| MoleMan | is there a way to change a symlink from a command line? I have a samba share, that consists of lots of symlinks to consolitdate files, with tidier names etc, and I've moved the root folder they are in, and would like to be able to bulk update the symlinks without having to manually recreate... any ideas? | 01:40 |
|---|---|---|
| sarnold | MoleMan: There's nothing terribly easy available; the readlink(1) program makes reading a symlink's pointed-to path easy enough, you can pipe those through sed or awk as you need, and use the result in a new ln -sf blah blah command... | 01:43 |
| MoleMan | hmmm, second attempt at google returned this http://superuser.com/a/157832 | 01:44 |
| MoleMan | hopefull will do the job | 01:44 |
| sarnold | MoleMan: it'll be a bit ugly, but starting with something like: for f in * ; do ln -sf $(readlink $f | sed -e ...) $f ; done | 01:45 |
| === furian is now known as shantorn | ||
| === Akendo is now known as Akendo_ | ||
| === chilicuil_away is now known as chilicuil | ||
| lvmer | I'd like to stick with Ubuntu-Server or a linux variation. But what is a good way to RAID 6 - 8x 3TB HDD's for LAN storage. I was planning on software raid & samba & shorewall, etc. But now that I think about it, idk if I've ever seen a RAID 6 option for ubuntu-server. | 03:23 |
| * patdk-lap would like to know, when raid6 wasn't an option | 03:32 | |
| RiXtEr | Hey guys, just installed 12.04.1 Server and my syslog and dmesg are getting flooded with nouveau messages... I don't have any X installed, can anyone tell me why or where to start looking for the answer? | 03:43 |
| patdk-lap | rixter, kernel | 03:49 |
| patdk-lap | nouveau driver :) | 03:49 |
| RiXtEr | patdk-lap, I didn't install anything that should be using nouveau and I am using nomodeset | 03:51 |
| RiXtEr | patdk-lap, shouldn't that turn it off? | 03:51 |
| RiXtEr | (cause it to not load that module) | 03:51 |
| patdk-lap | heh? nomodeset means, don't set a mode, use whatever it was using | 03:51 |
| patdk-lap | did you blacklist nv? | 03:51 |
| RiXtEr | patdk-lap, so is there a grub switch I can use to 'turn it off' ? | 03:51 |
| patdk-lap | dunno | 03:52 |
| RiXtEr | patdk-lap, no... | 03:52 |
| RiXtEr | patdk-lap, blacklist where? | 03:52 |
| lvmer | patdk-lap: raid 6 was always an option... I'm an idiot and I missed it. Do you know about Raid 10? I'm pretty sure I saw it. Also I could RAID 1 2x 30gb OS drives & then partition & raid 6 the other hdds right? 2 independent? | 03:53 |
| === cpg is now known as cpg|away | ||
| === Ursinha_ is now known as Ursinha | ||
| === n0ts_off is now known as n0ts | ||
| === cpg|away is now known as cpg | ||
| === SinZ is now known as SinZ|Tennis | ||
| goddard | anyone know how I can get a live view of connections on my server and maybe even apache error logs? | 06:43 |
| sarnold | goddard: netstat -anp | 06:44 |
| === chilicuil is now known as chilicuil_away | ||
| sarnold | goddard: tail -F /var/log/apache*/error.log | 06:44 |
| goddard | sarnold: nice foo must use that alot | 06:45 |
| sarnold | goddard: I've got a tail -F running right now, funny enough. :) | 06:46 |
| === n0ts is now known as n0ts_off | ||
| === n0ts_off is now known as n0ts | ||
| === n0ts is now known as n0ts_off | ||
| === n0ts_off is now known as n0ts | ||
| === n0ts is now known as n0ts_off | ||
| === n0ts_off is now known as n0ts | ||
| adauthenticate | hi team | 07:28 |
| Tm_T | howdy | 07:28 |
| adauthenticate | i like to ad authenticate with ubuntu server .. kindly help any one !!! | 07:28 |
| adauthenticate | no one is in the chat ??? | 07:30 |
| === SinZ|Tennis is now known as SinZ | ||
| RoyK | morning | 08:32 |
| * RoyK somewhat doesn't like people joining irc, asking a question, waiting two minutes, complaining, and parting... | 08:33 | |
| SinZ | inorite | 08:38 |
| balboah | it's better like me who idles for an hour before replying :) | 08:38 |
| RoyK | :) | 08:42 |
| === Guest30579 is now known as Ard1t | ||
| === mcclurmc_away is now known as mcclurmc | ||
| SysTom | Having some issues booting to grub under hyper-v, I simply get grub loading, and sometimes it'll show the menu for a split second | 11:16 |
| SysTom | Any ideas? | 11:16 |
| Daviey | jamespage: seen https://docs.google.com/document/d/14502bjf6Y8htQSOi3Eo9WdnNfjEWhtMtw37JYIWCwkU/edit?hl=en-GB&forcehl=1 ? | 11:19 |
| === Johnswing is now known as Jswing | ||
| jamespage | Daviey, I had not but its good | 11:53 |
| sarthor | HI, here is my wlan0 Unknown iwlwifi - [phy0].. when i started #airmon start wlan0, my machine says " mon0 is on channel -1, but the AP uses channel 11 " when i apply this command, " sudo aireplay-ng -1 6000 -a C8:D5:FE:0C:6a:74 -h b8:03:05:ca:7c:f0 mon0 " What exactly I need to do, Guide me to the direction, then I will search on the Internet please. Thanks in Advance. | 12:11 |
| radiske | here's the situation, i have a ubuntu server, running firewall and it does all the router to my network... there's one address my network can't resolv... but if i ping this address directly from the server, it resolvs... | 12:20 |
| radiske | someone can help? | 12:23 |
| === cpg is now known as cpg|away | ||
| asac | heyho! | 12:33 |
| asac | sorry if a bit out of touch:). wonder, what gets installed when selecting "basic server install" during install of ubuntu server (http://www.ubuntu.com/download/help/install-ubuntu-server) ... which seed/task/meta-package is it? | 12:33 |
| FauxFaux | I'm guessing it's from tasksel, which is, as I understand it, maintained outside of apt (but I have never looked). | 12:35 |
| sarthor | HI, here is my wlan0 Unknown iwlwifi - [phy0].. when i started #airmon start wlan0, my machine says " mon0 is on channel -1, but the AP uses channel 11 " when i apply this command, " sudo aireplay-ng -1 6000 -a C8:D5:FE:0C:6a:74 -h b8:03:05:ca:7c:f0 mon0 " What exactly I need to do, Guide me to the direction, then I will search on the Internet please. Thanks in Advance. | 12:36 |
| asac | ogra_: hi! do you know ? see a bit further above :) | 12:43 |
| radiske | why my ubuntuserver can resolv an address but my network can't? someone can help? | 12:44 |
| ogra_ | asac, whatever is in the server seed i would guess :) | 12:45 |
| ogra_ | asac, http://bazaar.launchpad.net/~ubuntu-core-dev/ubuntu-seeds/ubuntu.raring/view/head:/server i think | 12:47 |
| asac | ogra_: thats not much :) | 12:49 |
| ogra_ | yeah | 12:49 |
| === niemeyer_ is now known as niemeyer | ||
| ibiris | folks what subset of packages defines the "basic server install" mentioned in http://www.ubuntu.com/download/help/install-ubuntu-server? | 13:55 |
| ogra_ | ibiris, see above | 13:57 |
| ibiris | ogra_: checking from irclogs - thanks | 14:04 |
| === furian is now known as shantorin | ||
| === matsubara is now known as matsubara-lunch | ||
| stgraber | hallyn: can I let you review Dwight's [lxc-devel] [PATCH] Make config api items const ? | 15:08 |
| halvors | Hi! I have no experience setting up a mail server, i want to use postfix and dovecot and as simple authentication as possible, is this guide somthing to choose then? https://help.ubuntu.com/12.04/serverguide/postfix.html and use Mail Stack Delivery | 15:16 |
| halvors | Anyone willing to help me out? | 15:19 |
| halvors | Should i skip the sections "SMTP Authentication" and "Configuring SASL" when using the Mail-Stack Delivery package? | 15:21 |
| ivoks | halvors: yes | 15:49 |
| ivoks | halvors: mail-stack-delivery package configures everything for you; you might want to put your own certificate in place | 15:50 |
| ivoks | sigh | 15:52 |
| ivoks | halvors: yes | 15:54 |
| ivoks | halvors: mail-stack-delivery package configures everything for you; you might want to put your own certificate in place | 15:54 |
| hggdh | rbasak: there? A few Qs... | 15:58 |
