Re: oaf race conditions ...

[Maciej Stachowiak <mjs noisehavoc org>]

On 26Sep2001 12:37AM (-0400), Michael Meeks wrote:
> 
> Hi Maciej,
> 
> 	There seems to be a rather annoying oaf activation race still
> occuring - I have the latest (HEAD) oafd installed, and if I do:
> 
> 	oaf-slay
> 	test/oaf-run-query & test/oaf-run-query & test/oaf-run-query &
> 	ps ax
> 
> 12538 ?        S      0:01 oafd --ac-activate --ior-output-fd=6
> 12541 ?        S      0:01 oafd --ac-activate --ior-output-fd=6
> 12543 ?        S      0:01 oafd --ac-activate --ior-output-fd=6
> 
> 	of course subsequent oaf-run-queries don't activate yet more oafs;
> but ... this is really not good news, especialy at gnome-session startup
> time where we can have a whole load of things batch forked.

Hi Michael,

Wow, that sounds like a serious problem. I will make it my top
priority as soon as I get the releases for the alpha done.

I really appreciate having the reproducible case (I tried it myself
and saw the same bad results).
 
> 	I tried to verify the problem on HEAD bonobo-activation, but it
> seems that running bonobo-activation-run-query crashes the
> bonobo-activation server for reasons that are totaly opaque to me:

It definitely worked at some point (in fact I hade make check passing
too), and I have not made many changes lately, so I am a bit surprised
by this problem. But I will investigate it as well.

 - Maciej