[00:02] <Bitwise> When I do an nslookup I get this: http://pastebin.com/UVXUf6az
[00:02] <Bitwise> I have a feeling the first two lines aren't how they should be.
[00:04] <sarnold> Bitwise: well, yes and no. it does show what a local caching proxy has discovered. But that's of no help to other machines on the internet. nslookup lets you query specific servers. Once you figure out why your registrar threw away your domain, you'll be able to figure out which DNS servers to query to make sure it is alive again
[00:11] <Bitwise> sarnold, Would it matter that I'm not accepting connections on port 80?
[00:11] <sarnold> Bitwise: no. what matters most is that your "whois example.org" queries show you don't exist.
[00:11] <Bitwise> I don't want anyone using the server unless they're on localhost.
[00:12] <sarnold> wait, what?
[00:12] <Bitwise> I'm just trying to host an email server for my own use only.
[00:16] <sarnold> oh, so you're surprised that your domain doesn't exist. this is an odd use case to say the least, I don't know how well most existing tools will handle this.
[00:16] <sarnold> s/you're surprised/you're not surprised/
[00:18] <sarnold> Bitwise: if you can tolerate pop you might want to consider a vastly different alternative: https://www.powerdns.com/powermail.html
[00:18] <Bitwise> What exactly is different? I didn't think having the server go down whenever my computer is off was very interesting.
[00:19] <sarnold> Bitwise: most people want their mail servers to community with the internet. :)
[00:19] <Bitwise> The DNS entry won't propagate if my computer/server is off/down?
[00:21] <sarnold> it can't; you don't have a domain.
[00:23] <Bitwise> sarnold, https://www.whatsmydns.net according to this website my site is on all but one and my mx record is on all
[00:25] <sarnold> awesome site
[00:27] <Bitwise> sarnold, Shouldn't I be able to send/receive mail locally?
[00:29] <sarnold> Bitwise: hehe, at this point I'm so confused about what you're trying to achieve and how you're doing it that I'm out of ideas. I suspect if you want to keep using the tools you've chosen that you'll need to configure things to make them happy, but your setup seems so far from normal that I just don't know what to suggest next. sorry.
[00:31] <Bitwise> I'm just trying to turn my computer into a simple mail server that can send/receive to/from gmail. I like the RoundCube user interface so I suppose I'll use that but I'll gladly use something else if setup proves easier.
[00:32] <sarnold> Bitwise: okay, that makes some sense -- except for "receive from gmail" -- what do you mean by that?
[00:33] <sarnold> Bitwise: well, okay, I guess now that I think about it I'm also confused by "send to gmail" :) do you have a gmail account that you'd like send through? or ..
[00:33] <Bitwise> sarnold, A common user sending me emails from their gmail account.
[00:35] <sarnold> Bitwise: how will gmail's mail servers find your mail server if you intend it to be reachable only from localhost?
[00:36] <Bitwise> sarnold, I meant only I should be able to access the sending interface.
[00:37] <Bitwise> As in, I will be the only person who can send an email from my server.
[00:37] <sarnold> Bitwise: ah. the 'sending interface' is going to be the /usr/sbin/sendmail binary...
[00:40] <Bitwise> Yes, which I'm hoping can be simplified a bit by using RoundCube.
[00:52] <sarnold> Bitwise: sorry, I don't have the time today to properly confine an application not in the archive
[00:53] <Bitwise> Alright, no problem.
[01:08] <hallyn> ok the problem appears to be the trust kernel.  qemu version don't really matter.  i'll try to verify that tomorrow with various host kernels
[01:08] <lifeless> hallyn: what bug?
[01:09] <hallyn> lifeless: nested qemu, where bottom host is trusty kernel, cannot start a qcow2 image - grub can't load the kernel or something
[01:09] <hallyn> lifeless: i've at least verified it on a precise and trusty userspace, both with trusty kernel
[01:10] <hallyn> the first level qemu guest is trusty
[01:10] <hallyn> (have not varied that yet)
[01:10] <hallyn> tomorrow i'll see whether different firs tlevel guests help, and then i guess i'll have to bisect the host kernel
[01:10] <hallyn> anyway i'm outta here.  can't set that test up now
[01:10] <lifeless> yuck!
[01:11] <lifeless> gnight
[01:11] <hallyn> gnight
[01:55] <shreezbot> Hello!  I need a little help troubleshooting an issue I'm having with a server rebooting for no apparent reason...
[01:55] <Patrickdk> check your psu
[01:55] <Patrickdk> or did you want us to throw out other random things to check?
[01:56] <shreezbot> I've looked in the syslog and in almost all of the logs in /var, but none of them give me any clues as to why it could.