| === matsubara-lunch is now known as matsubara | ||
| rbasak | hggdh: pong | 16:04 |
| halvors | ivoks: But what about postfix? | 16:07 |
| ivoks | halvors: what about it? | 16:07 |
| ivoks | halvors: mail-stack-delivery package will pull in postfix and dovecot, configure the link between them and that's it | 16:10 |
| ivoks | halvors: all you have to do is answer couple of questions | 16:10 |
| halvors | ivoks: How is mail delivered to dovecot? On what port? | 16:21 |
| ivoks | halvors: mail from postfix to dovecot is delivered by dovecot's LDA | 16:22 |
| ivoks | (local delivery agent) | 16:22 |
| jamespage | Daviey, did you manage to ascertain what exactly is the plan for the kernel on the 12.04.2 iso for server? | 16:23 |
| jamespage | Daviey, ignore me | 16:23 |
| Daviey | /ignore jamespage | 16:23 |
| halvors | ivoke: I mean remote... | 16:36 |
| ivoks | halvors: smtp, imap, pop3 | 16:36 |
| ivoks | and one can enable submission and smtps very easy | 16:36 |
| ivoks | iirc, imaps and pop3s are enabled by default | 16:37 |
| ivoks | including TLS on imap and pop3 ports | 16:37 |
| iohnizer | Hello, is there anyone willing to help me regarding an issue with the 12.04 installer? I keep getting stuck. | 17:29 |
| radiske | why my ubuntuserver can resolv an address but my network can't? someone can help? | 17:36 |
| zooko | Dear people of #ubuntu-server. We're investigating a failure on a production server... | 17:38 |
| zooko | (We = https://LeastAuthority.com ) | 17:38 |
| zooko | and it *looks* like a "sudo apt-get upgrade" led to the removal of postfix and installation of sendmail, which led to the resulting failure. | 17:38 |
| zooko | Now, I thought that "apt-get upgrade" would never add or remove packages, only upgrade to new versions of currently installed packages. | 17:38 |
| zooko | Am I wrong? | 17:38 |
| sarnold | zooko: your understanding matches mine, upgrade shouldn't make decisions that drastic; only dist-upgrade should, but even that should be relatively rare | 17:42 |
| zooko | sarnold: yeah, weird. This is a 10.04 server. | 17:43 |
| zooko | The auth log and the apt log and everything put together make it look like I ran "sudo apt-get update && sudo apt-get upgrade && shutdown -r now". | 17:43 |
| zooko | Which is a thing I normally do on an Ubuntu LTS server without much hesitation. | 17:43 |
| sarnold | indeed, I'd think nothing of that | 17:44 |
| zooko | But then the apt log says "Okay, so you want to uninstall postfix and install sendmail? Fine!", and goes and does that. | 17:44 |
| zooko | Weird. | 17:44 |
| zooko | I have one non-Ubuntu package installed -- nginx as packaged by the nginx maintainers -- but I looked at its deps and it didn't look like it would interact wil mail to me. | 17:44 |
| * zooko looks again. | 17:44 | |
| sarnold | (tohugh in my case it's usually apt-get update && apt-get -u dist-upgrade && pm-suspend or similar :) | 17:44 |
| zooko | Yeah, even dist-upgrade I generally consider "safe" on Ubuntu LTSes. :-{ | 17:45 |
| sarnold | zooko: walking way out on a limb here, what are the chances someone used dpkg's --set-selections or dselect or similar to change the intended package disposition? | 17:48 |
| sarnold | zooko: .. or aptitude or software center? | 17:48 |
| zooko | aptitude, yes. Someone had been running aptitude in the past. | 17:48 |
| * zooko looks at aptitude log | 17:48 | |
| zooko | But I don't see any mention of sendmail, postfix, sensible-mda, or "mail" anything in the aptitude logs. | 17:50 |
| zooko | Okay, I give up on explaining it for now. | 17:50 |
| zooko | Thanks for the chat, sarnold! | 17:50 |
| sarnold | zooko: good luck! | 17:50 |
| smoser | jamespage, so the errors we're seeing on raring are bug Merge-Method: { "dicts": "overwrite", "lists": "append" } | 17:53 |
| smoser | oops | 17:53 |
| smoser | https://bugs.launchpad.net/ubuntu/+bug/1078926 | 17:53 |
