[00:00] <Vog-work> Quiero una version de eso en ingles.
[00:04] <jmedina> Vog-work: ayuda a traducir :D
[00:17] <Vog-work> Mi espanol, es muy "rusty" . I haven't been fluent since 1993-94
[00:20] <Vog-work> Hasta todos
[01:51] <MagicFab> How do I *disable* update checks from command line ?
[01:51] <MagicFab> 9sorry wrong channel)
[02:50] <genii> MagicFab: Enthusiastic ubuntu-ing from you earlier in -classroom was interesting to see, by the way :)
[02:52] <MagicFab> genii,
[02:52] <MagicFab> :)
[02:53]  * genii hands out a round of coffees
[03:00] <genii> Because min uid is supposed to be 1000 if you don't want system uid and root names logging in
[03:01] <janauss> anyone here know how to get grub to install on a nvidia fakeraid/dmraid? jaunty install borked the grub configuration
[03:29]  * mattt could use a coffee
[03:30]  * genii hands mattt a large coffee in an Ubuntu mug
[03:39]  * mattt ^5s genii
[03:40] <ajmitch> genii: as long as you've got enough for everyone
[03:42]  * genii hands ajmitch an Ubuntu mug as well, filled with life-giving coffee
[03:42] <ajmitch> excellent
[03:43]  * ScottK goes for a life giving Scotch
[03:44] <ajmitch> a bit early in the day for that here
[03:44] <ajmitch> as nice as it would be
[03:45] <PhotoJim> good time of day for it here :)
[03:45] <PhotoJim> but I'd have to crack a fresh bottle of 18-year-old Tallisker to do it
[03:45] <genii> Nice
[03:45] <ScottK> Tallisker is good.
[03:46] <PhotoJim> I've not tried it yet... thought I should.
[03:46] <PhotoJim> I was really close to the distillery a year and a half ago.  should have visited it.
[03:47] <carmony> Question: I have a Ubuntu 8.04 server running apache and php, we just moved from another ubuntu server, but for some reason top is showing 90% wa cpu usage during peaks times
[03:47] <carmony> what can I do to debug what is causing such a huge IO issue?
[03:57] <twb> Isn't wa idle?
[03:58] <twb> Rather, waiting for I/O.
[03:59] <twb> "wa -- iowait    Amount of time the CPU has been waiting for I/O to complete."
[03:59] <twb> So basically what that's telling you is that your PHP crap is highly I/O bound.
[03:59] <|dthacker|> carmony: what is your storage on the new server?
[04:00] <mattt> carmony: and what changed hardware-wise?
[04:02] <|dthacker|> carmony: are you connecting to the same DB instance.?   Is the DB tuning different?
[04:04] <carmony> k, we have a db server and a web server
[04:05] <carmony> we moved the web server to a beefier dual quad core machine
[04:05] <mattt> and what's changed?
[04:05] <carmony> it was on a dual dual
[04:05] <carmony> it was on a 8.10 Ubuntu server machine, now its on a 8.04 server
[04:05] <mattt> sure you're running an SMP kernel and that all procs are recognized?
[04:05] <carmony> lol,
[04:05] <twb> mattt: how will that help with his I/O problem?
[04:06] <carmony> I'm a developer first, and sys admin second, so I'm not sure what that all means
[04:07] <mattt> twb: won't directly, but it's a new box ... these are simple things to check
[04:08] <mattt> carmony: on that note, has the disk configuration changed?
[04:10] <mattt> twb: wb :)
[04:10] <carmony> new HD
[04:10] <mattt> carmony: what was it before and what is it now?
[04:10] <carmony> but nothing fancy
[04:10] <carmony> no raid or anything
[04:12] <carmony> it was on virtual machine before
[04:12] <twb> mattt: fbi crashed my laptop :-(
[04:13] <mattt> twb: :(
[04:16] <captainkirk> is anyone aware of some app that will allow me to sync data from a W2K server to drives on ubuntu server?
[04:16] <twb> captainkirk: Samba
[04:16] <captainkirk> ok, so I already running samba in my ubuntu server on the same subnet
[04:18] <captainkirk> ok. another q. how do i rename my ubuntu server?
[04:20] <PhotoJim> captainkirk: sudo hostname newhostname
[04:21] <micahf> hi, I'm sort of new to TCP/IP but here is my question: I've set up my ubuntu-server box as a gateway using iptables forwarding.  The computer picks up internet and sends it to the WAN port on my wireless router.  The problem is that I cannot access the gateway using its local domain name from the router's LAN.  How can I get this to work?
[04:22] <micahf> the address is "server.local" but it doesn't resolve
[04:22] <micahf> avahi is set up and it worked before when the computer was part of the LAN
[04:23] <micahf> it just doesn't resolve when the computer is the gateway for the router
[04:24] <captainkirk> how do i restart the network after changing the static ip in interfaces?
[04:24] <micahf> captainkirk: you could do "sudo /etc/init.d/networking restart"
[04:25] <micahf> captainkirk: that way you would also know that it should work on reboot
[04:25] <captainkirk> micahf: thanks. i was using network instead of networking.... :)  lots to learn
[04:25] <micahf> I hear you
[04:26] <Kamping_Kaiser> captainkirk, i'd suggest restarting networks from inside screen
[04:27] <micahf> any ideas why I can't access my router's gateway by hostname?
[04:27] <captainkirk> i restarted from ssh console.... it froze (of course) but now logged back in with new ip and all good
[04:31] <captainkirk> twb: where should i look to find info to configure samba to sync data files from W2K server?
[04:32] <twb> It depends what you mean by "sync data"
[04:33] <captainkirk> twb: i would like to have a realtime sync if possible (like a networked mirror) otherwise a schedule file sync arrangement
[04:34] <captainkirk> twb: so main data store is on w2k server, using ubuntu and samba as a backup data store via sync
[04:34] <twb> Eh, I don't know too much about that.
