On Thu, Jul 01, 2004 at 10:53:03AM -0700, Jeremy Allison wrote:
> On Thu, Jul 01, 2004 at 11:05:18AM -0500, Chris Garrigues wrote:
> > > From: Chris Garrigues <[EMAIL PROTECTED]>
> > > Date: Wed, 30 Jun 2004 07:54:53 -0500
> > >
> > > (I'm still trying to get a new debug log for you.)
> >
> > Sorry this took a while, but unfortunately I have other responsibilities.
> >
> > New log at
> >
> > http://www.trinsics.com/~cwg/smblog
> >
> > The delay was from 10:09:52 to 10:11:17.
>
> At last ! Thanks !!!!!
>
> The problem is here :
>
> 249646 [2004/07/01 10:09:52, 5] lib/smbldap.c:smbldap_search(932) 249647
> smbldap_search: base => [dc=trinsics,dc=com], filter =>
> [(&(sambaSID=S-1-5-21-904170827-898255731-2295272421-1407)(
> objectclass=sambaSamAccount))], scope => [2]
> 249648 [2004/07/01 10:09:52, 5] lib/smbldap.c:smbldap_search(950)
> 249649 smbldap_search: waiting 1117501 milliseconds for LDAP replication.
> 249650 [2004/07/01 10:11:17, 6] param/loadparm.c:lp_file_list_changed(2687)
> 249651 lp_file_list_changed()
> 249652 file /etc/samba3/smb.conf -> /etc/samba3/smb.conf last mod_time: Wed Jun
> 30 16:11:03 2004
> 249653
> 249654 [2004/07/01 10:11:17, 5] smbd/connection.c:claim_connection(170)
> 249655 claiming 0
> 249656 [2004/07/01 10:11:17, 0] lib/util_sock.c:get_peer_addr(1000)
> 249657 getpeername failed. Error was Transport endpoint is not connected
>
>
> Nothing to do with the new open code. There is a bug (not sure
> what yet) with calculating the sleep time in smbldap_search()
> (in fact I have *NO* idea what this code is for....) that causes
> it to sleep for much more than 30 second - client times out, fun
> ensues....
Ok, I've fixed up the smbldap_search code and added a max sleep replication
time of 5 seconds. I've committed to 3.0 svn if you want to try it. This
should fix the issue you've reported here. Let me know if not....
Thanks,
Jeremy.
--
To unsubscribe from this list go to the following URL and read the
instructions: http://lists.samba.org/mailman/listinfo/samba