| uvirtbot | Launchpad bug 1078926 in ubuntu "raring instance failed to find EC2 datasource" [High,Confirmed] | 17:53 |
| ikonia | window 14 | 17:54 |
| ikonia | oops | 17:54 |
| smoser | stgraber, around ? | 17:59 |
| stgraber | smoser: yep | 18:00 |
| smoser | stgraber, https://bugs.launchpad.net/ubuntu/+bug/1078926 | 18:04 |
| uvirtbot | Launchpad bug 1078926 in ubuntu "raring instance failed to find EC2 datasource" [High,Confirmed] | 18:04 |
| smoser | see the final comment, i'm wondering if you have help with my 'c' there. | 18:04 |
| smoser | and if you might have hints on 'b' although i think that might be more of a slangasek thing. | 18:04 |
| iohnizer | During the install from USB I get the message: main-menu[334]: DEBUG: resolver (libc6-udeb): package doesn't exist (ignored). The same thing goes for libnewt0.52. I have been trying to google my way around this issue, but can't seem to find any helpful answer. Any ideas? | 18:05 |
| stgraber | smoser: yeah, b) would be for slangasek. For c), it's a bit weird. Usually what should happen is that the kernel or 'udevadm trigger' will emit a net-device-added event on the netlink bus, which will then be forwarded to upstart triggering the network-interface job and ultimately calling ifup | 18:07 |
| stgraber | smoser: as eth0 is brought up by the fallback job, this most likely means that upstart never received the net-device-added event from udev | 18:08 |
| stgraber | smoser: which can be explained by either a kernel bug causing the uevent to simply never be emitted or because somehow your instance doesn't call udevadm trigger so never "fakes" the event if it was emitted before the udev bridge started | 18:08 |
| stgraber | that's what I'm seeing after a quick read, there may be some weirder things happening | 18:09 |
| smoser | stgraber, right. it could cause that. | 18:09 |
| smoser | but the kerel driver on kvm should be solid | 18:09 |
| smoser | why would the image never call udevadm trigger ? | 18:10 |
| smoser | (where woudl / should that occur) | 18:10 |
| smoser | but it doesn't fail all the time | 18:10 |
| smoser | so that would seem to not be valid. | 18:10 |
| stgraber | no idea why it wouldn't trigger but I'm never too sure of what weird changes you have in your environment with your cloud init jobs :) | 18:11 |
| stgraber | it should be triggered from udevtrigger.conf | 18:11 |
| stgraber | so requires udev to have been started by upstart which itself depends on virtual-filesystems | 18:11 |
| smoser | which should occur now in parallel with mounted / | 18:12 |
| smoser | so we shouldn't hit a race there on that. | 18:12 |
| stgraber | you may want to add a generic job to confirm that net-device-added is emitted | 18:13 |
| stgraber | something along the line of: | 18:13 |
| stgraber | start on net-device-added | 18:13 |
| stgraber | task | 18:13 |
| stgraber | script | 18:13 |
| stgraber | echo $INTERFACE >> /run/debug-network | 18:13 |
| stgraber | end script | 18:13 |
| stgraber | this will then be called by upstart whenever a network interface appears and it gets the event from udev. If you don't see eth0 in /run/debug-network, then you have a kernel/udev problem | 18:14 |
| smoser | stgraber, well static-networking-up does run | 18:21 |
| smoser | hm.. i have to look mroe. | 18:21 |
| smoser | well, i'll poke more later. | 18:21 |
| === cmagina_ is now known as cmagina | ||
| stgraber | smoser: sure, because the networking.conf job will eventually trigger and call ifup -a which will bring eth0 up and emit the required events, eventually triggering static-network-up, but I'm assuming that by the time it does that, it's way too late for you | 18:23 |
| stgraber | and networking.conf is really meant as a last fallback for when the event based bring up doesn't work, so depending on it would be wrong. (Except for virtual devices without any physical children where it's the only way they'll ever be brought up) | 18:24 |
