[00:06] <RoAkSoAx> kirklandi here waiting for shuttle to take me to airport
[00:06] <RoAkSoAx> i mean hotel
[00:11] <pHro> hello
[00:23] <panfist> is there any info on best practices regarding system time?
[00:23] <panfist> i vaguely remember reading about how it should be set to gmt in BIOS and in the os stored as an offset from that or something...?
[00:31] <kirkland> RoAkSoAx: cool
[00:31] <kirkland> RoAkSoAx: ping me when you're at the hotel
[00:31] <RoAkSoAx> ykirkland will do im still on my way
[00:34] <kirkland> zul: i can't build cobbler more than one
[00:35] <kirkland> zul: the build leaves a bunch of garbage around
[00:51] <kirkland> zul: and are we still using lp:~ubuntu-virt/cobbler/ubuntu ?
[00:51] <kirkland> zul: it looks out of date
[01:00] <Jasonn> Ok, well I ALMOST got PPTP VPN service to work on my server, just i cant connect to it
[01:03] <jeeves_moss> how do I set up a local DNS server behind a NAT?
[01:03] <kirkland> zul: can you get that branch fixed so that I can work on a couple of bugs?
[01:10] <pmatulis> Jasonn: PPTP?  are you serious?
[01:11] <jeeves_moss> how do I set up a local DNS server behind a NAT?
[01:19] <Jasonn> pmatulis: Yeah
[01:20] <Jasonn> pmatulis: Why?
[01:29] <e_t_> jeeves_moss: there are two options for DNS that I know about. One is Bind9, essentially THE standard of DNS. The other is dnsmasq. In my opinion, dnsmasq is easier to configure. Once you have your DNS server running, you just have to point your client machines to use it. This can be done either manually or through DHCP.
[01:44] <jeeves_moss> e_t_, can you point me to a simple "how to" URL
[01:46] <e_t_> jeeves_moss: There's this for Bind: http://news.softpedia.com/news/How-to-Host-Your-Own-Domain-With-Bind9-on-Ubuntu-49585.shtml
[01:47] <e_t_> jeeves_moss: For dnsmasq, just install it and read the /etc/dnsmasq.conf file. It has extensive explanation and instructions in the comments.
[01:50] <aliverius> i get a lot of hdd usage on my generally idle server
[01:50] <aliverius> how can i see what processes use it?
[01:52] <e_t_> There is a program called iotop that shows (more or less) which processes are writing to the disk and how much.
[02:16] <MetaJake> something keeps changing my /etc/resolve.conf to a different nameserver then the one I save it as...  any idea what might be causing this?
[02:27] <Jasonn> What is something useful i can throw on my server>
[02:27] <Jasonn> ??
[02:27] <e_t_> MetaJake: check to see if you have the package "resolvconf" installed. NetworkManager also makes changes resolv.conf.
[02:27] <MetaJake> e_t_ I see thank you
[02:28] <e_t_> Jasonn: what do you want your server to do?
[02:29] <Jasonn> anything useful
[02:29] <Jasonn> Not: seedbox, apache, proxy, vpn
[02:30] <e_t_> You could make it a virtual machine host with kvm.
[02:30] <Jasonn> only has 10GB dick space
[02:30] <Jasonn> disk**
[02:30] <Jasonn> lol;
[02:30] <e_t_> CUPS would make it a print server.
[02:31] <Jasonn> its a VPS, so dont really have a printer
[02:31] <e_t_> If it's got more than one Ethernet card, you could turn it into a router.
[02:31] <Jasonn> Hmmm
[02:31] <Jasonn> It doesnt
[02:31] <Jasonn> lol
[02:31] <Jasonn> Its a server outside of my network, in a diff country
[02:32] <Jasonn> Anyways, Im thinking of something awesome, along those lines
[02:32] <Jasonn> like
[02:32] <Jasonn> :(
[02:33] <e_t_> If you play World of Warcraft, you could set up your own server with bnetd.
[02:34] <Jasonn> Hmm
[02:34] <Jasonn> dont play
[02:34] <Jasonn> it
[02:34] <Jasonn> D:
[02:34] <Jasonn> something to do with networking
[02:35] <e_t_> After eliminating Apache, proxy, and VPN, there's not much left.
[02:35] <e_t_> Do you want to try an email server?
[02:35] <Jasonn> Yesh
[02:35] <e_t_> Or IRC?
[02:35] <Jasonn> Could i host a server for freenode
[02:36] <Jasonn> with only 256mb ram?
[02:37] <e_t_> That's probably below what they'd like. They also want at least 1.5Mbit up and down connection speed. You could ask them.
[02:37] <Jasonn> nope
[02:37] <e_t_> Is this a personal project, or is it work-related?
[02:37] <Jasonn> personal
[02:37] <Jasonn> just to gain knowledge and stuff
[02:38] <e_t_> You're paying for a VPS "just to gain knowledge?" That sounds expensive. I just have an old laptop with a busted screen for my server.
[02:38] <Jasonn> Its $5/mo
[02:38] <Jasonn> really not much
[02:38] <Jasonn> plus 20% off
[02:39] <e_t_> Mine is $0/month.
[02:39] <Jasonn> I have a server at home too
[02:39] <Jasonn> Its just i run a website too
[02:39] <Jasonn> So i want it for that
[02:39] <Jasonn> and actually, your probably costs you more in electricity bills than mine does in total u.u.
[02:40] <Jasonn> plus, i dont have a fast connection at home
[02:41] <e_t_> An email server is really easy to set up (if you already have a domain). In Ubuntu, the default mail setup is Postfix for SMTP and Dovecot for IMAP/POP3. You can also use Exim4. Either one will prompt you for information during installation and you should be up and running quickly.