[04:34] <twb> You could try asking #samba
[04:34] <captainkirk> kk
[04:34] <captainkirk> is that on freenode?
[04:35] <captainkirk> dont worry... found it
[05:59] <captainkirk> can anyone tell what command line to use to search for file on the server?  like a wildcard all folder type search
[06:00] <captainkirk> or maybe someone could just tell me where to find the rsync conf file
[06:01] <twb> captainkirk: the rsync client config file, or the rsync server config file?
[06:01] <twb> rsyncd is basically an FTP server, but using the rsync protocol.
[06:02] <captainkirk> twb... i am looking for the rsync server config file
[06:03] <captainkirk> is it correct that the 'server' is the computer storing the copy of the files?
[06:04] <twb> The rsync server config file is in /etc/rsyncd.conf or so, but I don't think you want the rsync server.
[06:04] <captainkirk> twb: well i have files on a w2k server that i want backed up to ubuntu via rsync (or any other method if you have a suggestion)
[06:05] <twb> That would be the rsync *client*.
[06:05] <captainkirk> so the server is the one sending the files?
[06:05] <twb> You write a cron job that runs rsync with particular options.  The "config file" is then just that cron job.
[06:05] <twb> captainkirk: it's a pull-based operation.
[06:06] <twb> rsync connects to your windows server using ssh, and copies files from it.
[06:06] <twb> Note that due to brokenness in Windows' design, you cannot copy files that are in use (e.g. the operating system).
[06:06] <captainkirk> ok, so just invoke the rsyncd with command line options scheduled via cron
[06:06] <twb> captainkirk: not rsyncd.
[06:07] <twb> rsync.
[06:07] <captainkirk> i only intend to copy data files at night.
[06:07] <captainkirk> twb: and the rsync software on the w2k server runs in server mode then?
[06:08] <twb> You may also want to look at rsnapshot, which is a wrapper for rsync intended to help you use it for incremental backups.
[06:08] <captainkirk> twb: wrapper for windows or linux?
[06:08] <twb> captainkirk: yes, but the rsync "server" on windows is only launched on demand.
[06:08] <twb> On linux, the cron job runs "rsync windows:/foo" and that in turn calls "ssh windows rsync -serve /foo"
[06:09] <captainkirk> i have found a prog called deltacopy which runs on w2k as a service
[06:28] <captainkirk> twb: Hey!! great excitement... i actually got it to work!!
[07:09] <billybigrigger> why does every openvpn howto i read seem like a pain in the arse to setup? is there no quick and dirty way to setup a vpn?
[07:10] <billybigrigger> vpn's seem very complex, maybe its just 12am and i dont have the patience right now, am i right?
[07:16] <twb> The quick and dirty way is to use ssh instead of openvpn
[07:48] <Scix> I have to change olcSizeLimit, but I'm a bit confused about where this settings is stored. When using ldapmodify -x -D cn=admin,cn=config -W, which dn am I supposed to connect to?
[07:48] <Scix> using ubuntu server 8.10
[07:49] <Scix> god morning, day and evening from noway by the way :)
[08:08] <Scix> found it. It wasn't that hard after all :)
[08:12] <sluimers> Hi, dovecot is an imap server right?
[08:12] <sluimers> I've got a server-computer running ovecot and I would like to reach my imap server it on my desktop computer using thunderbird
[08:13] <sluimers> However, I'm at a loss on how to accomplish this
[08:18] <soren> sluimers: dovecot is both an imap and pop3 server (and a bit more, but that's probably not really important for you right now).
[08:19] <soren> sluimers: If you want it to serve imap, you need to install dovecot-imapd
[08:20] <oh_noes> I just created a new filesystem
[08:20] <oh_noes> how do I find it's UUID to add an entry into fstab
[08:21] <oh_noes> found it, blkid
[08:21] <sluimers> soren, I have that already installed
[08:22] <soren> sluimers: Ok. then what exactly is the problem?
[08:22] <sluimers> I try to mail to this server and get this errror:
[08:23] <sluimers> Google tried to deliver your message, but it was rejected by the recipient domain. We recommend contacting the other email provider for further information about the cause of this error. The error that the other server returned was: 550 550 sorry, mail to that recipient is not accepted (#5.7.1) (state 14).
[08:25] <soren> sluimers: "mail to this server"? How so?
[08:33] <sluimers> server-computer
[08:33] <sluimers> sorry
[08:37] <sbeattie> sluimers: that's not a dovecot issue, that's an issue either with your mail server (likely postfix) or with your dns setup (i.e. the mx record pointing to the wrong machine)
[08:37] <sluimers> ah
[08:38] <sluimers> I'm such a newbie with this :/
[08:38] <sbeattie> sluimers: you want to look in /var/log/mail to verify that its getting to your server and see what errors are reported there.
[08:38] <sbeattie> sorry /var/log/mail.log
[08:40] <sluimers> the only thing I read is me logging in and out of mutt I think
[08:40] <sluimers> Login... Disconneced.. Login.. Disconnected
[08:43] <sbeattie> sluimers: can you connect to the mail server's smtp port? e.g. 'nc [servernameorip] smtp'
[08:46] <sluimers> Like I said, I'm very newbie, I don't know what my servernameorip is
[08:46] <sluimers> I tried 'nc localhost smtp' but connection is refused
[08:47] <sbeattie> while logged in to the server?
[08:47] <sluimers> yes
[08:48] <sluimers> I have managed to send an e-mail from my server-computer to my gmail though, using mutt
[08:48] <sluimers> and exim4
[08:51] <sluimers> wait... I should tell you a bit more. I want this to work because I want to make a contact page on my website. I bought a domain-name. The website is running on my server-computer
[08:51] <sluimers> I typed in nc <mail server from where I bought the domain name from> smtp and it's giving me:
[08:52] <sluimers> 220 rblsmtpd.local
[08:56] <sbeattie> sluimers: is the hostname returned by "host -t mx your.domain.name" the same as your server-computer?