| === yofel_ is now known as yofel | ||
| yolanda | hi zul, is there any issue where i can help | 18:40 |
| yolanda | ? | 18:40 |
| zul | yolanda: umm...not right now... | 18:40 |
| zul | yolanda: you might want to go through http://bugs.launchpad.net/ubuntu/+source/nova and see if there is anything you can fix or getting stale | 18:41 |
| yolanda | ok, i'll take a look | 18:41 |
| zul | yolanda: also you can check jenkins and look for broken builds and fix them up if you want | 18:42 |
| === Malediction_ is now known as Malediction | ||
| === n0ts is now known as n0ts_off | ||
| yolanda | zul, although this change is assigned to other person? | 18:59 |
| zul | which one? | 18:59 |
| yolanda | i was talking generally, looking at changes that were marked as failed | 19:02 |
| hallyn | stgraber: yup. boy there's a huge set of patches. what happened no lxc-devel? :) | 19:02 |
| stgraber | hallyn: well, looks like having a reactive upstream attracts developers, who'd have guessed ;) | 19:04 |
| hallyn | except this is sort of like a financial bubble | 19:05 |
| hallyn | we're gonna build up more and more patches ingithub, 0.9 won't get released, and people will be furious | 19:06 |
| halvors | I have a mail server, i realize that everybody can send email thru it, but how test it? | 19:08 |
| hallyn | stgraber: have you by chance done a build with that patch applied? | 19:09 |
| stgraber | hallyn: nope | 19:10 |
| halvors | I've used Mail-Stack Delivery, it's ok for me that other domain's use it, as long as they login using valid username and password :) | 19:11 |
| halvors | What is the most compatible webmail for the Mail-Stack Delivery service? | 19:12 |
| halvors | I would prefer a package that does all the needed apache2 configuration :) | 19:12 |
| === mcclurmc is now known as mcclurmc_away | ||
| sarnold | halvors: google around a bit, someone used to offer a service to check your mail server; you'd telnet to their machine from your mail server and they'd run 20-odd tests and show you the results in the telnet session.... | 19:14 |
| halvors | Does this look ok? | 19:17 |
| halvors | http://mxtoolbox.com/SuperTool.aspx?action=smtp%3amail.test.halvors.org | 19:17 |
| yolanda | jdstrand, you there? i wanted to do some work on https://bugs.launchpad.net/ubuntu/+source/nova/+bug/945177 | 19:17 |
| uvirtbot | Launchpad bug 945177 in nova "not lintian clean" [Medium,Confirmed] | 19:17 |
| halvors | sarnold: Looks ok? | 19:19 |
| sarnold | halvors: it's a pity their tests are as short as they are, but it's definitely a good start | 19:19 |
| halvors | I also have this problem that i think my ISP is blocking port 25, at least i cannot connect to my mail server from outside, is there a simple way to confirm that? | 19:20 |
| sarnold | halvors: 220 halvors.org ESMTP Postfix (Ubuntu) | 19:21 |
| sarnold | halvors: my isp lets me through just fine | 19:21 |
| halvors | hmm... | 19:21 |
| halvors | Sure that's not on port 587? | 19:21 |
| halvors | Cause thru VPN i cannot connect using a SMTP client. | 19:22 |
| sarnold | halvors: "telnet mail.test.halvors.org smtp" -- yeah :) | 19:22 |
| halvors | Does not work on windows right? | 19:22 |
| sarnold | halvors: no windows here to test with, but I'd hope it would work fine there, too | 19:23 |
| halvors | I doesn't it only take 1 parameter (The hostname) | 19:27 |
| halvors | I get this in log when trying to use SMTPS (Port 465) via VPN. What does it mean? | 19:30 |
| halvors | Nov 27 20:28:57 halvors-server postfix/smtpd[5540]: warning: hostname ip-130.privitize.com does not resolve to address 46.246.31.130 Nov 27 | 19:30 |
| halvors | 20:28:57 halvors-server postfix/smtpd[5540]: connect from unknown[46.246.31.130] | 19:30 |
| halvors | sarnold: :) | 19:30 |
| sarnold | halvors: dns provides both forward and reverse lookups | 19:31 |
| sarnold | halvors: to get an ip from www.google.com, and to turn that ip address back into www.google.com | 19:31 |
| sarnold | halvors: getting forward dns is extremely easy; every dynamic dns provider does that | 19:32 |
| halvors | Yeah. But what do you mean? | 19:32 |
| sarnold | halvors: getting reverse is different; that's tied to the ip address closely. whoever actually _owns_ that IP address provides the reverse dns. | 19:32 |
| halvors | I know. | 19:32 |
| halvors | But what does that have with the postfix server to do? | 19:33 |
| sarnold | halvors: dropping email from servers that do not have matching forward <-> reverse data is a very common anti-spam step | 19:33 |
| halvors | sarnold: May you try telneting my server again, just to be sure that it's not my isp that is responding. Now you should not get answer on port 25... | 19:34 |
| halvors | Ah ok. | 19:34 |
| halvors | So that means SMTPS (Port 465) actually works | 19:34 |
| halvors | But i can send mail from that ip using STARTTLS (587) | 19:35 |
| sarnold | halvors: 25 is dead (network unreachable! rather than "connection refused"), 465 works fine | 19:35 |
| halvors | Iknow. | 19:35 |
| halvors | What about 587? | 19:36 |
| sarnold | 587 works too | 19:36 |
| halvors | hmm, so wounder why i can't use port 25... | 19:36 |
| halvors | BTW: it should be back up now :) | 19:36 |
| halvors | sarnold: You've got comcast as ISP? | 19:38 |
| halvors | sarnold: If so i see your connection in /var/log/mail.log :) | 19:38 |
| sarnold | halvors: yes :) woot. | 19:41 |
| halvors | And you get thru on port 25, 465, and 587? | 19:42 |
| sarnold | halvors: yes | 19:43 |
| gucki | hi guys | 19:44 |
| gucki | anybody knows what's the maximum length of a network interface name? | 19:44 |
| jdstrand | yolanda: I am here | 19:58 |
| yolanda | hi jdstrand, i wanted to do some clean on that package, but i wonder if that bug is too old? changelog numbers don't match | 19:59 |
| halvors | sarnold: Shouldn't i be able to send mail not using STARTTLS or SSL/TLS, on port 587? | 19:59 |
| halvors | Or do postfix only allow Plain on Port 25, STARTTLS on Port 587 and SSL/TLS on Port 465? | 20:00 |
| sarnold | halvors: no idea :) | 20:01 |
| escott | halvors, i wouldn't be surprised if it refused. many don't run anything on 25 so they would be unhappy is something unsecured just got shoveled through 587/465 | 20:03 |
| halvors | Ah ok, but is there a way to accept STARTTLS and SSL/TLS on port 25? | 20:04 |
| raub | halvors: I believe you can control that in master.cf | 20:05 |
| escott | halvors, whether or not it is possible depends on the SSL/TLS handshake. if it can begin with a simple HELO maybe | 20:05 |
| jdstrand | yolanda: just download the current binaries and run lintian on them. if it is clean, close the bug, if not, then add a comment to the bug on what is still not clean | 20:09 |
| yolanda | jdstrand, essex stable ones? | 20:10 |
| jdstrand | yolanda: well, it would need to be against raring (ie, grizzly) at this point | 20:10 |
| yolanda | oh, ok | 20:10 |
| yolanda | understood! | 20:10 |
| jdstrand | yolanda: the bug is targeted at 'nova' not a particular release of ubuntu | 20:10 |
| jdstrand | and we don't usually fix linitan errors in stable releases | 20:11 |
| yolanda | ok, so when it's that "nova" target it intends to be to the latest version? | 20:11 |
| halvors | escott: Mostly thinking about my customers, since they think everything that is outgoing email is port 25. | 20:12 |
| halvors1 | sarnold: Seems like SSL/TLS works, just that client need to have a reverse dns record. But that's not common when using IPv6... | 20:18 |
| halvors1 | ' | 20:18 |
| halvors1 | So i got this problem... | 20:18 |
| sarnold | halvors1: there may be a way to disable that check | 20:18 |
| halvors1 | sarnold: Everyone got reverse for IPv4, but that's not common on IPv6... | 20:24 |
