[00:00] <teward> not really.  needs to work with IMAP clients and SMTP (I know postfix does SMTP), and needs SSL/TLS security if possible
[00:00] <teward> i haven't had the patience to sift through the (virtually nonexistent) postfix documentation on getting this all set up though
[00:02] <Joe_knock> teward if you're strapped for time and have some cash, somebody on freelancer/elance could do it for you.
[00:02] <teward> Joe_knock, maybe it's just my being in a rush to get other things done that's prevented me from actually digging into it
[00:02] <teward> and it's safer to assume I don't have cash :P
[00:02] <Joe_knock> teward, I can understand where you're coming from. Is the mail server part of some web project?
[00:02] <TJ-> teward: postfix with virtual aliases; deliver all email to a domain to a single user account on the server, and then have something like dovecot (IMAP4) for clients to access it
[00:03] <teward> TJ-, it's the dovecot part i get lost at, I think
[00:03] <teward> brb, firewall's being stupid again
[00:03] <teward> for the 10th time today
[00:24] <teward> back
[00:25] <teward> Joe_knock, yeah, its going to be the web server for my entire site, I'm tired of using Google Apps as the backend, would rather move all the services in-house.
[00:25] <teward> s/backend/email backend/
[00:25] <teward> and email's the last thing to move
[00:26] <teward> TJ-, it's getting dovecot set up where I can't find step-by-step documentation...
[00:26] <Joe_knock> teward, maybe you should try my method of having the webserver do all outgoing automated mail and use an external mail hosting server for incoming and response mails
[00:27] <TJ-> teward: Dovecot's config will depend on how you organise the mailboxes, so getting postfix configured in a way that supports the ways dovecot can access mailboxes is the issue, I think
[00:27] <teward> Joe_knock, I could probably do that, I still want to move all the mail for the site off of Google Apps, though, and I do need the response emails.  The other problem is there's a ticketing system on the servers that needs to pull from an IMAP address for emailed-in tickets so meh
[00:29] <Joe_knock> teward google apps isn't the only external mail server option. What does your app/site do?
[00:30] <teward> Joe_knock, the site that's relevant is the ticket system, it's for a site that provides free IRC bouncers
[00:30] <Joe_knock> interesting.
[00:33] <teward> the ticketing system is the one which needs to pull from IMAP for emailed-in tickets
[00:33] <teward> but meh
[00:33] <teward> Joe_knock, I do have other domains I can experiment setting up, I'm trying to move from my gmail.com addresses to my own domain emails anyways so meh :P
[00:35] <teward> Joe_knock, ultimately, this is an experiment, first off, before I move the data to a postfix/dovecot setup.  But I still need a guide for setting it up... :/
[00:41] <sheptard> google?
[00:42] <teward> sheptard, i'd use it if my firewall weren't causing hell...
[00:42] <teward> I guess it's time to replace my network equpiment... again...
[00:53] <ahmadgbg> Hi, i have setup my web server and everything works as it should. The only problem i have is postfix. I cant get it to send mail outside the local area. I am a newbie so its possible that i have done some config error.
[01:09] <Joe_knock> Hey RoyK
[03:47] <Guegs> Is it considered a good idea to update to 14.04 from 12.04 when it is released?
[05:01] <zanzacar> I have an ubuntu-server 13.10 and wanted to install a gui. I was looking at this site https://help.ubuntu.com/community/ServerGUI for tips and it does mention sinstall ubuntu-desktop as an option.
[05:01] <zanzacar> I thought that I would juts check to see if that is still up to date and everything, I thought maybe it might have changed since the site was modified in 2012.
[05:02] <zanzacar> I almost rather uninstall the server and install just unbuntu-desktop but I figured I could just install the desktop and be pretty much the same correct?
[05:04] <zanzacar> eh nm I am still going to have the same problems I would have had with my other desktop. I need another desk/computer
[08:28] <lordievader> Good morning.
[11:46] <pmatulis> morning
[15:41] <zul> coreycb: ceilometer rc1 is out lemme know when you are ready for me to upload it to the archive
[15:43] <Phibs> can anyone tell me why this fails to add the default gw to the interfaces config file: http://p.bsd-unix.net/puaf10jim
[15:57] <jamespage> bug 1293177
[16:19] <zul> jamespage:  have you seen this? http://pastebin.ubuntu.com/7185546/ (trusty-trunk packages)
[16:19] <jamespage> nope
[16:20] <zul> jamespage:  nm...im getting hit by that utf8-bug
[16:20] <jamespage> zul, yeah - that sucks
[16:22] <zul> jamespage:  horizon rc1 is out
[16:25] <patdk-wk> I need to diagnose a trusty upgrade issue :(
[16:26] <patdk-wk> upgraded a machine yesterday, and it hung before grub finished loading :(
[16:26] <cfhowlett> patdk-wk, trust support is in #ubuntu+1
[16:26] <patdk-wk> who was asking for support?
[16:28] <cfhowlett> patdk-wk, pretty sure diagnosing an upgrade issue = support
[16:28] <cfhowlett> !trusty
[16:28] <patdk-wk> heh?