[08:58] <sluimers> no it's the <mail server from where I bought the domain name from> hmmm... I should just mention it. it's 0 smtp.secureserver.net
[08:58] <sluimers> and 10 mailstore1.secureserver.net
[08:59] <sluimers> I typed in smtp.secureserver.net as server name in thunderbird
[09:00] <sluimers> username as <e-mail adress I get when sending to my gmail adress>
[09:00] <sluimers> secure connection as TLS, as I find that in the mail.log
[09:01] <sluimers> port 143
[09:01] <sluimers> oh and e-mail name is the same
[09:01] <sluimers> as my username
[09:07] <Knirgh> What is the command to add a user with it's home in /var/www and able to use ftp services?
[09:10] <sbeattie> sluimers: there's two elements to receiving email: (1) getting your dns set up so that other mail servers know how to find the ip address of your server and (2) configuring your mail server (exim4 in your case) to accept mail; at a minimum it needs to listen on port 25 and accept mail for your domain.
[09:13] <sbeattie> I assume secureserver.net (lightedge?) is handling dns for you, so you'll need to figure out how to get them to set up an mx record that points to an A name record for your server.
[09:14] <sbeattie> then you'll need to figure out why exim isn't listening on port 25 on your server; port 143 is imap, which is strictly for clients wishing to pull mail off of the server, not for sending.
[09:15]  * sbeattie needs to head to bed.
[10:09] <simplechat_> hey, i have a few issues
[10:09] <simplechat_> my server just stopped booting about 15 minutes ago
[10:10] <simplechat_> its having issues finding /dev/md0
[10:10] <simplechat_> it complains that it can't fsck it
[10:10] <simplechat_> but the md0 array won't build
[10:10] <simplechat_> because it can't find any devices
[10:10] <simplechat_> it says
[10:10] <simplechat_> Assembling MD array md0 failed (no devices found)
[10:10] <simplechat_> then it fails to read it
[10:22] <simplechat_> anyone?
[10:24] <sluimers> no idea, I'm here with problems myself
[10:26] <ploum> Hello
[10:26] <ploum> I'm using Dovecot and postfix
[10:26] <ploum> and I want the user+folder@email feature described here : http://blog.init.hr/?p=3
[10:26] <ploum> but I don't want to install dovecot-postfix as I already have a working configuration with LDAP
[10:27] <ploum> I've tried to read the whole file provided by dovecot-postfix but I can't find where the "email extension" (or "subaddress") is declared
[10:27] <ploum> nor how I can put this feature in my setup
[10:27] <ploum> (I suppose it's a sieve feature but I'm not sure)
[10:28] <sluimers> oh cool, maybe you can help me ploum, I'm a newbie with dovecot and use exim4. I'm trying to send mail fom my gmail account to my imap server dovecot-imapd
[10:29] <sluimers> drat
[10:29] <sluimers> wb ploum
[10:30] <ploum> sluimers: can you repeat ? Pidgin crashed
[10:30] <sluimers> ah
[10:30] <sluimers> I'm a beginner with mail servers, so I thought you might be able to help me.
[10:31] <sluimers> I don't really know how to set it up
[10:31] <sluimers> I've got dovecot installed though
[10:31] <sluimers> and dovecot-imapd
[10:32] <sluimers> and exim4 and I managed to send an e-mail to my gmail-account
[10:32] <sluimers> but not the other way aound
[10:32] <sluimers> around
[10:32] <ploum> where do you want to store your users
[10:32] <sluimers> I have no idea
[10:32] <sluimers> <- complete newbie
[10:33] <ploum> Ok, is there any reason for you to choose exim4 over postfix ?
[10:33] <sluimers> it was already running when I installed ubuntu on it
[10:34] <ploum> ok
[10:34] <ploum> so it means that any of your packages needed a mail sender
[10:34] <ploum> not a big deal
[10:34] <ploum> My advice would be to use postfix
[10:34] <ploum> install dovecot-postfix
[10:34] <ploum> (be sure to remove exim4)
[10:35] <ploum> and see if it works
[10:35] <ploum> in theory, it should work out-of-the-box
[10:37] <ploum> If later you want to switch to LDAP, I advise you to keep the following howto somewhere :
[10:37] <ploum> http://www.vogelweith.com/debian_server/07_postfix.php
[10:37] <ploum> it's really really good
[10:37] <ploum> but more avdanced
[10:37] <ploum> try to get basic stuffs working first
[10:37] <ploum> (sorry, I just realized it was in french)
[10:38] <sluimers> lol, yeah, my french isn't that good
[10:38] <sluimers> bonjour ploum, ca va? :P
[10:39] <sluimers> <- all I remember from french lessons
[10:39] <ploum> bien merci :-D
[10:39] <ploum> I'm sorry, as I'm always switching from french to english, I don't always know the language of something I've just finished reading
[10:40] <sluimers> je suis un hollandaise
[10:41] <sluimers> errmm... is there a package called dovecot-postfix or did you mean to install them both?
[10:44] <sluimers> hmmmm... wait a second, shouldn't exim/mutt be enough?
[10:46] <ploum> (Ik ben sorry maar mijn Nederlands slecht is. Ik moet het meer gebruiken)
[10:46] <sluimers> (want?)
[10:46] <ploum> There's a package called dovecot-postfix (on Jaunty)
[10:47] <ploum> mutt is a mail client
[10:47] <ploum> just like evolution or thunderbird
[10:47] <sluimers> Ah, I still use intiprid
[10:47] <sluimers> Jaunty is still in betaphase right?
[10:47] <ploum> no
[10:47] <sluimers> ah
[10:47] <ploum> it was released last week
[10:48] <ploum> and one of the big improvement for mail servers is this new packages "dovecot-postfix" which enable a working configuration "out-of-the-box"
[10:49] <marthe> PLEASE !!! can anyone help me with my postfix-procmail? How to configure UNOCONV with a mail server ?
[11:32] <twb> marthe: if you get no help here, try #postfix
[11:34] <marthe> i have try it, but it's not explain how to configure UNOCONV
[11:35] <marthe> twb:  i have try it, but it's not explain how to configure UNOCONV
[11:37] <marthe> twb: thanks
[12:20] <dayo> i'm trying to set up ldap authentication, so clients can access their home dirs on the nfs server. i'm following this guide https://help.ubuntu.com/8.10/serverguide/C/openldap-server.html and when i tried to run `sudo ldapadduser...` i got this error: http://paste.ubuntu.com/161360/   can someone help me with this?
[12:25] <simplechat_> dayo, did you install ldap?
[12:26] <dayo> on the client? no
[12:27] <simplechat_> that could be the issue
[12:27] <dayo> i need to install ldap on the clients??
[12:27] <dayo> but it's already installed on the ldap server
[12:27] <dayo> :-/
[12:28] <dayo> or maybe i should be running `ldapadduser` on the server...
[12:31] <dayo> says command not found
[12:42] <simplechat_> hmmmm
[13:02] <TarBar> I'm trying to install ubuntu server, I select my NIC from the list and it does the DHCP autoconfig but fails it, My NIC has a cat5 cable going to router and the lights on the nic show that its connected. any ideas?
[13:06] <ploum> can someone help me to have address extension on my dovecot/postfix setup ?
[13:07] <ploum> postfix is refusing user+folder adresses telling me that the use doesn't exist
[13:07] <ploum> (which is true, only "user" exists)
[13:07] <lamont> ploum: set recipient_delimiter = + in main.cf
[13:07] <ploum> lamont: it's already done
[13:08] <ploum> that's why I don't understand
[13:09] <ploum> but I'm using ldap
[13:09] <ploum> and postfix-ldap
[13:11] <lamont> ploum: it could be that it's rejecting the address in one of the early user maps
[13:11] <lamont> local_recipient_maps maybe
[13:11] <lamont> not sure without diving into the rathole
[13:13] <TarBar> Can anyone help me get networking setup in ubuntu server?
[13:13] <ploum> lamont: it could be, indeed
[13:14] <ploum> do you know how I can test that ?
[13:15] <lamont> ploum: a goodly scattering of -v options in master.cf should do the trick, once  you wade through the possibly thousands of lines of logging
[13:17] <ploum> I don't understand the following lines :
[13:17] <ploum> 127.0.0.1:10025 inet n  -       -     -       -  smtpd
[13:17] <ploum>     -o content_filter=
[13:17] <ploum>     -o local_recipient_maps=
[13:17] <ploum>     -o relay_recipient_maps=
[13:17] <ploum> (in master.cf)
[13:22] <TarBar> Right, I've tried a gui version of ubuntu and the internet works, however it fails setting up on the ubuntu server install?
[13:23] <ploum> TarBar: wifi or cable ?
[13:24] <VK7HSE> ploum:  127.0.0.1:10025 inet n  -       -     -       -  smtpd  This is setting postfix up to listen on localhost on port 10025 for mail re-injection once it has passed through external filtering... like Amavis, Spamassassin, clamav etc...
[13:24] <TarBar> cable mate.
[13:24] <ploum> lamont: Final-Recipient: rfc822; ploum+bugzilla@ploum.net <https://mail.fritalk.com/src/compose.php?send_to=ploum%2Bbugzilla%40ploum.net>   so it looks like it doesn't even try to use the recipient_delimiter
[13:24] <ploum> I don't understand why
[13:24] <ploum> TarBar: without GUI, you don't have NetworkManager
[13:25] <TarBar> mmm
[13:25] <VK7HSE> ploum: there are many ways to configure postfix... ;)
[13:25] <ploum> you have to configure /etc/network/interfaces and do /etc/init.d/networking restart
[13:26] <ploum> VK7HSE: a bit too much, indeed
[13:26] <ploum> But if I put a recipient delimiter, it should work
[13:27] <VK7HSE> ploum: It also depends on just what & how your want it all to work!...
[13:28] <ploum> VK7HSE: I really think that 99% of people want to have mails but it's a tradition in the software world to be flexible that even the basic needs are nearly impossible to understand
[13:29] <ploum> that's why I like the principle of dovecot-postfix
[13:29] <ploum> It makes the most frequent typical really easy
[13:29] <ploum> (the most typical usecase)
[13:29] <ploum> I admit that here, it's not a basic need
[13:29] <ploum> just a nice feature
[13:30] <VK7HSE> Yeah I had to wade through quite a bit of documentation to get my head around it! but I'm by no means an "expert" in this field!
[13:30] <ploum> me neither
[13:31] <VK7HSE> like I haven't setup my postfix to use mysql etc.. just for local accounts only...
[13:31] <ploum> but here, I'm really frustrated. It's a feature that should be, on the paper, really easy to have with dovecot-postfix but I cannot make it works without this package
[13:31] <ploum> VK7HSE: I use it with LDAP and it was *not* nice to set up ;-)
[13:32] <VK7HSE> as I have just upgraded from 8.10 to 9.04, and haven't done a fresh install, does the dovecot-postfix duo include amavis ??? or any filtering by default?
[13:33]  * VK7HSE just topping up the teacup!
[13:33] <ploum> not that I'm aware of
[13:34] <ScottK> VK7HSE: No.  We're looking at that for the future.
[13:35] <kwork> dont take the fun out of it by making it easy :P
[13:35] <ScottK> There is a postfix-filter-add (IIRC) script shipped in the postfix package that should make integrating amavisd-new quite easy.
