Thank you very much for your response. Actually that example worked, but the
Dtrace guide defines the system action as destructive. So, is the following
action safe to use then?
system("getent passwd %d | cut -d: -f5",uid);
I am not changing anything at the system level , just reading it.
Regards.
Deniz.
On 9/11/07, Bryan Cantrill <[EMAIL PROTECTED]> wrote:
>
> On Tue, Sep 11, 2007 at 08:55:48PM +0100, Jon Haslam wrote:
> > Hi,
> >
> > >Is there anyway for dtrace to report me the actual user name instead of
> a uid.
> > >
> > >For example If I have the following code somewhere in in my script, I
> know that I am getting uid with printf function
> > >
> > >syscall::open:return,syscall::open64:return
> > >{
> > > .............
> > > printf("%d,%s,... ",uid, ......);
> > >
> > >}
> > >
> > >Is there any way to get the username corresponding to that uid? I know
> that I can look at /etc/password and get the username but is there anything
> I can do directly from Dtrace?
> > >
> >
> > You can't obtain the username from within probe context but you can
> > use the system() action to do this. Actually, this very example is given
> > in the docs. Check out the system() action example at:
> >
> > http://docs.sun.com/app/docs/doc/817-6223/chp-actsub#hic
>
> Boy does that example take me back. As you might guess, it was written
> by me on or about Tue Jul 20 11:56:30 CDT 2004. ;) Firefox was still
> called Firebird, and I was on the plane, on the way to a demo that
> would become infamous:
>
> http://blogs.sun.com/bmc/date/20040726
>
> Needless to say, the return flight was not quite so relaxing: I spent
> it root-causing what exactly had gone wrong during the demo. Given
> that they have since given a public reference for DTrace, I think it's
> safe to reveal that the customer in question was FedEx, and that the demo
> in question was in Memphis. Not sure if any on the list were in the
> room that day (if you are, thank you for having been so forgiving!),
> but I think it remains my most stressful moment in front of a customer...
>
> Anyway, that documentation will forever for me be the calm before the
> storm... ;)
>
> - Bryan
>
> --------------------------------------------------------------------------
> Bryan Cantrill, Solaris Kernel Development. http://blogs.sun.com/bmc
>
_______________________________________________
dtrace-discuss mailing list
[email protected]
