sender=org.freedesktop.DBus -> dest=:1.33 path=/org/freedesktop/DBus; interface=org.freedesktop.DBus; member=NameAcquired
   string ":1.33"
signal sender=org.freedesktop.DBus -> dest=(null destination) path=/org/freedesktop/DBus; interface=org.freedesktop.DBus; member=NameOwnerChanged
   string ":1.34"
   string ""
   string ":1.34"
signal sender=org.freedesktop.DBus -> dest=(null destination) path=/org/freedesktop/DBus; interface=org.freedesktop.DBus; member=NameOwnerChanged
   string ":1.34"
   string ":1.34"
   string ""

Comment 10 Samuli Suominen (RETIRED) gentoo-dev 2007-11-11 06:43:00 UTC
on stable amd64,

unique / # resolvconf -u
 * Caching service dependencies ... [ ok ]
Comment 11 Christian Faulhammer (RETIRED) gentoo-dev 2007-11-12 11:28:12 UTC
(In reply to comment #10)
> on stable amd64,
> 
> unique / # resolvconf -u
>  * Caching service dependencies ... [ ok ]

 So, that is a bit weird.  On an amd64 test system I can't reproduce either, and the DBus monitor output tells me that the core of DBus issues the error.  Anyone against closing this bug?  Erik, if you want to avoid nasty warnings, you will maybe profit from bug 196276, as soon as I introduce a USE flag for resolvconf-gentoo.
Comment 12 Erik Quaeghebeur 2007-11-12 12:47:06 UTC
(In reply to comment #11)
> Anyone against closing this bug?  

Not me, go ahead. WORKSFORME (well, for you and samuli) seems appropriate.

> Erik, if you want to avoid nasty warnings,
> you will maybe profit from bug 196276, as soon as I introduce a USE flag for
> resolvconf-gentoo.

I quite like (the idea of) resolvconf, so I think I'll live with the warning (I even added the resolvconf use-flag to my make.conf). If it really has an impact later on, I'll reopen the bug. But I guess it'll just disappear in time, when my system feels like it...