[02:43] <Jasonn> Hmmm
[02:43] <MetaJake> e_t_ sorry to wander away there. I'm just testing elementary functions of apache.. httpd, php mysql. eventually I will be using python and its' webdev frameworks. but...
[02:43] <Jasonn> But then the reverse DNS of the server would have to be the one of the email?
[02:43] <Jasonn> brb
[02:44] <MetaJake> i made some changed to some apache config files and now /etc/resolve.conf keeps changing on me like i said and I can't get a connection to my internet at all.
[02:44] <e_t_> Jasonn: Actually getting your mail accepted by other servers (Gmail, Yahoo, Hotmail) is a different matter.
[02:45] <e_t_> MetaJake: resolve.conf only affects DNS resolution. Try "ping -c 4 72.14.213.105" (google.com). If that fails, then DNS is the least of your problems.
[02:46] <MetaJake> ok. actually I just got my connection back (tested using apt-get update)
[02:47] <MetaJake> ... I think  I'm back in business.. I had made some changes to the configuration of my virtual machine that I forgot about.
[02:47] <MetaJake> been a spaghetti sort of day.. in all its learning
[02:47] <e_t_> OK. What you could do is make resolv.conf how you want, then "sudo chmod 444 /etc/resolv.conf" to prevent even root from writing to the file afterwords.
[02:49] <MetaJake> ah, i gotcha. Nice tip, I'm writing that down for next time, similar situation
[03:04] <Jasonn> e_t_: THen how do i get the mail accepted?
[03:06] <e_t_> I don't know. I have my server set to use Gmail as a smarthost. Gmail then trusts my server because I gave it my login credentials, and everybody else just gets a message from my Gmail account.
[03:08] <e_t_> Jasonn: http://www.techrepublic.com/article/get-it-done-host-your-own-e-mail-server/5032914
[03:08] <e_t_> It's not a how-to, but it's got useful information.
[03:10] <e_t_> Also, http://serverfault.com/questions/9594/cannot-send-email-to-hotmail-or-msn-addresses
[03:16] <Jasonn> e_t_: How do i install it?
[03:16] <e_t_> Which?
[03:17] <Jasonn> mail server
[03:17] <e_t_> Do you want Postfix or Exim4?
[03:18] <Jasonn> difference?
[03:20] <Jasonn> e_t_: ^
[03:20] <Jasonn> difference?
[03:20] <e_t_> Exim4 offers (I think) a more detailed installation. Postfix is Ubuntu's default mail server; Exim4 is Debian's.
[03:21] <Jasonn> Ill take exim
[03:21] <Jasonn> just apt-get install exim4 ??
[03:21] <e_t_> I think so.
[03:23] <Jasonn> dont work
[03:24] <Jasonn> It gives me an error, but i cant copy and paste because putty doesnt let me
[03:28] <e_t_> You're in Windows? take a screenshot of the putty window.
[03:31] <shauno> putty lets you; drag to select & it goes straight into the clipboard  (it's more difficult to tell it to _not_ copy)
[03:34] <Jasonn> Can i get a domain, and set the a records to 192.168.1.140 and then when im in my local network, it will go to 192.168.1.140??
[03:35] <thesheff17> Jasonn: yes I do it all the time.
[03:35] <Jasonn> Works?
[03:35] <Jasonn> Thanks
[03:35] <e_t_> No. Any address beginning with 192.168 is not routable on the Internet. You need a static public IP.
[03:36] <Jasonn> :OO
[03:36] <Jasonn> Then what can i do
[03:36] <Jasonn> ?
[03:36] <e_t_> How do you connect to the server?
[03:37] <thesheff17> Jasonn: I create domain names all the time that are webServerPriv.domain.com -> 192.168.7.15 and it works fine.
[03:37] <shauno> the "when I'm in my local network" bit is real relevant there.  you can put any address you want in the A record.  you'll only be able to reach it from places it's routable tho
[03:37] <thesheff17> A host records
[03:37] <Jasonn> e_t_: Local server
[03:37] <e_t_> I thought this was your VPS.
[03:39] <Jasonn> nope
[03:39] <Jasonn> talking about another serrver
[03:41] <e_t_> OK. If you want a server on your LAN to just do stuff on the LAN, you don't need a domain at all. If you want people on the Internet to be able to email you though, you need a registered domain name and a public IP address. You will then configure your router or gateway with the public IP address and forward all traffic to the server on the LAN.
[04:09] <pHro> i like this channel much better than #ubuntu, much better conversation
[06:40] <itsNe> Hi! Are there any sendmail guru's on?
[07:44] <qwebirc44173> How can I disable all the power management ?
[08:07] <mithran> hi can someone suggest me a good native tool to use for ubuntu as a client, to take ssh/telnet of an appliance and store the logs, execute commands, set term type, , etc?...I already tried teraterm via wine and putty, but is there any other tool that maybe better?
[08:07] <mithran> the client will be run on ubuntu 10.04 desktop
[08:08] <qwebirc44173> ^
[08:09] <mithran> hi  qwebirc44173: is xshell any good, ive seen some people run it?
[08:10] <xampart> a bit of a problem with dovecot+postfix vacation autoreply: our mailaddresses are the form $firstname.$lastname@$domain and they are delivered to $user@$dovecot.server. when i use sieves autoreply, the from-address in the reply message is $user@$dovecot.server, not the proper mail address.
[08:11] <xampart> is there a way to use the $first.$last@$domain in reply messages from-address?
[08:27] <shaggy2> hey guys, I need to change the ip address in bind files from 192.168.0.8 to a different ip is there a quicker way to do it?
[08:29] <shaggy2> better explination, I have about 15 bind9 database files, all with about 10 refferences to 192.168.0.8 I want to know if there is a quicker way to change it rather then manualy opening every file
[08:32] <andreasf_> shaggy2: Something like "sed 's/192.168.0.8/192.168.0.100/' < file1 > file1.new"
[08:33] <soren> Or "sed -i"
[08:33] <soren> Also, s/192.168.0.8/192.168.0.100/ might replace 192.168.0.87 with 192.168.1007, for example.
[08:33] <soren> So: Be careful.
[08:33] <shaggy2> wouldn't that copy the file and apply the changes in the new file?
[08:34] <soren> shaggy2: andreasf_'s solution would put the changes into a new file (which you can then proceed to rename, if you're happy with the changes).
[08:34] <andreasf_> Back up your original files before attempting this :-)
[08:34] <soren> shaggy2: "sed -i -e 's/blah/blah/g' file.conf" will act on file.conf in-place.
[08:35] <shaggy2> I might just stick with the way I know. might be safer. if I stuff something up I was planning to format anyway, I thought I would just try to change the ips so I didn't have to, bind 9 aint the only thing that needs to be changed
[08:37] <_ruben> use 'sed -i~', it'll backup the existing files with the ~ suffix
[08:38] <qwebirc44173> How can I disable all the power management/saving ?
[08:39] <soren> qwebirc44173: Why?
[08:41] <qwebirc44173> soren: problem shutting down.
[08:41] <qwebirc44173> soren: we are discussing at #ubuntu
[08:41] <soren> qwebirc44173: So you want to disable power management so that you can be sure it won't shut down?
[08:41] <qwebirc44173> lol
[08:41] <qwebirc44173> I want to be sure it shuts down*
[08:42] <soren> qwebirc44173: This makes no sense.
[08:42] <qwebirc44173> soren: power saving
[08:42] <soren> qwebirc44173: It's sort of like removing the wheels from your car, because you don't feel it's driving very well.
[08:42] <qwebirc44173> soren: how to disable power saving
[08:42] <qwebirc44173> maybe it's the problem. i need to know
[08:44] <soren> Powering off is power management.
[08:44] <soren> If you're talking about CPU throttling, that's a different story.
[08:45] <soren> I'm not sure you can thoroughly disable CPU throttling or frequency scaling.
[08:45] <soren> It used to be a module, but I think it's built-in no.
[08:46] <soren> now.
[08:46] <soren> qwebirc44173: You could talk to the good folks in #ubuntu-kernel.
[08:49] <qwebirc44173> ok
[08:49] <qwebirc44173> thanks
[08:51]  * sky1 is looking for an user which has experience with the tracking issue software Request Tracker  /msg me
[09:18] <huats> morning
[09:20] <jamespage> ttx: hey - snap!
[09:21] <ttx> jamespage: snap?
[09:21] <ttx> oh :)
[09:21] <jamespage> :-)
[09:21] <ttx> jamespage: young padawan needs to learn more if he wants to be as fast as the jedi master
[09:24] <jamespage> I think we must have been looking at it at exactly the same time - I though I had discovered a new 'feature' in launchpad that recognised the words 'duplicate' and 'bug xxx' and did automagic stuff :-)
[09:41] <soren> sky1: If you really want help, I suggest you just start explaining your problem.
[09:42] <soren> sky1: Even if I was a requesttracker expert, why would I /msg you?
[09:43] <sky1> hm... ok
[09:47] <soren> sky1: Here, in the channel, there are tons of people who can help. If I get bored along the way, someone else can pick up where I left off. What would be my motivation to give that up?
[09:47] <sky1> i would like to use a template  from an copy of a document for Improvement suggestions for RT..  it's no problem to convert these in a format like html or sth like that ... if i do so is it possible to integrate these html document  for a ticket creation in RT?
[09:48] <soren> sky1: For stuff like that, I recommend looking at one of the requesttracker mailing lists:
[09:48] <soren> http://bestpractical.com/rt/lists.html
[09:49] <sky1> ok ...thnx
[09:51] <eagles0513875> hey guys whats the name of the mod_zlib package for php on lucid?
[09:52] <_ruben> eagles0513875: there is none, as it's built in
[09:53] <eagles0513875> ok is it enabled by default or do i need to go digging around in the php.ini file
[09:53] <_ruben> i assume it's enabled by default
[09:55] <eagles0513875> ok will dig around in php.ini
[09:55] <eagles0513875> thanks
[10:07] <volkan> How do i remove a package pin? Can i just delete the /etc/apt/preferences.d/ files?
[10:10] <volkan> I checked the documentation but the method proposed doesn't work: echo PACKAGE_NAME install | dpkg --set-selections returns nothing and after apt-get update apt-cache policy PACKAGE_NAME is still set to the pinned version...
[10:13] <qwebirc44173> output of /sys/power/disk is test testproc [shutdown] reboot
[10:13] <qwebirc44173> is this the reason I cant shutdown the system properly?
[10:38] <qwebirc44173> ...........
[10:46] <rickspencer3> Daviey, hey, beta 1 is out, what's the word on the street wrt server?
[10:47] <Daviey> rickspencer3, haven't yet had much community feedback.. which isn't uncommon... looking at the incoming bug count, looking reasonable.  the beta1 qa'ing is looking pretty good.
[10:48] <Daviey> rickspencer3, The documented (release notes) eucalyptus bug should now be fixed in the archive, so apt-get upgrade away
[10:52] <rickspencer3> Daviey, nice!
[10:52] <qwebirc44173> Allocate IRQ for PCI VGA on
[10:52] <rickspencer3> Natty! Natty! Natty!
[10:52] <qwebirc44173> good?
[10:55]  * Daviey passes rickspencer3 a pair of pompoms.
[11:09] <ttx> Daviey, zul: saw http://www.stackops.org/ ?
[11:10] <ttx> Daviey, zul: Lucid-based openstack ISO, with agent-based post-install config
[11:13] <Daviey> ttx, I did!  I downloaded the iso.. not had a chance to complete installation with beta1 this week.
[11:14] <Daviey> but thanks for pointing it out.
[11:27] <zul> ttx: yep
[11:27] <jamespage> Daviey, ttx: we have had quite a few reports of bug 654549
[11:28] <jamespage> The fix is quite small; SRU worthy?  Happy to handle if you think so
[11:29] <Daviey> jamespage, looking
[11:30] <jamespage> Daviey: this is the fix - http://pastebin.com/Bq6tHhW6
[11:31] <Daviey> jamespage, is it just upgrades or new installs aswell?
[11:31] <jamespage> So it would be upgrades only where extra stuff has been added to /etc/default/tomcat6 with '/' characters
[11:32] <jamespage> normally JAVA_OPTS
[11:33] <Daviey> jamespage, So the only catchment is people updating from lucid -> maverick?
[11:34] <jamespage> Daviey: yes - and people applying security fixes to based Maverick installs which have had extra options added
[11:34] <jamespage> TBH it probably impacts Lucid as well
[11:34]  * jamespage goes to look
[11:35] <Daviey> Ah, if it does, and the fix looks trivial - and backported, shoot for it :)
[11:35] <jamespage> So this was added in Maverick by the looks of things - so does not impact Lucid
[11:36] <jamespage> Daviey: ^^
[11:37] <Daviey> jamespage, ahh... looks totally Fix Released then?
[11:37] <jamespage> Daviey: fixed release in Natty; however Maverick still has the issue
[11:38] <jamespage> Lucid not impacted
[11:38] <Daviey> jamespage, It doesn't seem that high impact to me... but if you have the patch, and can throw out the SRU dance, comments... seems a good idea.
[11:39] <jamespage> Daviey: well it can be worked around but it is annoying - I'll stick it on my list as low priority - have some ec2 testing stuff todo from beta-1 testing.
[11:40] <Daviey> jamespage, ec2 testing, fixes or using it?
[11:41] <jamespage> daviey: changes to the way it runs; broader use during beta-1 threw up a few things that don't work so well.
[11:41] <Daviey> Oh dandy.
[11:41] <jamespage> daviey: going to move to a new type of Jenkins project for it to make reporting easier - http://jenkins.qa.ubuntu-uk.org/job/natty_server_ec2/
[11:42] <Daviey> jamespage, Am i able to sign up to get mails on failures?
[11:42] <Daviey> That reporting matrix looks good!
[11:44] <jamespage> daviey: I think we should get some mailing lists setup for failure notifications - what do you think?
[11:45] <Daviey> jamespage, That sounds a really, good idea
[11:46] <jamespage> ack
[11:47] <Daviey> jamespage, Is it looking like your jenkins instance will cover more than just server eventually?
[11:48] <jamespage> daviey: TBH it already is - we have desktop and alternate ISO testing running in it at the moment
[11:48] <Daviey> Ah, i didn't know that had been achieved...
[11:49] <jamespage> daviey: it does not really fit into a t1.micro any more - I broke it yesterday when I used a new Jenkins feature - it ran out of memory....
[11:49] <Daviey> jolly good :)
[12:17] <doko_> what is the status of xen in main? still at -3.3?
[12:21] <raphink> rpinson@rpinson:~$ rmadison xen-3.3 | grep natty
[12:21] <raphink>    xen-3.3 | 3.3.0-1ubuntu13 |         natty | source
[12:24] <doko_> ahh, there is a patch in the bug tracker ...
[12:24] <doko_> server team, please have a look at the openldap and postfix build failures
[12:44] <qwebirc44173> I need help enabling APM support!
[12:44] <qwebirc44173> please!
[12:45] <soren> qwebirc44173: What sort of hardware are we talking about?
[12:47] <qwebirc44173> amd pcchips 810lmr
[12:47] <qwebirc44173> m810lmr
[12:47] <qwebirc44173> it supports APM
[12:48] <soren> What processor is it?
[12:48] <qwebirc44173> soren: http://www.ciao.co.uk/Productinformation/PCCHIPS_M810_LMR__5409872
[12:55] <jmazaredo> 0 0 * * 1-6 am I right? this will run the program at midnight every monday to saturday except sunday?
[12:55] <qwebirc44173> jmazaredo: man crontab
[12:56] <jmazaredo> ya i did that im just confused
[12:57] <qwebirc44173> theres a bunch of guides on the internetz
[14:36] <zul> Daviey: the ftbfs is a transient error i think
[14:37] <RoAkSoAx> mornin all
[14:38] <Doonz> is there an open source alternative to cpanel?
[14:44] <semiosis> Doonz: webmin, maybe
[14:45] <semiosis> hi RoAkSoAx
[14:51] <Doonz> semiosis: any recommendation on a vm server? don twanna use vmware
[14:51] <Doonz> is virtual box decent?
[14:58] <semiosis> Doonz: i havent used virtualbox, just vmware and centos xen.  have you checked out ubuntu enterprise cloud?  i plan on setting that up sooner or later
[14:59] <Doonz> im just starting to play around with it
[15:08] <iclebyte_work> Doonz, yes it is
[15:08] <iclebyte_work> very handy
[15:09] <iclebyte_work> I've run production VM's on it too
[15:09] <iclebyte_work> my experience has been that it's a solid product
[15:30] <MagicFab> hi all - if anyone speaks French here, you're invited to contribute to this translation (SAGE Sysadmin code of ethics): si qq'un veut participer à cette traduction, vous êtes invités: http://sketchpad.cc/PPM2DE5gDL
[15:39] <RoyK> MagicFab: #ubuntu-fr might be worth a try
[15:39] <MagicFab> RoyK, tx, I am inviting a few people
[15:48] <kirkland> cjwatson: hi, do you have an idea when you'll rebuild d-i for beta2?
[15:49] <cjwatson> I'll do it now
[16:02] <zul> Daviey: yeah so openldap ftbfs is bogus
[16:12]  * rcconf hello all
[16:13] <RoyK>  
[16:13] <rcconf> How can I use Ubuntu Server as Live CD?
[16:13] <RoyK> you can't
[16:13] <rcconf> no?
[16:13] <rcconf> I just want to get to cli
[16:14] <RoyK> oh
[16:14] <rcconf> :)
[16:14] <RoyK> press alt+left
[16:14] <RoyK> s/left/right/
[16:14] <RoyK> or alt+f<something>
[16:14] <rcconf> RoyK: in the first menu?
[16:14] <RoyK> I think you have to go through the language settings before the terminals spawn
[16:15] <rcconf> RoyK: where it say
[16:15] <rcconf> s
[16:15] <rcconf> Try Ubuntu
[16:15] <rcconf> no
[16:15] <RoyK> or even - choose 'repair' in the menu
[16:15] <rcconf> it only shows Install
[16:15] <rcconf> etc
[16:15] <RoyK> ubuntu server doesn't have the 'try ubuntu' choice (iirc)
[16:16] <rcconf> yes it doesnt
[16:16] <rcconf> so how I do boot and access the disks
[16:16] <RoyK> but if you choose install, and then go through the kbd/language settings, new terminals should be available on virtual terminal 2-4 IIRC
[16:17] <genii-around> 2 and 3 I think, 4 used to be for error messages from the install, etc
[16:17] <rcconf> RoyK: there is recue mode and stuff
[16:17] <RoyK> rcconf: that's the easy way, yes
[16:18] <rcconf> http://www.cyberciti.biz/faq/howto-boot-ubuntu-linux-rescue-mode/
[16:18] <RoyK> rescue will prompt for a root partition and mount that
[16:18] <rcconf> RoyK: it didnt mount
[16:18] <rcconf> dont know why
[16:18] <RoyK> well, if you get a terminal, then just mount it yourself :)
[16:18] <RoyK> cat /proc/partitions
[16:19] <RoyK> that'll show you what's there
[16:19] <rcconf> k
[16:24] <RoyK> rcconf: any luck?
[16:25] <rcconf> no
[16:25] <rcconf> RoyK: cat proc shows the partitions
[16:25] <rcconf> now what i do?
[16:26] <RoyK> mount <dev> <mountpoint>
[16:26] <rcconf> mount sda /media/ ?
[16:26] <rcconf> mount sda /mnt/
[16:26] <rcconf> or sda1
[16:26] <RoyK> sda is the drive
[16:26] <RoyK> sda1 is the first partition
[16:26] <rcconf> yes
[16:26] <RoyK> mountpoint can be anything
[16:27] <RoyK> I guess /mnt should do fine
[16:27] <RoyK> even /asdf will do :P
[16:28] <rcconf> RoyK: mount /dev/sda1 /media/ mount fail
[16:28] <rcconf> invalid argument
[16:28] <RoyK> rcconf: pastebin /proc/partitions
[16:28] <rcconf> cant
[16:28] <RoyK> well, what partitions do you have there?
[16:29] <rcconf> oh i will try the home partition
[16:29] <RoyK> if in doubt, run a fsck on them first
[16:29] <rcconf> :)
[16:29] <rcconf> i was trynig to mount swap
[16:29]  * rcconf facepalms
[16:30]  * RoyK hands rcconf a beer :)
[16:33]  * rcconf drinks water
[16:33] <rcconf> :D
[16:33] <jacobfogg> We currently have 9 web/db servers and I am trying to map them out and am looking for some advice as to what to include in my report... I have a good start, but don't want to miss anything worthwhile...
[16:35] <ttx> daviey, zul: if you want to have parallel LXC/KVM support in openstack, you should probably +1 the "Re: [Openstack] heterogeneous instance types" thread
[16:38] <RoyK> jacobfogg: start off with distro/kernel versions, list of installed packages and perhaps an OpenVAS scan
[16:38] <zul> ttx: done and done
[16:39] <jacobfogg> RoyK: got everything but the scan so far.
[16:40] <jacobfogg> I also have hardware specs: Proc's, Ram, HD's
[16:40] <RoyK> lshw is nice for that
[16:41] <jacobfogg> Nice! ... I also have Internal/External IP's, Primary and secondary (if applicable) purpose
[16:41] <RoyK> OpenVAS can be a bit paranoid, so try to weight those alerts on your mental scales :P
[16:52] <a456> hey everyone, i have an ec2 instance on amazon with ubuntu, i see my test page for apache, but inside of /var/www/ i see folders for cgi-bin, html but nothing inside them, any idea where the page is served from
[16:52] <RoyK> jacobfogg: make sure you run chkrootkit on those machines as well
[16:52] <RoyK> a456: check the apache config
[16:53] <jacobfogg> Thanks =)
[16:54] <a456> RoyK: i cant find the apache2 config
[16:54] <RoyK> a456: /etc/apache2
[16:55] <raphink_afk> grep -r "cgi" /etc/apache2
[16:55] <a456> hmm -bash: cd: /etc/apache2: No such file or directory
[16:55] <raphink_afk> how did you install apache?
[16:55] <RoyK> a456: perhaps lighttpd is running?
[16:55] <a456> it was preinstalled on amazons ec2 image
[16:55] <a456> http://50.17.186.175/
[16:56] <RoyK> lol - that's fedora
[16:56] <a456> so its apache
[16:56] <RoyK> not ubuntu
[16:56] <a456> oh fuck
[16:56] <a456> hah
[16:56] <a456> sorry ive been up since 4am working /west coast time/ grr
[16:56] <RoyK> a456: /etc/httpd
[16:57] <RoyK> a456: and perhaps /j #fedora :P
[16:57] <raphink_afk> ;-)
[17:02] <a456> thx guys hah
[17:02] <a456> i was confused because any host ive had had a test page liek that but included an html file
[17:03] <a456> that was served via the server? the dir was emtpy which is why i was co nfused once i "touch index.html"
[17:03] <a456> i could see, thanks!