[13:35] <VK7HSE> Ok! cool thanks ScottK  ... I had fun sorting mine out!.. :P
[13:36] <ploum> I've the following postfix maps with ldap : http://www.vogelweith.com/debian_server/07_postfix.php#x1-180003.3
[13:36] <VK7HSE> ScottK I have mine working just fine... but I will compare the new info once its complete...
[13:37] <ploum> maybe the /etc/postfix/ldap-aliases.cf in the link refuse the user+folder@domain ?
[13:37] <ScottK> I had it backwards.  It's /usr/sbin/postfix-add-filter
[13:38] <ScottK> If you just execute /usr/sbin/postfix-add-filter it will explain what it does
[13:39] <VK7HSE> ScottK hey that pretty neat!... you guys have been hard at it!... :D
[13:45] <ScottK> That's actually been there in Intrepid too, but didn't get the press that dovecot-postfix is getting.
[13:46] <VK7HSE> ScottK... gee that would have saved me many hours of "learning" !!! but hey I needed to understand just what it was I was attempting to do... not just cut'n'paste the whole config!
[13:47] <ploum> (damm, it looks like it's a old problem : http://www.irbs.net/internet/postfix/0611/0115.html )
[13:48] <ivoks> ah, ploum :)
[13:49] <VK7HSE> I now "understand" why I left ldap alone !!! :P
[13:50] <ivoks> i don't see why ldap/passd/sql would make any difference with delimiter
[13:50] <ivoks> if postfix knows what the delimiter is, then it knows what is the username
[13:51] <ivoks> and will ask ldap/sql/passwd for a correct username
[13:51] <VK7HSE> for me, I just didn't understand it! plus for a home server it was just overkill !!!
[13:52] <ivoks> easy with '!'
[13:52] <ploum> ivoks: I don't understand it more than you
[13:52] <ploum> what you say is logical
[13:52] <ploum> but I think I've tried all logical solutions
[13:52] <ivoks> ploum: it doesn't work for you?
[13:53] <ploum> so I'm trying the non-sense approach
[13:53] <ploum> ivoks: no
[13:53] <ivoks> ploum: do you use dovecot as MDA?
[13:53] <ploum> yes
[13:53] <ploum> with sieve and everything working fine
[13:53] <ivoks> you configured both dovecot and postfix to do that, right?
[13:53] <ivoks> ok
[13:53] <ploum> ivoks: how do you configure dovecot for that ?
[13:54] <ivoks> ploum: you have to enable mda in its configuration
[13:54] <ploum> this is just that **** recipient_delimiter which is not taken into account
[13:54] <ploum> ivoks: yes
[13:54] <ivoks> (protocol lda)
[13:54] <ploum> my config is working for nearly one year
[13:54] <ivoks> ok then
[13:54] <ivoks> how does mailbox_command looks like?
[13:54] <ploum> empty
[13:55] <ivoks> so, let's start again
[13:55] <ivoks> ploum: do you use dovecot as MDA?
[13:55] <ivoks> correct answer on this question is: no
[13:55] <ivoks> unless you made serious changes to master.cf
[13:56] <ploum> virtual_transport = dovecot
[13:56] <ploum> master.cf :
[13:56] <ploum> dovecot    unix  -       n       n       -       -       pipe
[13:56] <ploum>   flags=DRhu user=vmail:mail argv=/usr/lib/dovecot/deliver -d $recipient
[13:56] <ploum> so I really believe that I'm using dovecot as MDA
[13:56] <ivoks> you do, but you didn't set it up to look at extensions
[13:57] <ploum> really interesting
[13:57] <ploum> How can I do that
[13:57] <ploum> ?
[13:57] <ivoks> http://wiki.dovecot.org/LDA/Postfix
[13:57] <ivoks>   flags=DRhu user=vmail:vmail argv=/usr/local/libexec/dovecot/deliver -f ${sender} -d ${user}@${nexthop} -n -m ${extension}
[13:57] <ivoks> notice the -n -m ${extension}
[13:58] <ivoks>  /usr/lib/dovecot/deliver --help
[13:58] <ivoks> doh, no --help
[13:58] <ivoks> http://wiki.dovecot.org/LDA
[13:58] <ivoks> -m <mailbox>: Destination mailbox (default is INBOX). If the mailbox doesn't exist, it's created (unless -n is used). If message couldn't be saved to the mailbox for any reason, it's delivered to INBOX instead.
[13:59] <ivoks> so, you aren't exporting mailbox name to dovecot's MDA
[13:59] <ivoks> that's why it can't deliver in specified mailbox
[14:00] <ploum> ivoks: but there's another problem
[14:00] <ploum> It doesn't even try
[14:00] <ploum> <ploum+bugzilla@ploum.net <https://mail.fritalk.com/src/compose.php?send_to=ploum%2Bbugzilla%40ploum.net>>: user unknown
[14:00] <ivoks> of course it doesn't; you didn't tell it to try
[14:01] <ivoks> have you added -n -m ${extension} ?
[14:01] <ploum> yes
[14:01] <ploum> and reloaded postfix
[14:01] <ploum> and it's still the same
[14:01] <ivoks> so, how does that line in master.cf looks like now?
[14:03] <ploum> flags=DRhu user=vmail:mail argv=/usr/lib/dovecot/deliver -d $recipient -n -m ${extension}
[14:03] <ploum> doesn't work (user unknonw)
[14:03] <ploum> flags=DRhu user=vmail:mail argv=/usr/lib/dovecot/deliver -d ${@user}@{nexthop} -n -m ${extension}  :
[14:03] <ploum> mails are not coming
[14:06] <ploum> status=deferred (mail system configuration error)
[14:07] <ploum> I'm sure there's something I don't understand
[14:07] <ivoks> sorry, phone
[14:07] <ploum> (I always said that master.cf is the limit of my intelligence)
[14:07] <ploum> no problem ! You are helping me ! Thanks :-)
[14:08] <ivoks> so... correct line would be:
[14:09] <ivoks>   flags=DRhu user=vmail:vmail argv=/usr/lib/dovecot/deliver -d ${recipient} -n -m ${extension}
[14:10] <ivoks> save it, restart postfix
[14:10] <ivoks> send your self a mail on email+existing_folder@domain.com
[14:10] <ploum>  user unknown
[14:11] <ploum> I've a "bugzilla" folder
[14:11] <ploum> my mail is ploum@ploum.net
[14:11] <ploum> so I send to ploum+bugzilla@ploum.net
[14:11] <ploum> user unknown
[14:11] <ivoks> ok
[14:11] <ivoks> postconf recipient_delimiter
[14:11] <ivoks> ?