[01:56] <Patrickdk> then it's likely not an os issue
[01:56] <Patrickdk> time to start looking at hardware
[01:56] <shreezbot> Hmmm, is there a decent tool that might help me diagnose potential hardware issues?
[01:57] <Patrickdk> dmesg
[01:57] <Patrickdk> but if nothing is there, well
[02:00] <sheptard> shreezbot: memtest86
[02:00] <sheptard> shreezbot: also check psu voltages
[02:01] <shreezbot> lm-sensors is showing that my voltages look good
[02:01] <Patrickdk> psu voltages yes, but that isn't a reliable test though :(
[02:01] <Patrickdk> atleast for me, had 3 psu's die, but voltages where always good
[02:02] <shreezbot> Is there a better test other than breaking out a voltmeter?
[02:02] <Patrickdk> voltmeter is a horrible test
[02:02] <sarnold> shreezbot: I had a machine lock without warning when under load due to machine check exceptions; you could try installing mcelog and see if you get some that are fixed, which might help explain the ones that aren't fixed..
[02:02] <Patrickdk> you need to stress test it, put a good load on it, and see if the voltage is still stable then
[02:03] <Patrickdk> I assume no ecc memory? and you check the bios log if you have one
[02:03] <Patrickdk> I had a system that was set to reboot on ecc memory errors
[02:03] <shreezbot> One thing I did notice is that after the machine comes back up, the CPU seems to be hot as fuck.  The CPU fans spin up really high to cool it down...
[02:03] <sarnold> shreezbot: that's common when booting, many systems do that; if that's the only symptom of hot cpu, it might not actually be hot..
[02:04] <shreezbot> Well, when I check lm-sensors, it is actually over 55 degrees.
[02:04] <sarnold> ever been around a mac pro dual g5 tower? damn thing sounds louder than most servers when it boots..
[02:05] <sarnold> BIOS tends to be horrible code that draws surprising power, or so I've been lead to understand. 55 doesn't seem too horrible..
[02:05] <shreezbot> I'm just worried that I have a hardware issue with no real way to troubleshoot it...
[02:05] <th3joker> Crikey ever had one of the Mac 2U servers, damn thing is louder than a jet engine
[02:06] <shreezbot> Since I don't have a ton of spare parts lying around...
[02:06] <th3joker> 55 is ok, wouldn't worry about that
[02:08] <shreezbot> I'm guessing a bunch of "^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@" in the syslog means that's when it likely restarted, right?
[02:08] <th3joker> Check your ram that's usually the culprit for random reboots, if you have physical access to the server
[02:09] <th3joker> Download the memtest ISO and run it if there's a cd drive or dd it to a USB drive if not
[02:09] <shreezbot> How long do those tests usually take to run?
[02:10] <th3joker> Anything from a few minutes to a few hours
[02:10] <shreezbot> Unfortunately, this is a production system  :)
[02:10] <th3joker> Depends how much ram you have for starters
[02:10] <th3joker> Out of hours ;-)
[02:11] <shreezbot> This particular machine is a physical box and it has 24 GB
[02:11] <th3joker> I would pull the modules, clean them and then test, you can rule out bad connection and bad ram at the same time
[02:12] <th3joker> Save you running the test twice at least
[02:12] <shreezbot> I may have to do that
[02:13] <th3joker> Yeah, ugly but gotta be done sometimes
[02:13] <shreezbot> We don't have a lot of Linux in our shop, so I don't know much about it unfortunatley, but we have this ONE app that the vendor insists will only run on Ubuntu...
[02:14] <th3joker> Wait this is Ubuntu, there's a memtest built into grub... DOH
[02:14] <th3joker> When you boot select the memtest option
[02:14] <th3joker> Lol
[02:14] <shreezbot> lol, I'll do that
[02:14] <shreezbot> I just hope it doesn't take too terribly long
[02:15] <th3joker> Yeah it can take a few minutes, not always though
[02:15] <th3joker> Leave it overnight
[02:15] <shreezbot> Oh, I can't do that
[02:15] <th3joker> 24/7 access?
[02:15] <shreezbot> Yup
[02:16] <th3joker> is it hosting websites or db?
[02:16] <shreezbot> Unfortunately
[02:16] <th3joker> Etc.
[02:16] <shreezbot> db
[02:16] <th3joker> Pants
[02:16] <shreezbot> It is some rules engine db
[02:16] <th3joker> Any failover
[02:16] <shreezbot> Of course not!
[02:16] <th3joker> Lmao, I love those sort
[02:17] <shreezbot> If they gave me money for failover I wouldn't be awake right now!  :D
[02:17] <th3joker> They'll be moaning when it dies and you have to wait for new modules
[02:17] <shreezbot> Yea, and I'll have to worry if they're going to fire my ass when I was the one that insisted that we have failover
[02:18] <shreezbot> For an Operating system that I warned them ahead of time that I don't know shit about...
[02:18] <Patrickdk> heh, time to virtualize it
[02:18] <th3joker> Eek
[02:19] <shreezbot> Yea, I'm anxiously looking for an opportunity to get out of this mess...
[02:19] <th3joker> Contract it out
[02:19] <Patrickdk> that is one thing I do love about virtualizeation, move the vm's, then take the host down for inspection
[02:20] <shreezbot> Unfortunately this is one of two Linux servers in the entire shop, and the other one is physical as well.
[02:20] <shreezbot> And our virtualization guys won't touch Linux...
[02:20] <shreezbot> Windows only...
[02:21] <th3joker> Yeah gotta love vm
[02:21] <shreezbot> Sorry, Hyper-V only...
[02:21] <shreezbot> :)
[02:22] <shreezbot> Honestly, this experience has given me a newfound love for Linux.  I intend to expose myself to it as much as possible..
[02:22] <shreezbot> It just runs so much better than Windows in the server world.  Windows servers have entirely too much overhead just for the OS
[02:22] <th3joker> Ubuntu server is pretty awesome, in fact Linux distros are all awesome
[02:23] <shreezbot> I'll agree
[02:23] <shreezbot> I've converted all of my machines at home to Ubuntu exept for my gaming laptop
[02:23] <th3joker> Switched to Linux servers years ago and never looked back
[02:23] <th3joker> I quite like freenas too for home use
[02:23] <th3joker> Have a look if you get the time
[02:23] <Patrickdk> I do a mixture of windows and linux servers
[02:23] <th3joker> Has zfs built in which is nice
[02:24] <shreezbot> Oh, that is cool1
[02:24] <Patrickdk> ubuntu has zfs too
[02:24] <th3joker> Yeah but it's not au-naturel in Ubuntu
[02:24] <shauno> worth noting that hyper-v doesn't mean you can't host linux guests; msft list ubuntu 12.04 onwards as supported - http://technet.microsoft.com/library/cc794868(WS.10).aspx
[02:24] <Patrickdk> th3joker, why not?
[02:25] <Patrickdk> I wonder if the new freebsd release fixed up their zfs stuff
[02:25] <shreezbot> shauno, Yea, I know it is supported, and our virtualization guys know that as well, they just don't want to touch it because they don't know anything about it...
[02:25] <Patrickdk> cause zfs on freebsd was a lot of issues
[02:26] <sarnold> llnl supports ZoL.. is there a similar Big User of zfs on freebsd?
[02:27]  * Patrickdk is using illumos, pretty happy with it
[02:27] <shreezbot> My primary job responsibility is a z/OS systems programmer and we use zfs for the UNIX system services
[02:27] <dthorman> Hello I was wondering if anyone else was having issues installing packages in us-west-2 I am getting E: Failed to fetch http://us-west-2.ec2.archive.ubuntu.com/ubuntu/pool/main/libe/liberror-perl/liberror-perl_0.17-1_all.deb: 403  Forbidden
[02:27] <shreezbot> It is amazing
[02:27] <dthorman> I was also getting that error for some other packages
[02:27] <Patrickdk> I was thinking about moving half my systems to illumos, but then though, dunno if I want to package all that software
[02:31] <dthorman> the above 403 occurred while just attempting to install git, it doesn't look like all packages fail but I wasn't sure if anyone had a work around... maybe another package source I could point to
[02:34] <shreezbot> It is weird stuff like this that makes me question my sanity.  I have a cronjob that runs fine when I manually, but when cron runs it, it starts and then just sits there...
[02:34] <sarnold> dthorman: thanks, being looked at
[02:35] <dthorman> sarnold: Ok, thank you :)
[02:35] <sarnold> dthorman: us-west-1 ought to work, might be cheaper than jumping to us-east* or mirror.anl.gov ..
[02:36] <dthorman> sarnold: Ok, I will use that for now
[02:37] <dthorman> sarnold: is there an ETA on a fix or is there someone who might know?
[02:39] <sarnold> dthorman: our guy said it ought to fall back to the main servers within a few minutes, but a real fix will take longer
[02:40] <shauno> shreezbot: that's a very typical problem.  when you run it manually, there's a lot of environment variables (especially the current path) set; when cron runs it, there isn't.  so you need to check your script for such assumptions
[02:43] <Bitwise> Will DNS even work if I have mydomain.no-ip.org rather than my very own domain.tld ?
[02:44] <Bitwise> As in could I have to point to my computer and then have mail.mydomain.no-ip.org also point to my computer?
[02:46] <shauno> what are you actually trying to do?  this sounds a lot like you're trying to solve the wrong problem
[02:51] <Bitwise> shauno, I'm trying to host my own basic mail server.