[16:28] <patdk-wk> the first step to making a support issue, is asking for help
[16:28] <patdk-wk> I don't remember doing that
[16:28] <Phibs>  can anyone tell me why this fails to add the default gw to the interfaces config file: http://p.bsd-unix.net/puaf10jim
[16:29] <jamespage> coreycb, zul: pls can you check for fixed ubuntu bugs when prepareing rc1's
[16:29] <patdk-wk> I was making a comment about my *current* activities
[16:29] <jamespage> patdk-wk, that sucks a bit
[16:30] <jamespage> patdk-wk, tbh that's foundations territory
[16:30] <cfhowlett> patdk-wk, until release, installing, cofiguring, troubleshooting and spotchecking of trusty belongs in the #ubuntu+1 channel.  thank you.
[16:30] <Phibs> does anyone in here use preseed with a static net config
[16:32] <patdk-wk> cfhowlett, but I didn't do any of those things, so thanks
[16:34] <jamespage> zul: https://code.launchpad.net/~james-page/horizon/icehouse-rc1/+merge/213509
[16:37] <zul> jamespage:  +1
[16:42] <Phibs> nobody preseeds eh
[16:42] <patdk-wk> phibs, I normally don't build from scratch like that
[16:42] <patdk-wk> but build the machine, snapshot it, then clone
[16:50] <jamespage> zul, tested and uploaded
[16:50] <zul> jamespage:  awesome
[16:50] <jamespage> pending approval from release team - I did not reference the FFe as I don't want to close it now
[16:51] <Phibs> patdk-wk: nod
[17:29] <luminous> is there anything special you can put into /etc/network/intefaces that would say "if you don't have a config for the interface, fallback to dhcp" ?
[17:30] <zul> coreycb:  ceilometer uploaded
[17:30] <coreycb> zul, thanks!
[17:40] <Phibs> is there somewhere to set the default gw in ubuntu, where it is not tied to an interface config stanza ?
[17:45] <sarnold> hey Phibs :) I don't think I've seen a way to do that; how would that help? you've got to have at least one interface configured with an ip in order to know how to route to the gateway anyhow..
[17:46] <henrik> Anyone here interested in bug reports for the trusty serverguide? I found a couple of typos in the lxc section.
[17:48] <pmatulis> henrik: sure, can you open a single bug for them?
[17:49] <pmatulis> http://pad.lv/b/serverguide
[17:50] <henrik> pmatulis: thanks, will create a launchpad account. (and do a quick search to see if anyone spotted the same things..)
[17:51] <pmatulis> henrik: ty bro
[17:57] <vlad_starkov> TJ-: I'm here
[17:58] <TJ-> :)
[17:58] <TJ-> The elevator=deadline is supposed to fix ultra-fast SSDs causing hangs
[17:59] <TJ-> Which made me wonder if you might have also benefited from using "rootdelay=$SOMETHING"
[17:59] <vlad_starkov> TJ-: ahh, I disconnected all drives
[18:00] <TJ-> vlad_starkov: Oh yeah! Well, try it anyhow, we seem to firing shotguns at the barn doors anyhow :)
[18:00] <vlad_starkov> "rootdelay=5" ?
[18:01] <sarnold> why wouldn't you use elevator=none with ssds?
[18:01] <TJ-> I think it's 30 by default, so I usually use 90 when I suspect it might help... If I recall the delay between the last kernel message and the bug warning, is about 21 seconds
[18:01] <TJ-> sarnold: Because there are no SSDs connected :)
[18:01] <sarnold> err, elevator==noop of course..
[18:01] <sarnold> oh :)
[18:04] <TJ-> sarnold: Summary. Supermicro X7DBR-3, no IPMI, with 2x SDD and 4x HDD, new system trying to run even the installer, BUG soft lockup early in boot. In default config (with an initrd) hangs at init-bottom, without initrd, still hangs.   Let me link you to a pastebin of the kernel log, captured over serial, from last Friday: http://paste.ubuntu.com/7169894/
[18:04] <vlad_starkov> TJ-: Here fresh logs with different set of boot params http://paste.ubuntu.com/7186017/
[18:06] <TJ-> vlad_starkov: Was this the system that "noacpi" worked on? And one thing I was trying to do was isolate which ACPI facility was causing the hang?
[18:06] <TJ-> vlad_starkov: We tried "nolapic", did we ever try "nolapic noapic" ?
[18:07] <vlad_starkov> TJ-: yes we did
[18:08] <jamespage> coreycb, zul: was the epoch bump on ceilometer intentional?