[14:12] <ploum> recipient_delimiter = +
[14:12] <ivoks> and how about ldap filter?
[14:12] <ivoks> have you looked at ldap to see which user postfix is requesting
[14:13] <ploum> virtual_mailbox_maps = ldap:/etc/postfix/ldap-accounts.cf
[14:13] <ploum> virtual_alias_maps = ldap:/etc/postfix/ldap-aliases.cf, ldap:/etc/postfix/ldap-maildrop.cf
[14:13] <ploum> the .cf files are here : http://www.vogelweith.com/debian_server/07_postfix.php#x1-180003.3
[14:13] <ploum> basically :
[14:14] <ploum> query_filter = (&(objectClass=mailAccount)(mailalias=%s)(mailenable=OK))  result_attribute = mail
[14:15] <ploum> so, of course, ploum+bugzilla will not match any ldap alias
[14:15] <ploum> but that's the goal of the stuff
[14:15] <ploum> Postfix should only ask for "ploum@ploum.net"
[14:16] <ploum> http://www.irbs.net/internet/postfix/0611/0189.html
[14:18] <ivoks> er...
[14:19] <ivoks> what's the username, ploum or ploum@ploum.net?
[14:19] <ivoks> bah, never mind
[14:19] <ploum> that's maybe the problem. The username is "ploum@ploum.net"
[14:19] <ploum> so maybe it looks for "ploum" only
[14:20] <ivoks> probably
[14:20] <ivoks> try mailalis=%s@ploum.net
[14:20] <ivoks> mailalias
[14:20] <ploum> but I'm hosting several domains so it's not a possible solution
[14:21] <ivoks> propagate_unmatched_extensions = canonical, virtual, alias, forward, include
[14:21] <ivoks> er...
[14:22] <ivoks> make that only canonical, virtual
[14:24] <ploum> mmm
[14:24] <ploum> no
[14:24] <ploum> it doesn't help
[14:24] <ploum> But anyway, thanks for taking the time to help me :-)
[14:24] <ivoks> hm...
[14:25] <ivoks> we have a error in our logic :)
[14:25] <ivoks> result_attribute = mail
[14:25] <ploum> ?
[14:25] <ploum> I should put that in main.cf ?
[14:25] <ivoks> what's mail attribute?
[14:25] <ivoks> an alias or mailbox?
[14:26] <ivoks> i'll have to investigate this
[14:27] <ploum> an alias
[14:27] <ivoks> i'm planing on having ldap as a backend for karmic, so i'll have to tackle this
[14:27] <ivoks> ?
[14:27] <ploum> wait
[14:27] <ploum> ivoks, are you ante ?
[14:27] <ivoks> yes
[14:27] <ploum> oooh !
[14:28] <ploum> thanks for your time *and* your mail :-)
[14:28] <ivoks> hehe np
[14:28] <ploum> I've tried to write a spec about that : https://wiki.ubuntu.com/LdapIntegrationSpec
[14:29] <ploum> I'm using an Ubuntu server with LDAP for mails, Jabber and webservices : it works great but was really hard to set up
[14:30] <ploum> ivoks : If you work on LDAP and you want to see some specific bit of my configuration, don't hesitate to drop me a mail
[14:35] <ploum> ivoks: thanks a lot for your time and everything else (your blog, your work)
[14:35] <ploum> I will try it again another time
[14:54] <Hungry> hello there.
[14:55] <Hungry> Think I may have got myself into a bind.
[14:55] <Hungry> Is there a way to get hold of the update-manager-core and other packages required to upgrade a server from gutsy to hardy? Or now the repository is gone is the only option to reinstall from cd (with a train journey to visit the server)?
[14:58] <ScottK> Hungry: old-releases.ubuntu.com (or something similar) has it.
[14:58] <Hungry> will check that out. thanks.
[15:13] <Sam-I-Am> wow, nfs4 infinitely fails on 8.04 lts...
[15:13] <Sam-I-Am> foxbuntu: figured out the problem... already a bug reported on it.
[15:13] <Sam-I-Am> foxbuntu: reported last july, still not fixed in LTS, but fixed in intrepid.
[15:20] <Sam-I-Am> actually, using nfs4, LTS is totally unusable
[15:20] <Sam-I-Am> nice to see :)
[15:23] <Faust-C> i thought cyrus was being taken out of repos?
[15:23] <Faust-C> or was that courier
[15:23] <Sam-I-Am> i'd hope cyrus isnt leaving...
[15:43] <foxbuntu> Sam-I-Am, Im not really sure at this point what I helped you with, but if there is a fix in intrepid for it, you could submit a backport request on it
[15:47] <Sam-I-Am> would they consider backporting a kernel?
[15:47] <Sam-I-Am> foxbuntu: we were talking about the kernel panic
[15:47] <foxbuntu> Sam-I-Am, no, but you might get them to backport the fix
[15:47] <foxbuntu> Sam-I-Am, but it would need regression testing and such
[15:48] <foxbuntu> Sam-I-Am, what is the panic related to?
[15:48] <Sam-I-Am> i set up a serial console on the VM and traced it to an NFSv4 problem... which is documented as https://bugs.launchpad.net/debian/+source/linux/+bug/253004
[15:48] <Sam-I-Am> a recent post indicates that they're considering fixing it in hardy
[15:48] <Sam-I-Am> but nothing official
[15:49] <Sam-I-Am> i could just toss the intrepid/jaunty kernel on hardy but that sorta defeats the purpose of running LTS
[15:51] <foxbuntu> Sam-I-Am, that bug report shows its been nominated for backport already
[15:51] <Sam-I-Am> yeah
[15:51] <Sam-I-Am> thats what i was thinking
[15:52] <Sam-I-Am> except its been a looooooong time since it was reported
[15:52] <foxbuntu> Sam-I-Am, so its likely just a matter of time till it shows up in one of the lts kernels
[15:52] <Sam-I-Am> considering it makes lts unusable if you run nfs4
[15:52] <foxbuntu> yea
[15:52] <ScottK> Bug nomination doesn't mean more than someone thought it would be an appropriate thing to fix post release if there was a fix.  It doesn't mean someone is actively working on it.
[15:52] <jmedina> nfs is importan in today virtualization systems,
[15:54] <foxbuntu> Sam-I-Am, my only suggestion is to try to bump it and get the fix backported
[15:55] <foxbuntu> find out what was actually fixed
[15:58] <Sam-I-Am> foxbuntu: how would i 'bump' it?
[16:02] <foxbuntu> Sam-I-Am, on the kernel, I am not totally sure, you might want to go ask someone on the kernel team for advice. #ubuntu-kernel
[16:02] <Sam-I-Am> not a bad plan...
[16:18] <AnRkey> can i create an lpr queue for our as400 to print to with cupsys on ubuntu?
[16:19] <AnRkey> if so how?
[16:33] <Hecate> AnRkey, i dont know exactely what your as400 is, but it sounds pretty much like a printer and since that would disqualify the question from being asked in this channel, i'd recommend asking it somewhere else.
[16:34] <AnRkey> Hecate, as400's are a kind of server
[16:34] <LHC> hey all
[16:34] <AnRkey> the as400 server must print send a print job to a server here running ubuntu desktop for our noobs to work on
[16:35] <AnRkey> Hecate, thanks anyway :)
[16:35] <Hecate> oh, sorry for "showing you off", then. ;)
[16:37] <soren> AnRkey: Doesn't cups-bsd also run an lpd?
[16:37] <soren> AnRkey: It certainly provides the lpr-like cli tools, but I think it also runs a corresponding daemon.
[16:38] <AnRkey> soren, now we are talking
[16:38] <jmedina> yeap cups supports lpd
[16:38] <AnRkey> i want to simply set up a queue
[16:38] <AnRkey> jmedina, how do i create a queue then?
[16:38] <AnRkey> jmedina, save my life please!!!! :P
[16:39] <soren> I would expect cups-lpd to expose a queue for each of cups' configured printers.
[16:39] <soren> http://www.cups.org/doc-1.1/sam.html#8_2
[16:40] <jmedina> use cups-lpd daemon
[16:40] <jmedina> http://www.cups.org/documentation.php/doc-1.4/man-cups-lpd.html
[16:40] <jmedina> you can launch it from xinetd
[16:40] <jmedina> so you point your as400 lpr cliento to cups
[16:42] <AnRkey> jmedina, where do i send the beer to?
[16:43] <jmedina> mexico city plase
[16:43] <jmedina> I dont have incluensa
[16:43] <jmedina> XD
[16:43] <AnRkey> oh geez, ok a bit far
[16:43]  * AnRkey gives jmedina a noddy badge instead
[16:43] <AnRkey> :D
[16:44] <AnRkey> ok that was about as korny as i should be today
[16:44] <AnRkey> bbl
[16:47] <AnRkey> jmedina, now i just need to get this company to switch to ubuntu server from that blasted as400 server
[16:50] <jmedina> AnRkey: wahat is as400 function?
[16:50] <jmedina> only print server?
[16:50] <jmedina> file sharing?
[16:50] <jmedina> XD
[16:58] <AnRkey> no it's a server that runs there accounting stuff
[16:58] <AnRkey> it's terminal based
[16:58] <AnRkey> text
[16:59] <AnRkey> runs over a telnet type session
[16:59] <AnRkey> they are getting ripped off
[16:59] <AnRkey> i have already saved them a boatload of cash with ubuntu desktop deployments
[16:59] <AnRkey> we just got a call that they are considering ubuntu server and getting in a new package
[18:02] <stas> hi, why after editing motd.tail and running bootmisc.sh my motd doesn't change? have you modified something again?
[18:05] <genii> !info update-motd
[18:34] <jbernard> stas: that should work, is /etc/motd a symlink to /var/run/motd?
[18:34] <stas> jbernard: yes it is
[18:35] <stas> I mean it's a symlink to /var/run/motd
[18:35] <stas> but it doesn't change on running /etc/init.d/bootmisc.sh start
[18:36] <stas> genii: it worked, thx
[18:37] <jbernard> stas: ahh, the updating of motd is no longer done by bootmisc.sh for jaunty
[18:37] <jbernard> that was removed, probably in favor of installing update-motd
[18:38] <stas> jbernard: yeah :)
[18:38] <jbernard> i think i was the last one to actually realize that ;)
[18:39] <stas> I just wanted to know what changed :)
[18:52] <theuser1>  what are the ports used by irc. ?
[18:58] <infinity> theuser1: 6667 is the most common one, but IRC networks tend to run on a range of ports, to allow people to get out of permissive firewalls.
[18:59] <theuser1> need all ports used. i have to run ircd
[18:59] <infinity> theuser1: It's entirely up to how you configure your ircd.
[18:59] <infinity> theuser1: But, like I said, 6667 is the most common default.
[19:00] <theuser1>  6665, 6666, 6667, 7000, 7070, 8000, 8001, 8002 ?