[02:52] <dthorman> sarnold: Ok, thanks again!
[02:53] <sarnold> dthorman: thanks for the report! :)
[02:54] <th3joker> Bitwise: Never tried to run a mail server on a non static ip
[02:54] <th3joker> Internally you would point your clients at the local ip
[02:55] <th3joker> No need to go external to come back in
[02:55] <Bitwise> My IP is static.
[02:57] <th3joker> Which mail server are you planning on running
[02:57] <Bitwise> postfix/courier
[02:57] <th3joker> Yeah postfix is nice and easy
[02:58] <th3joker> Do you have a domain registered?
[02:58] <th3joker> With some form of external DNS setup
[02:58] <shauno> I'm not sure I'd use the word 'easy'; trying to run a mailserver on a residential connection is harsh on the best of days
[02:59] <th3joker> Nah, I have 80MB fibre here would run a mail server easy peasy
[02:59] <th3joker> Not sure what Bitwise has though
[02:59] <sarnold> shauno,th3joker, I believe Bitwise intends to turn this machine off from time to time too; it'd be best to ask for details before solving problems ;) I got lost though...
[02:59] <th3joker> Lol
[02:59] <th3joker> Well that's not really going to work then
[03:00] <shauno> it's not so much the speed, it's that residential isps don't score to well in dns blacklists, since most the mail coming from them is from compromised windows boxes
[03:00] <Bitwise> I'm the only one who is going to be sending mail from it. I have a no-ip.org subdomain and I registered an mx record as mail.mydomain.no-ip.org doesn't seem to working as whois lookups fail, as sarnold said earlier.
[03:01] <shauno> do you plan on receiving inbound mail at the same server?
[03:01] <th3joker> Why set up a mail server to simply send mail for one client, seems a bit like overkill
[03:02] <th3joker> Try zoho and point your domain mx record at that
[03:02] <th3joker> Bit like google apps when it was free
[03:03] <th3joker> Free mail server
[03:03] <th3joker> No need to reinvent the wheel for a single client
[03:03] <Bitwise> Yes, shauno. th3joker I'm just doing this for the experience.
[03:03] <th3joker> Gotcha
[03:03] <th3joker> Well then that's cool
[03:04] <th3joker> You want an old machine that's on 24/7 for starters
[03:04] <th3joker> Then you can set up your own DNS, nameservers etc, might as well go the whole hog
[03:05] <th3joker> What's your Internet connection like
[03:07] <Bitwise> 4MBps down, 10 Mbps up
[03:07] <shauno> just because I know the pitfalls that are coming, I'd step back and make sure you can get 25 in & out first, it's not unusual for isps to block those, which will sink your ship
[03:09] <shauno> then worry about getting local delivery working first; you shouldn't need dns at that stage, as long as you're sending to/from domains listed in postfix as myhostname / mydestination
[03:09] <Bitwise> I'm going to be the only one sending mail from the server and I don't plan to receive much mail.
[03:10] <shauno> once you're happy it's actually functional, is when you need to start look at making it reachable from the outside
[03:11] <shauno> if you worry about the outside before the inside, you'll end up as what I like to call "nigeria's best friend"
[03:11] <Bitwise> Hahaha. I don't plan to even use this server more than once.
[03:13] <drewery> Bitwise are you gonna spam people ? :)
[03:14] <sheptard> prolly
[03:14] <Bitwise> No. I'm just doing this for the experience. The one time I'll use it is the first time it works and that will just be a test email to my gmail account.
[03:15] <Bitwise> Well I guess I'll show it off a bit so I'll actually probably end up sending more emails.
[03:15] <sarnold> shauno: haha, love that :)
[03:16] <shauno> it'd be funnier if it wasn't true ;)  it seems to be the default configuration for "yay, it works" mailservers
[03:20] <shauno> Bitwise: https://help.ubuntu.com/community/PostfixBasicSetupHowto  actually works quite well.  the only change I'd make is that if you're using no-ip, just use your regular hostname as the MX name.  trying to create a 'mail' subdomain will be more trouble than it's worth
[03:23] <shauno> but seriously, watch /var/log/mail.log and be amazed at how quickly the rest of the world wants to use your server too.  hopefully that'll scare you into making sure they can't
[03:25] <Bitwise> I'm just trying to get the basics down right now. :p
[03:28] <shauno> oh I realise that.  just sharing experience; I'm the only person that uses my mailserver.  a quick grep of my logs shows I've rejected 919 'Relay access denied' in the last 3 hours
[03:28] <Bitwise> Ohmy..
[03:32] <sheptard> shauno: you don't run spamd or anything?