[17:30] <jamespage> hggdh: around?
[17:45] <Nafallo> hallyn: hi. I've just updated bug 746497
[17:45] <Nafallo> hallyn: hopefully that's all of the questions answered...
[17:46] <RoyK> heh - starting an openvas scan on this old single cpu machine: load average: 57.30, 53.72, 39.70
[17:50] <Kiall> RoyK, didn't you're parents talk to you about playing with matches? The same rules apply to what you're doing right now ;)
[17:52] <RoyK> Kiall: lol
[17:52] <RoyK> Kiall: just scanning a network I manage :)
[17:53] <RoyK> but from the outside to see if there are any cracks
[17:53] <Kiall> ... and starting a fire at the datacenter that server is hosted at while you're at it ;)
[17:53] <RoyK> haven't seen a server that can't handle 100% load yet
[17:54] <RoyK> anyone that produces a server that stops at 50% should be hanged
[17:54] <Kiall> I have .. and it did end in a fire too .. but it was a crappy old dust filled desktop someone was doing stupid shit with ;)
[17:55] <RoyK> then the problem probably was noone had vacuumed the server recently...
[17:55] <RoyK> dust can be quite a bitch
[18:09] <hggdh> jamespage: now I am
[18:31] <cjwatson> kirkland: uploaded a new debian-installer; should be ready for tomorrow morning's daily builds
[18:31] <kirkland> cjwatson: rock, thanks
[18:31] <kirkland> cjwatson: will test then
[18:38] <hallyn> Nafallo: so restarting libvirt-bin did reinstate the routes right?
[18:40] <Nafallo> hallyn: routes? who talked about routes?
[18:40] <hallyn> Nafallo: filter rules i meant
[18:41] <hallyn> all right well let me try to reproduce
[18:41] <hallyn> thanks for the info
[18:41] <Nafallo> it did indeed, yeah :-/
[19:32] <RoAkSoAx> SpamapS: any ideas why this is showed? while upgrading eucalyptus start: Unknown parameter: IF
[19:32] <RoAkSoAx> start: Unknown parameter: IFACE
[19:38] <Daviey> RoAkSoAx, It has always done that, it's when, aiui, an emit doesn't fill $iface... the problem is that upstart doesn't let it pass unoticed now
[19:38] <Daviey> not sure of the fix tho :)
[19:40] <SpamapS> hmm
[19:40] <SpamapS> is IFACE used in the script/exec lines?
[19:44] <Decepticon> is amazon AWS doing some april fools joke? i cant login since april 01 12:00 or ping the httpd
[19:44] <Decepticon> it shows as a running instance though
[19:49] <RoAkSoAx> SpamapS: no it is not
[19:52] <RoAkSoAx> SpamapS: this is the job http://paste.ubuntu.com/588357/
[19:53] <RoAkSoAx> SpamapS: ARGH.. this one.. http://paste.ubuntu.com/588358/
[19:54] <RoAkSoAx> SpamapS: but the upstart job has not changed in quite a while
[19:54] <RoAkSoAx> and because of that, the upgrade /install in natty fails
[19:59] <Daviey> SpamapS, IFACE is not used... previously that was not a problem with upstart... it would give a warning, but not fail to start
[19:59] <SpamapS> oh but it *IS* used
[19:59] <Daviey> SpamapS, interetingly, when i appended IFACE=eth0, it still blocked
[19:59] <SpamapS> instance $IFACE
[19:59] <Daviey> SpamapS, yeah
[19:59] <SpamapS> thats very very important
[19:59] <Daviey> but in natty, it matters that it's not provided
[20:00] <SpamapS> Its *critical*
[20:00] <Daviey> in <natty, it didn't
[20:00] <SpamapS> and actually a good thing that its refusing to start
[20:00] <Daviey> SpamapS, it wasn't critical pre-natty
[20:00] <SpamapS> its exposing a confusing thing though
[20:00] <SpamapS> why is it used in instance, and not in the scripts
[20:00] <Daviey> SpamapS, one moment
[20:01] <SpamapS> I'd like to see a comment explaining that added to the file
[20:01] <Daviey> SpamapS, it gave a warning before natty, but didn't fail...  I questioned it during the cycle, but because it worked - we just carried on with it.
[20:01] <Daviey> as a unknown :/
[20:01] <SpamapS> It looks like the script needs to run multiple times.. every time a net-device-up runs..
[20:02] <SpamapS> So, you just need to add this
[20:02] <SpamapS> env IFACE=manual-start
[20:02] <Daviey> SpamapS, well yeah, but postinst, we don't emit the interfaces
[20:02] <Daviey> SpamapS, Ahh!
[20:02] <Daviey> SpamapS, to the postinst or the upstart job?
[20:02] <SpamapS> upstart job
[20:03] <Daviey> SpamapS, when was that introduced?
[20:03] <SpamapS> that way any time somebody runs 'start eucalyptus-network' it works
[20:03] <SpamapS> Daviey: natty of course. :)
[20:04] <SpamapS> it bit me a few times as well
[20:04] <Daviey> SpamapS, ok.. but doesn't change anything else?
[20:04] <SpamapS> No, as you guys have noted, its not used anywhere else.
[20:04] <Daviey> ie, the current job works as is, after a reboot.
[20:04] <SpamapS> I see the point of the thing..
[20:04] <SpamapS> its saying "check to see if there's a default route every time a net device is up"
[20:05] <Daviey> oooo
[20:05] <SpamapS> and w/o the instance stanza, it would race if multiple interfaces came up
[20:06] <SpamapS> I would suggest that it be   'start on net-device-up IFACE!=lo' .. since the code makes no sense on loopback only.
[20:06] <Daviey> SpamapS, so add both of them entries, the env and amend the start on line?