| hallyn | zul: did you have any changes heading to raring libvirt soon, or sh ould i just push my debdiff for not setting up default autostart when 192.168.122.1 exists right now? | 20:25 |
| === Corey_ is now known as Corey | ||
| halvors1 | sarnold: May this be it? smtpd_senicder_restrtions = reject_unknown_sender_domain | 20:33 |
| sarnold | halvors1: seems like it should be better named, if that's the one. I'd say keep looking... | 20:33 |
| halvors1 | sarnold: You mean like this: smtpd_sender_restrictions = reject_unknown_sender_domain | 20:34 |
| sarnold | halvors1: oh, hah, thta does look like the point of that setting: http://serverfault.com/questions/319842/postfix-client-ip-reverse-dns | 20:36 |
| uvirtbot | New bug: #1063177 in mysql-5.5 (main) "update MySQL 5.5.28" [Wishlist,Fix released] https://launchpad.net/bugs/1063177 | 21:05 |
| uvirtbot | New bug: #1079897 in walinuxagent (main) "walinuxagent mangles server identity and access on upgrade" [Critical,Confirmed] https://launchpad.net/bugs/1079897 | 21:05 |
| uvirtbot | New bug: #1083598 in vsftpd (main) "package vsftpd 2.3.2-3ubuntu4.1 failed to install/upgrade: ErrorMessage: Package is in a very bad inconsistent state - you should reinstall it before attempting configuration." [Undecided,New] https://launchpad.net/bugs/1083598 | 21:05 |
| uvirtbot | New bug: #1083719 in open-vm-tools (multiverse) "open-vm-dkms 2011.12.20-562307-0ubuntu1: open-vm-tools kernel module failed to build" [Undecided,New] https://launchpad.net/bugs/1083719 | 21:05 |
| uvirtbot | New bug: #1083407 in samba (main) "subtle upstart race condition with rc-sysinit scripts " [Undecided,New] https://launchpad.net/bugs/1083407 | 21:06 |
| uvirtbot | New bug: #1083513 in php5 (main) "installation IDN on PECL at PHP > 5.3 fails" [Undecided,New] https://launchpad.net/bugs/1083513 | 21:06 |
| uvirtbot | New bug: #1083542 in maas (main) "sudo dpkg-reconfigure maas-cluster-controller breaks" [Undecided,New] https://launchpad.net/bugs/1083542 | 21:06 |
| zul | hallyn: i have nothing | 21:08 |
| uvirtbot | New bug: #1083155 in python-webob "Unable to set Content-MD5 header when using chunked transfer encoding" [Undecided,New] https://launchpad.net/bugs/1083155 | 21:08 |
| uvirtbot | New bug: #1083158 in zookeeper (universe) "package does not put startup script in /etc/init.d/ and also lacks an rc script" [Undecided,Invalid] https://launchpad.net/bugs/1083158 | 21:08 |
| hallyn | zul: ok, pushing. hopefully i haven't missed some upgrade corner case... | 21:22 |
| === cpg|away is now known as cpg | ||
| === cpg is now known as cpg|away | ||
| timrc | Hello, I've created a lucid container on a precise host, and I notice when I ssh into the container I get some locale problems that I don't get when I use the "console". As an example, when I run bzr I get this: http://pastebin.ubuntu.com/1392843/ | 21:37 |
| timrc | Here is what I get when I ask Python what it thinks, http://pastebin.ubuntu.com/1392853/ | 21:38 |
| timrc | sudo locale-gen en_US.UTF-8 seems to "fix" the problem | 21:40 |
| hallyn | what *is* that thing? | 21:40 |
| hallyn | ANSI_X3.4-1968 that is | 21:41 |
| timrc | no clue, to be honest | 21:41 |
| hallyn | i'm building a lucid container right now... | 21:41 |
| hallyn | stgraber: ^ look familiar at all? | 21:41 |
| stgraber | timrc: apt-get install language-pack-en | 21:42 |
| stgraber | hallyn: we don't have any langpack installed by default, that's the cause of this | 21:43 |
| stgraber | hallyn: ssh passes your locale along, if the target server doesn't have it, then you get that kind of problem | 21:43 |
| timrc | ah | 21:44 |
| stgraber | hallyn: when logging in on the console, you just get the good old C locale, so nothing should attempt to do UTF-8 and you're "fine" (until you try to print anything that actually requires unicode) | 21:44 |
| hallyn | aha :) yeah i just got as far as reproducing it. interesting | 21:44 |