[03:34] <shauno> that's before spamd; that's clients connecting and trying to use my host to send to other hosts.  spamd comes into play once postfix has already decided to accept & deliver
[03:35] <shauno> (which is a good thing.  "you're not in my subnet, go away" is a lot less computationally expensive than spamd)
[04:26] <adam007> Defered connection refused by 127.0.0.1
[04:27] <adam007> what i must do ? i want setup sendmail
[04:27] <adam007> for php mailing..
[04:28] <adam007> any suggestion for this ?
[04:28] <adam007> need help :(
[04:29] <Bitwise> adam007, I'm having the exact same problem.
[04:30] <adam007> Bitwise did you arleady get the solution ?
[04:30] <Bitwise> Not yet.
[04:30] <Bitwise> The immediate problem I'm having right now is logging into postfixadmin
[04:50] <Seven_Six_Two> I'm setting up apache with ssl for a diaspora node, and I see that ssh is listening on 443, blocking apache. Is there a "best practice" for using ssl in this case?
[04:52] <Seven_Six_Two> I don't mind using 22 for ssh, as the port is not forwarded at router
[05:37] <Bitwise> to=<me.mydomain@localhost>, orig_to=<me@mydomain>
[05:38] <Bitwise> It then bounces because status=bounced (unknown user: "mydomain")
[05:39] <Bitwise> I remember reading about this. There is a setting for it. I can't remember what it was. :\
[05:59] <Bitwise> "yay, it works!!!!!!"
[05:59] <Bitwise> "what's a segfault?"
[09:08] <abzmodeel> I'm trying to connect through to a ftpd and it won't work. I'll reinstall distro and always after restart the problem occurs.
[09:13] <TJ-> Do you mean the ftpd service on the server doesn't start automatically?
[15:30] <med___> stgraber, congrats on the LXC 1.0 release. Nicely done.
[15:43] <stgraber> med___: thanks!
[16:22] <wrale> hello world.. I'm seeing an issue on Ubuntu 12.04.4 LTS 64-bit, where SR-IOV virtual functions don't appear as expected in 'ip link show'... Instead of getting the VF(s) below the PF(s), I get what appears to be additional PF(s).  Trying to configure them via 'ip link set DEV vf 0 mac aa:bb:cc:dd:ee:ff'  (or anything similar, appears to not work)
[16:46] <fishcooker> is there any network manager for command line?
[16:48] <wrale> fishcooker: there is the network service.. closet thing to what you're asking is probably editing /etc/network/interfaces.. or perhaps the 'ip' command
[16:48] <wrale> *closest
[16:48] <fishcooker> how about network manager cli?
[16:50] <fishcooker> actually this is wireless network
[16:50] <wrale> i know of none, but there may be something out ther
[16:54] <rickbeldin> someone wrote up using network manager with 'nmcli' :  http://arstechnica.com/civis/viewtopic.php?t=1163023
[16:55] <rickbeldin> example:  sudo nmcli dev status
[16:55] <rickbeldin> DEVICE     TYPE              STATE
[16:55] <rickbeldin> wlan0      802-11-wireless   connected
[16:55] <rickbeldin> eth0       802-3-ethernet    unavailable
[17:00] <fishcooker> is it work when the box will functioning as AP, rickbeldin?
[17:00] <rickbeldin> No idea.  I just tried for the first time on my laptop.  ;)
[17:01] <rickbeldin> I remembered having seen the command in the past, the result of hitting tab-tab too quickly.  ;)
[17:01] <fishcooker> LoL
[17:02] <fishcooker> but thanks for the advice.. cool nmcli
[17:02] <kantlive-> hey all. im using 4TB WD drives with 4K advanced format. 512 logical, 4k physical. should i format starting on 2048 or 4096 sector with parted?
[17:08] <bjf> stgraber, is --dir not supported with lxc-create?
[17:12] <stgraber> bjf: I thought it was when combined with -B dir, but apparently it's not... hallyn ^
[17:14] <hallyn> source says "--dir is not yet supported"
[17:15] <hallyn> You can however put the whole container somewhere else using -P lxcpath...
[17:15] <bjf> hallyn, stgraber thanks
[17:17] <hallyn> Another thing tha twould be nice to have is specifiable block device for -B loop.  We actually used to have it...
[17:17] <hallyn> (but then what we have *today* is less fragile :)
[17:18] <stgraber> hallyn: source says that but the manpage doesn't ;)
[17:19] <hallyn> Hm.  it might be a very simple patch to make it work.
[17:21] <bjf> stgraber, $ sudo lxc-start -n test -d
[17:21] <bjf> lxc-start: Executing '/sbin/init' with no configuration file may crash the host
[17:21] <bjf> stgraber, that's kind of disturbing
[17:22] <stgraber> bjf: if you used -P with create, you also need to pass -P to any of the commands after that
[17:22] <bjf> stgraber, ah!
[17:22] <stgraber> bjf: or you can set it locally once and for all in /etc/lxc/lxc.conf (lxc.lxcpath = /your/path)
[17:22] <bjf> stgraber, that makes obvious sense now that you pointed it out
[17:38] <hazmat> lxc 1.0 !! :-)
[17:47] <hallyn> grr.  so saucy kernel doesn't do nesting at all.
[17:47] <hallyn> smb: do you have a chart somewhere of your findings on supported qemu nesting combinations?
[19:18] <rostam> Hi using ubuntu 12.04 server. For testing propose, I would like configure my system after power cycle run an script after finishing the script reboot the system. This should happen only once and during this time no login is allowed.  Is there anyway I can do this? thx
[19:29] <Daviey> roaksoax: Hey!  I wanted to check if you intentionally used a different licence for iscpy's debian/* ?
[19:31] <roaksoax> Daviey: hey man! I guess I'm just used to doing GPL+ hen i do packaging. I can change that
[19:32] <Daviey> Yeah, unless it was intentional.. it can become a pain to differ.. especially if patches start being in debian/patches/
[19:33] <Daviey> roaksoax: the extended description does sound a bit sales pitchy... If it were me, i'f tone it down a little.. but just personal taste.
[19:34] <hallyn> smb: zul: bug 1283179 is calling for you :)
[19:34] <zul> hallyn:  the libxl one?
[19:34] <hallyn> yeah
[19:34] <roaksoax> Daviey: yeah the description is copy/paste from upstream
[19:34] <zul> yeah i know about it
[19:34] <hallyn> lost our bot?
[19:34] <hallyn> cool then i'll remove it from my inbox
[19:37] <roaksoax> Daviey: ok I uploaded again! Thanks for reviewing this
[19:45] <Daviey> zul: Why are the tests failng in manilla?
[19:46] <Daviey> zul: not a blocker, but you have python-keystoneclient and python-eventlet declared twice as deps.  Not sure on the bundled upstream gz files.. but they seem to be windows only.
[20:03] <Daviey> zul: Presumably you are MIR'ing this, if so - remove the
[20:03] <Daviey> zul: Presumably you are MIR'ing this, if so - remove the || true for the tests, and the duplicate depends.  Accepted.
[20:03] <GrueMaster> rostam: oem-config does something like this (although all of the <alt> tty's are still available).
[20:05] <GrueMaster> Shouldn't be too dificult to create an upstart job that runs before the tty jobs.
[20:07] <hochmeister> can someone help me with this upstart conf: http://pastebin.com/hXS525Ed ?? It's acting strangely. When the logger command in the pre-start stanza is commented, upstart says "Job failed to start". When uncommented I get a classpath related error.
[20:14] <sarnold> hochmeister: sorry, nothing really jumps out, except to notice that $DEFAULTFILE appears to be sourced twice; is that necessary?
[20:51] <Bitwise> hochmeister, CLASSPATH=$CLASSPATH this should be on a separate line.
[20:54] <Bitwise> Just kidding, I'm slow.
[20:55] <Bitwise> hochmeister, Does $OPTS contain $CLASSPATH?
[20:56] <Bitwise> If not then line 35 should actually be java -classpath $CLASSPATH $OPTS com.nextuc.services.ZuoraOrderProcessor >> /var/log/teamup/services/zuora-order-processor.log 2>&1
[21:25] <webfox> Hello folks!
[21:25] <hochmeister> Hey Bitwise. No, $OPTS doesn't contain $CLASSPATH. Just log4j configuration.
[21:26] <webfox> I need a help in order to set a server DNS because I guess it is not working properly here. I am guetting this message when trying to get-apt update : http://pastebin.com/jWaPd750
[21:26] <webfox> Could someone help me please?
[21:26] <hochmeister> adding that --classpath option as you suggested doesn't have any affect.
[21:26] <hochmeister> The job still fails to start.
[21:27] <Bitwise> hochmeister, -classpath not --classpath
[21:27] <hochmeister> no output to the upstart log, only "start: Job failed to start" to stdout.
[21:27] <hochmeister> Bitwise: that's what I meant.
[21:27] <guntbert> webfox: what does   dig br.archive.ubuntu.com   say?
[21:27] <hallyn> smb: ok so certainly saucy kernel - on either precise or kernel userspace - cannot support nested kvm
[21:28] <Bitwise> hochmeister, Open terminal to the proper directory and run (obviously handling the variables accordingly): java -classpath $CLASSPATH $OPTS com.nextuc.services.ZuoraOrderProcessor
[21:29] <hochmeister> Bitwise: this works as expected with and without (CLASSPATH set in the environment) the classpath option.
[21:30] <hochmeister> the issue is that logger command on like 25
[21:30] <hochmeister> when commented the job fails to start. When uncommented, I see the expected message in syslog.
[21:33] <Bitwise> What happens if it's uncommented? Does the Java program run?
[21:33] <hochmeister> no because it's not picking up the variables sourced in the defauls file.
[21:35] <webfox> guntbert: http://pastebin.com/xQz5eKDH
[21:37] <hochmeister> Bitwise: if I specify the classpath as so, it doesn't complain about not able to find the namespace but I need other values from the environment for the java service to run correctly.
[21:37] <guntbert> webfox: really looks like misconfiguration - does  dig br.archive.ubuntu.com @8.8.8.8  work?
[21:38] <webfox> guntbert: no, same thing
[21:38] <Bitwise> hochmeister, Try  reading the DFEAULTFILE between the logger and Java invocation. Another words it will be in there 3 times.
[21:38] <guntbert> webfox: same thing?  try ping 8.8.8.8
[21:39] <GrueMaster> webfox: Is there a proxy that you are missing, or an external DNS server you are not connecting to?
[21:42] <Bitwise> hochmeister, I'll brb for a moment. I just have to reboot my machine.
[21:43] <hochmeister> Bitwise: ok
[21:44] <hallyn> smb: whereas a precise kernel works splendidly
[21:44] <hallyn> (on the same box with precise userspace)
[21:44] <webfox> guntbert: ping: unknown host @8.8.8.8
[21:45] <guntbert> webfox: without @ :-)
[21:45] <webfox> GrueMaster: not sure man.
[21:45] <webfox> guntbert: From 192.168.1.100 icmp_seq=1 Destination Host Unreachable ... and ...
[21:45] <GrueMaster> Try a simple ping to google.com.  If it works, there are other issues.  If not, you aren't getting an external DNS connection.
[21:46] <guntbert> GrueMaster: no DNS seems to be available for webfox now
[21:46] <webfox> yes, no DNS
[21:46] <guntbert> GrueMaster: not even ping
[21:47] <guntbert> webfox: does your machine have a valid IP address?   ip ad    will tell you
[21:47] <webfox> When the Ubuntu 13.10 (not server, the host) sleeps it stops the internet connection I guess and it is causing vBox to crash.
[21:48] <webfox> guntbert: ip ad?
[21:49] <guntbert> webfox: it was meant as a command to be typed - therefore  the spaces
[21:49] <webfox> guntbert: -bash: ip: command not found
[21:50] <hochmeister> Bitwise: funny, with the pre-start stanza commented out it works as expected: http://pastebin.com/2gta6w3J
[21:50] <guntbert> webfox: what version of ubuntu are you running? paste the output of    lsb_release -a
[21:51] <Bitwise> hochmeister, all the variables are also being exported manually.
[21:55] <webfox> guntbert: http://pastebin.com/cbXhtNzB
[21:55] <webfox> guntbert: sorry, was my mistake. I had lost the shh connection.
[21:55] <guntbert> webfox: :)
[21:56] <webfox> guntbert: \o/ what?
[21:56] <webfox> :)
[21:57] <guntbert> webfox: please paste the output of    ip r   right into the channel (should not be more than 3 lines)
[21:58] <webfox> guntbert: http://pastebin.com/921VAKkD
[21:59] <guntbert> webfox: now try ping 192.168.1.254 (no need to paste the result, just tell me if it works)
[21:59] <hochmeister> Bitwise: what's funny is, if I uncomment the pre-start stanza and comment out the test on line 23 it works. If I uncomment that test it fails to start, even though the logger command on line 25 clearly shows that the value of $ENABLED is not zero-length.
[22:00] <webfox> guntbert: it is also happening : "userver@Userver:~$ Write failed: Broken pipe" from time to time and when I try to reconnect it refuses.
[22:01] <Bitwise> hochmeister,  [ -z "$ENABLED" ] && { stop; exit 0; } Shouldn't that be an if block?
[22:01] <Bitwise> The && doesn't short-circuit because ENABLED is true.
[22:02] <guntbert> webfox: did you restart the guest already? what network mode did you select?
[22:02] <webfox> guntbert: bridged
[22:03] <Bitwise> Is this true? https://discussion.dreamhost.com/thread-97931-post-98160.html#pid98160
[22:03] <guntbert> webfox: and 192.168.1.254 is the real router (you use it with your host too?)
[22:03] <Bitwise> Does that mean I MUST have my own domain name to have a mail server?
[22:04] <hochmeister> Bitwise: I just made it so and it works as expected. I adapted that solution from the upstart cookbook: http://upstart.ubuntu.com/cookbook/#pre-start-example-debian-and-ubuntu-specific
[22:04] <webfox> guntbert: it is the router, but it is a wireless router plugged into a ADLS modem
[22:04] <Bitwise> Right now I'm trying to have my own basic little mail server from mydomain.no-ip.org
[22:05] <guntbert> webfox: so your host is conneted only via wifi? did you bridge to the correct interface?
[22:06] <guntbert> *connected
[22:06] <webfox> guntbert: from ping 192.168.1.254 : From 192.168.1.100 icmp_seq=13 Destination Host Unreachable
[22:07] <guntbert> webfox: yes, you said that before (I assume this is from the guest)
[22:07] <webfox> guntbert: no, this computer is connected via cable.
[22:08] <webfox> guntbert: to the wifi router, because the computer has no wifi.
[22:08] <Bitwise> hochmeister, Tbh, I don't know. I'm not the best with this kind of scripting.
[22:08] <hochmeister> it's cool. I got it working now. Thanks for the assist.
[22:10] <webfox> guntbert: yes, the gest looses connection.
[22:11] <guntbert> webfox:  I see - doesn't matter then - lets just confirm a few things: host:ubuntu, using eth0 with 192.168.1.(something) and guest ubuntu server, using eth0 with 192.168.1.100, bridged to eth0 on the host
[22:12] <webfox> guntbert: exactly
[22:15] <guntbert> webfox: looks a little more like a vbox problem after all - step 1) disable the virtual NIC, check if it is correctly bridged, reenable it  step 2) shut down the VM, start it again   step 3) shut down the VM, restart the host, start the VM again
[22:16] <guntbert> of course you need the further steps only if the previous don't help
[22:16] <guntbert> :-)
[22:16] <webfox> what is the virtual NIC?
[22:17] <guntbert> webfox: in the settings of the VM
[22:21] <webfox> guntbert: couldn't find this NIC, but the Network setting seems alright (same I've used in previous versions of Ubuntu). The only consideration I never realized is the Adapter Type I should use, so I always used the default which is Intel PRO/1000 MT Desktop (82540EM).
[22:22] <webfox> I will be restarting everything now..
[22:22] <guntbert> webfox: looks sane
[22:23] <webfox> ok, everything is donne.
[22:28] <webfox> guntbert: here is the Network setting image : http://imgur.com/dFRNWfi
[22:29] <webfox> guntbert: is there anything I should do now?
[22:30] <phunyguy> hey folks, I am trying to follow https://help.ubuntu.com/lts/serverguide/postfix.html, and the portion under "Configuring SASL" mentions to look for "auth default" section in /etc/dovecot/dovecot.conf and change some stuff... problem is, the /etc/dovecot/dovecot.conf that it pulled in, has no such thing.  I am confused... :(
[22:30] <guntbert> webfox: looks good - just test - from the guest   ping the host, ping the router, ping 8.8.8.8 (the latter only if the previous works)
[22:32] <webfox> HOST : From 192.168.1.100 icmp_seq=1 Destination Host Unreachable, GUEST : 4 bytes from 192.168.1.100: icmp_seq=1 ttl=64 time=0.416 ms (OK!), 8.8.8.8 same as the host
[22:33] <webfox> The only working it the Guest
[22:35] <webfox> No, I am laying. Host it is IP is 192.168.1.102 (and is working from ping), the client is working as well (192.168.1.100) but nothing else works. The router 192.168.1.254 or 8.8.8.8
[22:38] <phunyguy> nevermind! :)  https://bugs.launchpad.net/serverguide/+bug/1018548
[22:38] <phunyguy> looks like a bug I can actually fix myself
[22:39] <webfox> So, Host IP = 192.168.1.102 (and pinging), Client IP = 192.168.1.100 (pinging as well, itself), the wifi router (with cable et0) = 192.168.1.254 (unreachable) AND Google DNS 8.8.8.8 (unreachable as well).
[22:41] <guntbert> webfox: if the host cannot reach the router the the VM can neither - sorry  I have to leave now, I was hoping to get you going
[22:42] <webfox> File /etc/resolv.conf is like "nameserver 192.168.1.254" (wifi router) and "search RTKAP"
[22:43] <webfox> guntthe host CAN reach the router.
[22:43] <webfox> guntbert: sorry, the host can reach the router.
[22:43] <webfox> guntbert: ok, thank you very much man!
[22:44] <webfox> Perhaps someone could follow helping me please?
[22:44] <guntbert> webfox: was not really successful - as for the next steps: try to get ping working, DNS comes later
[22:44] <guntbert> g'night
[22:46] <webfox> guntbert: g'night!
[22:46] <webfox> Good point ther, how come ping is not working, even with real IP addresses. :P
[23:05] <Bitwise> postfix: Recipient address rejected: User unknown in virtual mailbox table;
[23:06] <Bitwise> My alias table clearly has a catchall (@mydomain.no-ip.org)