Re: 2.1.13 cfexecd --no-splay requires argument

[Christian Pearce]

I looked at this a little while back, we should sit down and unify the
interface between cfexecd, cfagent and cfservd and maybe the others.  I
think cfexecd should have most if not all the cfagent arugments.
Especially -D and -N.  It would be nice to put -D or -N on the command
line for cfexecd in cron.  Especially if you need to add a new platform
like redhat_es_4 which isn't programmed in yet.  And you don't want to
upgraded.  But that is a whole other problem I suppose.

Maybe we can put together a formal set of requirements for the
arugements and document it in the wiki.  This pay someone through
donations to implement.  Provided Mark blesses the requirement.

We should also clean up all the bugs like --no-splay requiring an
argument.

On Thu, 2005-03-10 at 14:43 +0000, Jim Wight wrote:
> If given --no-splay as argument, cfexecd complains that --no-splay
> requires an argument. It is clear from the source why that is:
> 
> struct option CFDOPTIONS[] =
>    {
>    { "help",no_argument,0,'h' },
>    { "debug",optional_argument,0,'d' },
>    { "verbose",no_argument,0,'v' },
>    { "file",required_argument,0,'f' },
>    { "no-splay",required_argument,0,'q' },
>    { "no-fork",no_argument,0,'F' },
>    { "once",no_argument,0,'1'},
>    { "foreground",no_argument,0,'g'},
>    { "parse-only",no_argument,0,'p'},
>    { "ld-library-path",required_argument,0,'L'},
>    { NULL,0,0,0 }
>    };
> 
> Furthermore, -q and -f are rejected as invalid options. They've been
> forgotten about altogether. From the source:
> 
> while ((c=getopt_long(argc,argv,"L:d:vhpFV1g",CFDOPTIONS,&optindex)) != EOF)
>                                  ^^^^^^^^^^^
> 
> -V is not documented (at 
> http://www.cfengine.org/docs/cfengine-Reference.html#Cfexecd-reference)
> 
> Jim
> 
> 
> 
> 
> 
> _______________________________________________
> Bug-cfengine mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/bug-cfengine
-- 
Christian Pearce
http://www.sysnav.com
http://www.commnav.com
http://www.perfectorder.com

signature.asc
Description: This is a digitally signed message part