| stgraber | anyway, the fix is to have the template creation script to always install language-pack-en and any other locale you have outside the container | 21:44 |
| hallyn | stgraber: thanks | 21:44 |
| stgraber | I have a workitem to fix that for this cycle | 21:44 |
| timrc | stgraber, I can confirm that does indeed work :) | 21:44 |
| hallyn | really? i don't remember that int he blueprint - excellent :) | 21:45 |
| timrc | stgraber, thanks for the assistance | 21:45 |
| stgraber | timrc: np | 21:45 |
| stgraber | hallyn: I dumped a bump of LXC related workitems I had around after the session, that's one of them :) | 21:45 |
| timrc | While I'm here, I'm trying to run livebuild inside of my lucid container and I keep failing at: W: Failure trying to run: chroot /path/to/chroot mount -t proc proc /proc -- When I try running the command manually I get a little more information, specifically that the the block device is write-protected, mounting read-only and that I can't mount block device proc read-only | 21:51 |
| timrc | hallyn, stgraber: have either of you run into this? | 21:51 |
| stgraber | timrc: that'd be apparmor preventing you from mounting proc or sys in the container for security reason | 21:52 |
| stgraber | timrc: the easiest way to bypass this is to edit your container's configuration and uncomment the lxc.aa_profile line, setting it to unconfined | 21:52 |
| stgraber | timrc: then restart the container and you won't have the problem again | 21:53 |
| timrc | stgraber, is this a recent change? I don't recall having this issue a few weeks ago | 21:53 |
| hallyn | not that recent | 21:53 |
| timrc | hrm ok | 21:53 |
| stgraber | timrc: however, after doing that, your container will be allowed to mount unprotected copies of proc and sys which can damage your host, so don't give access to that container to someone you don't trust with the host | 21:53 |
| hallyn | actually, i was getting failures in debootstrappin ga debian unstable - on my HOST - at the mount proc part | 21:53 |
| hallyn | but i didn't have time to delve deeper | 21:53 |
| stgraber | timrc: that security has been introduced as of Ubuntu 12.04 | 21:54 |
| timrc | stgraber, That'd be my wife, thanks :) | 21:54 |
| hallyn | "she's crafty" ? | 21:54 |
| * hallyn looks for the beastie boys cd... | 21:54 | |
| timrc | hallyn, she has a talent for breaking things.. she should be a product QA consultant | 21:55 |
| hallyn | :) | 21:56 |
| timrc | So this is a first... "disable app armour"... it's usually "disable selinux" | 21:56 |
| stgraber | timrc: if this was some kind of public service, I'd have recommended creating a profile specific for this container under /etc/apparmor.d/lxc/ then set that profile as lxc.aa_profile. But if it's just for local hacking, it's probably not worth the time :) | 22:00 |
| timrc | local hacking only, so yeah | 22:01 |
| timrc | I honestly didn't give app armour any consideration, sadly, until you mentioned it though | 22:01 |
| timrc | I need to be more cognizant of it in the future | 22:01 |
| === ibiris is now known as ibiris|afk | ||
| === cpg|away is now known as cpg | ||
| sarthor | HI, Before 12.04, where ever I used to configure the dns farwarders for my network in the ubuntu-server, I used to edit /etc/bind/named.conf.options and used to write there my dns, and in the resolve.conf was somthing like, "nameserver 127.0.0.1" Now the /etc/resolv.conf says, Do not edit by hand, Where to put that forwarders now?? please help. | 23:20 |
| === pug is now known as Guest29428 | ||
| sarnold | sarthor: you could either have your dhcp server advertise the dns you want clients to use or you could stuff the directives in /etc/resolvconf/<something> | 23:21 |
| === Guest29428 is now known as DaveR | ||
| === Gallomimia_ is now known as Gallomimia | ||
| uvirtbot | New bug: #1083824 in php5 "php5 UTF-8 bug" [Undecided,New] https://launchpad.net/bugs/1083824 | 23:56 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!