Re: [nssldap] release 0.2 of nss-ldapd

[Howard Chu <hyc [at] highlandsun.com>]

Arthur de Jong wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
>
> On Tue, 19 Jun 2007, David Houlder wrote:
> > That seems like the main point of having a daemonised nss_ldap to me. 
> > The current in-process implementation has side effects. Some things are 
> > unavoidably shared with the calling application: file descriptors and 
> > signal handling are the two most obvious ones. The file descriptor issue 
> > is currently handled reasonably well I think, but it's definitely ugly.
>
> nss-ldapd still uses file descriptors for communicating with the daemon 
> (named socket) but these file descriptors are closed after each request.
>
> If the caller messes around with file descriptors with longer-lived 
> requests (eg setxxent(), getxxent(), endxxent()) he's getting what he 
> deserves (which probably also goes for other nss modules).

Right, there's really not a lot of alternatives. You could use a SysV shared 
memory region with a hardcoded key. That would also avoid the whole SIGPIPE 
problem.

--
  -- Howard Chu
  Chief Architect, Symas Corp.  http://www.symas.com
  Director, Highland Sun        http://highlandsun.com/hyc/
  Chief Architect, OpenLDAP     http://www.openldap.org/project/