[20:07] <Daviey> hggdh, are you around?
[20:07] <hggdh> Daviey: always
[20:07] <Daviey> hggdh, :D
[20:07] <Daviey> hggdh, Are you free? :)
[20:08] <hggdh> expept, of course, when I am elsewhere
[20:08] <hggdh> Daviey: I was until now. But, mind you, I am married
[20:08] <Daviey> heh
[20:08] <hggdh> Daviey: what can I do for you?
[20:08] <Daviey> hggdh, Are you able to re-provision these machines, with stock beta1?
[20:09] <Daviey> well.. actually.. only cempedak actually.
[20:09] <hggdh> Daviey: hum. For stock beta1, it might be better to reinstall
[20:09] <hggdh> no biggie
[20:09] <hggdh> (I ran a dist-upgrade, and some other packages got upgraded as well)
[20:10] <hggdh> Daviey: give me ~ 20 min, and we will be all set
[20:11] <Daviey> hggdh, yeah.. cempedak uncovered an upgrade issue
[20:12] <hggdh> Daviey: just what we needed
[20:12] <Daviey> hggdh, think we have a fix, thanks to SpamapS .
[20:12] <Daviey> just need to test it
[20:13] <hggdh> Daviey: installing stock beta1 now
[20:13] <Daviey> hggdh, groovy
[20:27] <kirkland> RoAkSoAx: ping
[20:28] <RoAkSoAx> kirkland: pong
[20:28] <kirkland> RoAkSoAx: i'm going to head your way shortly
[20:28] <kirkland> RoAkSoAx: sorry, i've been caught up here all day :-(
[20:28] <RoAkSoAx> kirkland: no worries :)
[20:28] <kirkland> RoAkSoAx: did you find some good wifi at a coffee shop?
[20:28] <kirkland> RoAkSoAx: there's some good places around there
[20:29] <RoAkSoAx> kirkland: actually end up staying in the room :). But, the Java Jives cafe seems like a good polace
[20:29] <kirkland> RoAkSoAx: it's a beautiful day though!
[20:29] <RoAkSoAx> and its right next to the hotel
[20:29] <kirkland> RoAkSoAx: i'm going to come pick you up
[20:30] <kirkland> we'll get you out the hotel shortly
[20:30] <kirkland> RoAkSoAx: once again, my apologies
[20:30] <RoAkSoAx> kirkland: no worries, was working here too :) and btw.. robbiew said he might be coming too
[20:31] <kirkland> RoAkSoAx: cool
[20:31] <kirkland> robbiew: want to meet us at Austin Java on Barton Springs road?
[20:31] <robbiew> nah...I need to drop off stuff at the venue
[20:31] <kirkland> robbiew: I figured we'd finish the work day there, or at Uncle Billy's before heading over to Threadgills
[20:31] <kirkland> robbiew: ah
[20:32] <robbiew> kirkland: they having the speaker's dinner at Threadgills again?
[20:32] <kirkland> robbiew: yes, but Riverside location, at least
[20:32] <kirkland> robbiew: which is better than the original
[20:32] <robbiew> ah, very nice
[20:32] <robbiew> for sure
[20:32] <kirkland> robbiew: do you need help @ the venue?
[20:32] <kirkland> robbiew: either today or tomorrow?
[20:33] <robbiew> not today...as for tomorrow, I can't sit there all day...but have folks lined up to man the booth
[20:34] <robbiew> just given out CDs at the table
[20:34] <robbiew> and ubuntu lanyards as bag swag
[20:34] <RoAkSoAx> robbiew: my presentation is afternoon so I can be there most of the morning
[20:36] <robbiew> ok
[20:36] <SpamapS> hmm
[20:36] <SpamapS> isn't 'telinit 1' supposed to bring down the network?
[20:43] <SpamapS> Daviey: sorry yeah, I'd say both changes would be good. Though the start on change may need some more thought/testing than we can put in between beta1 and beta2. ;)
[20:44] <Daviey> heh
[20:49] <SpamapS> Daviey: we should add an event..   default-route-configured
[20:50] <SpamapS> Daviey: any time the default route changes that should be emitted. Lots of things would do well to start at that point rather than net-device-up IFACE!=lo
[20:50]  * SpamapS puts it on the Oneiric idea pile
[20:50] <hggdh> Daviey: CLC/Walrus/CC/SC installed on cempedak, now installing the NCsa
[21:13] <hggdh> Daviey: all installed, pristine beta1. I have to go out for 20 min
[21:16] <hallyn> hggdh: so the problem yesterday was due to an upgrade issue?
[21:16] <hallyn> Daviey: ^ ?
[21:17] <SpamapS> DOHHH
[21:17] <SpamapS> ssh isn't restarted on runlevel 1->2 :(
[21:17] <SpamapS> or rather, sshd
[21:21] <Daviey> -otp
[21:22] <hallyn> yeah we need a new one time pad.  let's just do plaintext for now
[21:23] <hallyn> bbl
[21:38] <genii-around> Probably an upstart thing
[21:57] <hggdh> hallyn: I do not know, I don't know what SpamapS and Daviey found
[21:57] <SpamapS> genii-around: definitely an upstart thing :)
[21:57] <hggdh> Daviey: I am back, what do we need to do?
[21:57] <SpamapS> the job was changed recently to have sshd stop on runlevel [!2345]
[21:58] <SpamapS> but it only starts on 'filesystem'
[22:08] <hggdh> SpamapS: can you please summarise the changes to euca upstart -- I will then test the beast
[22:09] <SpamapS> hggdh: add 'env IFACE=manual-start' to the file
[22:09] <hggdh> SpamapS: to the eucalyptus-netkwork.conf, correct?
