[02:07] <[g2] > where would be a good place to start for building a customized server ?
[04:25] <tarvid> installed webmin-slbackup and it does not appear in the webmin menus
[04:26] <tarvid> the module appears to have been installed /usr/share/webmin/slbackup/WebminSLBackup.pm
[04:26] <tarvid> it is in webmin.acl
[04:27] <tarvid> it is not in module.infos.cache
[04:29] <tarvid> it is in module.info in the source
[04:34] <tarvid> i have both /usr/share/webmin-1.250 and /usr/share/webmin
[04:34] <tarvid> slbackup is in /usr/share/webmin but not in /usr/share/webmin-1.250
[04:35] <tarvid> could that be the problem?
[05:08] <tarvid> After much stumbling about, I found the problem. webmin updates via different directories in /usr/share
[05:08] <tarvid> ubuntu installs to a directory without the version suffix
[06:44] <Mercury> I'm trying to move my server over from the user land NFS server with ugidd (which more or less works, but which has some Issues), to NFSv4 with the kernel land NFS server.
[06:45] <Mercury> At the moment I have some progress, it gets mounted fine, ls -l on stuff shows the proper usernames, with the ones that only exist on the server shown as nobody.
[06:45] <Mercury> However as best I can tell, for actually _accessing_ files the raw UID is used, and I can't for the life of me figure out _why_.
[06:46] <Mercury> Any suggestions?
[06:51] <Mercury> Yep, it is most definitely doing the proper translation when going from the server to the client, but it doesn't seem to be for going from the client to the server.
[07:00] <Mercury> And..  I don't know enough about the workings of nfsv4 to debug this properly, hrm.
[07:05] <lionelp> NFSv4 is not currently considered as stable
[07:05] <lionelp> it is still exeperimenting stuff
[07:07] <Mercury> Understood, but I'd still like to track down this problem.
[07:07] <lionelp> I search the web few weeks ago, and i do not found viable informations :-(
[07:08] <lionelp> yes, no pb
[07:08] <Mercury> Did you have the same problem?
[07:08] <lionelp> non, i did not try yet to migrate NFSv4
[08:20] <Mercury> Alright, this has one, simple, itty bitty question..  Why the hell is the client sending back UIDs instead of names?
[08:25] <spike> Mercury: what's the subject? LDAP?
[08:26] <Mercury> spike: NFSv4.
[08:27] <Mercury> spike: I have a problem, after poking at things a bit I've come to the conclusion that the clients are not sending user names back, they are sending UIDs back, which makes no sense at all.
[08:35] <spike> Mercury: uh, sending back user names? what are you doing and what's not working?
[08:36] <spike> Mercury: have you run tcpdump and sniffed the traffic? that would clear out any doubt on what clients are sending back
[08:37] <Mercury> spike: I have.
[08:37] <Mercury> spike: The server is sending <user>@<domain>, nice and dandy.
[08:37] <Mercury> spike: The client is sending the UIDs, not even in ASCII.
[08:37] <Mercury> spike: This, of course, completely breaks the ID mapping.
[08:38] <Mercury> spike: And I can't figure out _WHY_.
[08:43] <Mercury> spike: Any ideas?  Any at all?
[08:43] <Mercury> spike: A known working configuration?  Anything that might help?
[08:54] <spike> Mercury: nope, sorry, I thought it was some ldap related prob
[09:02] <Mercury> spike: Alright.