[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: ORBit 2.8.1 uploaded to experimental



Today I was bit by this bug somehow. I'm not sure how. I haven't
upgraded to the Orbit in experimental. I was similarily at a loss of how
to fix it.

It took me about 5 hours to track down with my limited knowledge of C
and Bonobo.

I ended up editing libbonobo-activation4, in um...
bonobo-activation-something, in a function named get_tmpdir(). There is
a fall back condition there if tmpdir isn't properly set by linc, set it
manually to (/tmp/orbit-%s, username). I just made this action the
default (removed the linc attempt). This was obviously not the proper
way to fix it, but I didn't know how best to go about it givin my
limited understanding of the situation. It works now though.

Any clue how this bit me?

+++-==============-==============-============================================
ii  orbit2         2.6.3-2        a CORBA ORB
ii  bonobo-activat 2.2.4-1        Bonobo Activation Framework
ii  libbonobo-acti 2.2.4-1        Bonobo Activation Framework --runtime



On Wed, 2003-10-01 at 13:52, Sebastian Rittau wrote:
> Hi!
> 
> I've just uploaded ORBit 2.8.1 to experimental. Unfortunately, this
> version is troublesome. Since Michael Meeks (the upstream maintainer)
> had the great idea to internalize linc into this "binary-compatible"
> release of ORBit, bonobo-activation breaks. The problem is as follows:
> When a socket is about to created in /tmp/orbit-user the
> linc_get_tmpdir() function seems to fail. Therefore bonobo-activation
> tries to open a file in the root directory which obviously fails and
> leads to subsequent failures.
> 
> I don't know the exact reasons for this failure, so if anything has any
> useful insights or knows where to dig, please let me know.
> 
> When I simply tried to rebuild bonobo-activation, it failed telling me
> something along the lines of "bonobo-activation is obsolete, don't use
> it". So, for now I resorted to conflicting with libbonono-activation4.
> This means that the whole of GNOME would need to be rebuilt. At the
> moment, I am devoid of a clue what to try next. Any ideas?
> 
>  - Sebastian
-- 
Jerry Haltom <jhaltom@feedbackplusinc.com>
Feedback Plus, Inc.



Reply to: