Re: its#179 (ITS#181)

[gomez@lhola.engr.sgi.com]

hw,

You are right there are quite a few other places where error
recovery leaves messy stuff hanging around ...I will try to 
look into that soon.
		Juan

On Thu, 27 May 1999 hwr@pilhuhn.de wrote:

> This looks like being a different incarnation of ITS#101
> 
> 
> The Problem I see:
> 
> backends/back-ldbm/modify.c:
> 
> ldbm_back_modify()
> 
>     err = add_values( e, mod, op->o_ndn );
> 
>     shcemackeck(e)
> 
> 
>     if error goto error_return
> 
> error_return:
> 
> 	        cache_return_entry_w( &li->li_cache, e );
> 
> But at this point, e is already modified, so the modyfied value
> gets cached. I might be wrong though.
> 
> 
> -- 
>           See <a href="http://www.netbsd.org";>NetBSD</a> for a multiplatform OS
> Two signs from a Majorcan shop entrance:
>                 - English well talking.
>                 - Here speeching American.
> 
>