[19:41] <ichat> what whould be the  best way to (remote or local)  control  a ubuntu server instance - for home use -  ?
[19:42] <jmedina> ssh
[19:43] <ichat> sorry - i should have said other than ssh - cause im not verry fond of  cli command al the time
[19:43] <ichat> i googled for webmin instead but its said to have issues with ubuntu
[19:45] <kirkland> mathiaz: around?
[19:46] <kirkland> mathiaz: regarding bug #360825 ...
[20:21] <mathiaz> kirkland: yes - bug 360825
[20:22] <kirkland> mathiaz: so i've been talking with upstream about this issue, trying to get him to take our patch
[20:22] <kirkland> mathiaz: but he says that it's wrong
[20:22] <kirkland> mathiaz: wrong to use index= anything, with virtio
[20:22] <kirkland> mathiaz: because with virtio, the indexing is misleading
[20:23] <kirkland> mathiaz: it doesn't actually correspond to the order as seen by the guest
[20:23] <mathiaz> kirkland: hm ok.
[20:23] <kirkland> mathiaz: and uuid should be used instead
[20:24] <kirkland> mathiaz: my question for you ...  where did those index= entries come from?  libvirt?
[20:24] <mathiaz> kirkland: right - so if I define a virtio devices as being hda in the libvirt file it doesn't mean that it will be seen as hda in the guest
[20:24] <kirkland> mathiaz: well, vda you mean
[20:24] <kirkland> mathiaz: but, yes, exactly
[20:24] <mathiaz> kirkland: yes - in my libvirt definition, I have device=vda, device=vdb
[20:25] <mathiaz> kirkland: the libvirt definition goes like this: <target dev='hda' bus='virtio'/>
[20:26] <mathiaz> kirkland: then I guess that libvirt translates that to index option when it launches the kvm process
[20:27] <kirkland> mathiaz: gotcha
[20:27] <kirkland> mathiaz: okay, we'll need to fix that in libvirt
[21:08] <jimcooncat> anyone here use freenac?
[21:08] <jmedina> it is totally free?
[21:10] <jimcooncat> not if you use their consulting service :-)
[21:10] <jmedina> have you tried http://netreg.sourceforge.net/
[21:12] <jimcooncat> not sure that fits my scenario -- want to provide guest access in our conference room. I want them to see only a proxied gateway, not the other machines on the LAN
[21:13] <jimcooncat> ... wireless access (I've got a couple WRT54GL's handy) would be a bonus
[21:15] <jmedina> I use WPA2 Enterprise with FreeRADIUS
[21:15] <jimcooncat> I thought originally I'd make very small subnets and hand each guest their own via dhcp
[21:15] <jmedina> and FreeRADIUS with EAP-PAP or EAP-TTLS
[21:15] <jmedina> usring windows autentication or ldap auth
[21:16] <jmedina> you can add a user to a specific vlan depending of their authtentication credentials
[21:16] <jimcooncat> jmedina: does that require anything for the client to configure or install?
[21:16] <jmedina> it is much simple to setup a captive portal server :D
[21:16] <jmedina> like in hotels
[21:16] <jmedina> your user connect to a open wireless
[21:17] <jmedina> assigned a IP
[21:17] <jmedina> when taht user opens a browser it will apear a registration form
[21:17] <jmedina> so you can auth that user and give privileges
[21:17] <jmedina> I like ZeroShell distribution for that
[21:17] <jmedina> it has ldap and kerberos auth integrated
[21:17] <jmedina> it is easier
[21:18] <jimcooncat> thanks jmedina, I'll look into that -- it does sound simpler
[21:21] <jimcooncat> looks very feature-rich, but not simple to update. Shouldn't need much updating though
[21:22] <jimcooncat> I'm so spoiled with apt
[23:33] <captainkirk> does anyone know where i should look to find the rsync logs?
[23:36] <andol> captainkirk: from rsyncd? Well, unless otherwise specified it should go into syslog.
[23:38] <captainkirk> andol... just looking now
[23:40] <captainkirk> andol. details i am looking for are not in the system log, but I did find something else interesting.....
[23:40] <andol> You always find something interesting when you peak at your syslogs :)
[23:40] <captainkirk> andol: i have changed the hostname of the computer from emerlad2 to emerald but all the log entries are still showing the old hostname of emerald2?
[23:46] <captainkirk> andol: ok, i found a 1 line entry in the syslog of where the cron job has run the rsync command.  Is there some way of piping the rsync output to a file somewhere and then have a cron to email it to me as an attachment?
[23:50] <captainkirk> anyone ?  ^^^^^^^
[23:50] <andol> captainkirk: Since you were talking about logs I figured you were running a rsyncd, but obviously not.
[23:51] <andol> captainkirk: if you have a MTA set up properly all cron output will be sent as mail, to the user running the job.
[23:51] <captainkirk> andol: rsyncd is the server side of rsync right?
[23:51] <andol> captainkirk: If you want more rsync output, use the --verbose flag
[23:52] <captainkirk> andol: yes, i am using the --verbose flag and if i run the rsync command from cli, i get 'los' of info... :)
[23:52] <andol> captainkirk: Yes, but not neccesary the one you use. Often today you simply pipe rsync through a ssh tunnel, spawning a rsync session on the other wise.
[23:53] <captainkirk> andol: ok, my problem is the other host with the 'live data' to copy is a w2k server
[23:53] <andol> captainkirk: Another option, unless you have your mail working, is to simply add a ">> /path/to/logfile" after the command in your cron file.
[23:56] <captainkirk> andol, as this machine is not my site mta, it is not configured so i like the >>etc option just to capture the output of the most recent run event
[23:57] <captainkirk> andol: is the ssh tunnel option difficult to configure, or even necessary if both machines are in the same room?
[23:59] <andol> captainkirk: Actually, the ssh tunnel one is usually the easiest one, it only requiers the other server to be running an openssh-server and have the rsync client install