[22:13] <SpamapS> hggdh: right
[22:14] <hggdh> k, doing it. Will reboot after
[22:20] <hggdh> alright. I am just confused on *what* bug is this. I do not see it having any impact on the blocker we have
[22:21] <kthomas_vh_> howto install apache2-mpm-worker without forcing update of php to 5.3,  under 10.x?
[22:21]  * kthomas_vh_ immediate googles
[22:21] <kthomas_vh_> *-ly
[22:32] <hggdh> SpamapS: OK. Now, what exactly am I checking for here?
[22:32] <hggdh> all I see is that euca did not auto-start on a reboot
[22:32] <SpamapS> hggdh: that you can upgrade the package
[22:33] <SpamapS> RoAkSoAx: ^^ can you help hggdh with a more clear test case?
[22:36] <RoAkSoAx> SpamapS: will have to do it a bit later as I'm kinda fixing something here :)
[22:36] <hggdh> hallyn: so, by the above, this has nothing to do with the issue we found yesterday
[22:36] <RoAkSoAx> hggdh: but basically, once you do a fresh install, try to upgrade with a later euxalyuptus version
[22:36] <RoAkSoAx> and you should be able to upgrade without any issues
[22:36] <RoAkSoAx> and not having eucalyptus failing to upgrade because of IFACE thing oin the upstart job
[22:37] <hggdh> RoAkSoAx: will do. But, worth noting, I made the change & rebooted -- euca did *not* came up
[22:37] <hggdh> meanwhile, upgrading
[22:37] <SpamapS> hggdh: you can probably even get it with --reinstall
[22:38] <RoAkSoAx> hggdh: yeah try that
[22:42] <hallyn> hggdh: jjohansen: have you learned anything on bug 746751 ?
[22:44] <jjohansen> hallyn: sorry really just started looking as I was finishing up with another bug before switch context
[22:45] <hallyn> jjohansen: np, was just wondering
[22:52] <RoAkSoAx> kirkland: http://me.roaksoax.com/cobbler.debdiff
[22:53] <jjohansen> hallyn: is there anyway to get access to the hardware in question or should I have hggdh run some test kernels
[22:55] <hggdh> jjohansen: all you need is access to tamarind, you should be able to ssh from there to cempedak/mabolo/marula/santol/sapodilla/soncoya
[22:55] <hggdh> jjohansen: the first is the CLC/CC/Walrus/SC, the others are the NCs (where KVM is run)
[22:56] <jjohansen> hggdh: I don't have access to tamarind
[22:57] <hggdh> jjohansen: then I can do it -- but I have a dinner to go in a few
[22:57] <jjohansen> hggdh: well thats okay I won't have a kernel for a while
[22:58] <jjohansen> honestly not until some time on the weekend /me needs to look at it more first
[22:58] <RoAkSoAx> kirkland: #731616
[22:58] <hallyn> jjohansen: what're you going to try?
[22:58] <hallyn> hggdh: i got the access.  however, i don't really know my way around those systems...
[22:58] <hggdh> jjohansen: no prob. I will be fireflying the channels during the weekend
[22:59] <hallyn> hggdh: i've looked over the wiki page,
[22:59] <hallyn> is there a 'normal' way to start looking around?
[22:59] <jjohansen> hallyn: not sure yet, just wanted to know about testing for when I had smething
[23:00] <hggdh> hallyn: I usually start with ssh-copy-id to the systems
[23:01] <hggdh> hallyn: you might want to add some lines to your ~/.ssh/config, though
[23:02] <hallyn> but is any machine free/safe?
[23:02] <hallyn> hggdh: i was planning to look around next week
[23:02] <hallyn> was hoping to figure out this vmbuilder ec2 port
[23:03] <hggdh> hallyn: all of them have natty Eucalyptus installed; I am running a dist-upgrade right now
[23:03] <hggdh> hallyn: they are free to use, I can always re-gen the beast
[23:03] <hallyn> cool, thanks
[23:04] <kirkland> RoAkSoAx: SpamapS: http://paste.ubuntu.com/588420/
[23:04] <kirkland> RoAkSoAx: this includes your koan fixes
[23:04] <kirkland> SpamapS: and the debconf stuff
[23:04] <kirkland> SpamapS: eyeball that diff while i build and test it locall here
[23:10] <jjohansen> hggdh: this worked on maverick, tight?
[23:19] <hggdh> jjohansen: yes. But we were using CCISS then
[23:25] <jjohansen> hggdh: I am not convinced that is the reason, but it is worth looking at
[23:34] <kirkland> RoAkSoAx: http://paste.ubuntu.com/588426/
[23:34] <kirkland> RoAkSoAx: something's wrong in cobbler.postinst
[23:34]  * RoAkSoAx looks
[23:34] <kirkland> RoAkSoAx: on dpkg -i, it's hanging
[23:35] <kirkland> RoAkSoAx: i have some debug code in postinst right now
[23:41] <hggdh> jjohansen: neither are we convinced. But it is worth noting the difference.
[23:42] <jjohansen> hggdh: right, do we know if any of the natty kernels worked?
[23:43] <hggdh> jjohansen: no, we do not -- this is the first time in Natty that we have been able to run Euca on that hardware
[23:43] <jjohansen> hggdh: okay, then I think the place to start is with a mid point bisect on the kernels
[23:44] <hggdh> jjohansen: perfect
[23:45] <jjohansen> hggdh: hrmm, I'm not sure whether I will do bisects on the mainline or just grab older natty kernels for the first couple of tests.  But either way I'll have a kernel to point you at later tonight
[23:46] <hggdh> jjohansen: OK. I am going out to a dinner, but will be monitoring the channel
[23:46] <hggdh> of course, when I get back :-)
[23:47] <jjohansen> hggdh: /me was also thinking it would be good to get a trace with the maverick kernel installed on natty.  ie.  do the strace on kvm with the maverick kernel, that would give us a better comparison point for the kind of output expected in the strace log