--DocE+STaALJfprDB
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On 2008-04-21 at 21:56 +0200 Volker Lendecke sent off:
> Sure. I meant to try to default both to the "ldap timeout"


yes, I understood you this way.=EE=80=80But the current "ldap timeout"=20
default IS the "infinit 15s" that I mentioned. And as we currently do=20
not distinguish ldap timeout for connection and for operation we have=20
to wait so long for the next ldap host to be asked. Of course the=20
timeout is feasible for ldap operations but not for connection=20
timeouts, which even multiply if we have to fall back to the 3rd or=20
4th ldap server.


> in case this is set explicitly. If no explicit ldap timeout
> is set, I agree that a shorter connect timeout has benefits,
> but I don't want to change behaviour for those who
> consciously did set that parameter to something.


why don't you want to change the default here especially if the=20
proposed default is obviously more useful?

Cheers
Bj=C3=B6rn

--DocE+STaALJfprDB
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.9 (GNU/Linux)

iEYEARECAAYFAkgNqSUACgkQdoo0s+hIejkiWwCgru0jeAW4D0 v39JR5+pfmUZrJ
CRsAn2kNW5ffv0JdpMX3sv1DoGqi1KEW
=ZygC
-----END PGP SIGNATURE-----

--DocE+STaALJfprDB--