[18:08] <jamespage> -queuebot/#ubuntu-release- Unapproved: ceilometer (trusty-proposed/main) [2014.1~b3-0ubuntu3 => 1:2014.1~rc1-0ubuntu1] (ubuntu-server)
[18:09] <zul> jamespage:  er...no
[18:09] <zul> i missed that when reviewing it
[18:09] <vlad_starkov> TJ-: the interesting thing is that with some sets of boot params, for ex. "debug initcall_debug console=tty0 console=ttyS0,115200n8 elevator=deadline", the systems does no fall in hang and responds to Ctrl+Alt+Del
[18:10] <jamespage> zul, lets get it rejected and re-upload with epoch :-)
[18:10] <vlad_starkov> TJ-: "systemd-udevd[1024]: timeout: killing '/sbin/modprobe -qba dm-multipath' [1107]"
[18:10] <zul> jamespage:  yeah
[18:10] <vlad_starkov> TJ-: "dm-multipath" - is a module name?
[18:11] <TJ-> vlad_starkov: Does the "elevator=deadline" cure it completely... with disks connected?
[18:11] <vlad_starkov> TJ-: disks are currenly disconnected, I'd like to boot successfully in minimal environment and then enable devices one by one
[18:13] <coreycb> jamespage, probably a mistake on my part if anything.  I thought it was supposed to be 1.
[18:14] <henrik> pmatulis: took me a second to get the account due to grey listing, but does this look somewhat sensible? https://bugs.launchpad.net/serverguide/+bug/1300369
[18:18] <TJ-> dm-multipath is the device-mapper
[18:21] <pmatulis> henrik: looks good, somebody will take a look
[18:21] <vlad_starkov> TJ-: how about this boot params? console=tty0 console=ttyS0,115200n8 debug initcall_debug noapictimer edd=off clocksource=acpi_pm nohz=off highres=off
[18:22] <henrik> pmatulis: thanks
[18:24] <zul> jamespage:  ok we should be ok for stable/havana on friday
[18:36] <TJ-> vlad_starkov: I'm out of definitive ideas so keep trying anything that looks likely! I'll be afk frequently here (dinner time)
[18:39] <vlad_starkov> TJ-: Hey! It looks I'm booting slowly...
[18:39] <vlad_starkov> TJ-: Confirmed. Server is loading...
[18:40] <vlad_starkov> TJ-: with this set "console=tty0 console=ttyS0,115200n8 debug initcall_debug apic=verbose sysrq_always_enabled ignore_loglevel no_hz=off nmi_watchdog=0 nolapic_timer hpet=disable idle=mwait idle=poll highmem=512m nopat notsc acpi=off pci=nomsi"
[18:40] <sarnold> cripes
[18:47] <vlad_starkov> TJ-: OK, fine. The system looks booted. But I see blackscreen (had to add nomodeset boot param I think).
[18:47] <vlad_starkov> TJ-: Now will try to connect single ssd and boot with it
[18:54] <nszceta> anybody have experience with Dell EqualLogic systems?
[18:57] <Phibs> why the heck is this giving me 500G swap: http://p.bsd-unix.net/pbfxjtdjl  http://www.bsd-unix.net/seitz/jing/2014-03-31_1457.png
[19:00] <TJ-> vlad_starkov: OK, so you'll have to reduce that set to figure out which setting(s) actually have an effect :)
[19:01] <vlad_starkov> TJ-: this will take awhile))
[19:02] <TJ-> vlad_starkov: I know how it goes, but at least, once it works, you have a position to work back from, to isolate the workaround, which might also help to pin down a bug in the kernel eventually
[19:02] <vlad_starkov> TJ-: yep
[19:13] <TJ-> Phibs: I'd suggest "64000 1000000 10000000000 ext4" because you want that partition to really get the maximum size it requests
[19:13] <Phibs> ok...
[19:14] <vlad_starkov> TJ-: when I boot from SSD it boots very slooow...
[19:14] <TJ-> Phibs: The priority value should be in terms of the size values; closer to either size value will make that size more likely to win
[19:14] <Phibs> oh....
[19:14] <TJ-> vlad_starkov: OK ... with all those options I'm not suprised :)
[19:14] <vlad_starkov> TJ-: looks like CPU used intensivly, but no lockup yet
[19:14] <Phibs> I feel preseed partitioning... is retarded then :)
[19:15] <Phibs> but this is good info
[19:15] <Phibs> TJ-: seems to have fixed it thanks :)
[19:15] <TJ-> Phibs: See debian-installer package, file "/usr/share/doc/debian-installer/devel/partman-auto-recipe.txt.gz"
[19:16] <Phibs> THANKS
[19:16] <Phibs> grr caps
[19:16] <TJ-> CAPS are great when ITs raining on you :p
[19:16] <Phibs> I was fine with the auto layout, except using 100% swap
[19:16] <Phibs> 256G ram and 256G... no
[19:33] <vlad_starkov> TJ-: At the very beginning of boot log I have lines like " *BAD*gran_size: 64K 	chunk_size: 512M 	num_reg: 8  	lose cover RAM: -256M". Does it mean that I have corrupted memory?
[19:44] <vlad_starkov> TJ-: can't exit ыскуут
[19:44] <vlad_starkov> TJ-: can't exit screen
[19:45] <vlad_starkov> TJ-: Ctrl+A returns No other window
[19:45] <vlad_starkov> TJ-: oh, exited :)
[19:46] <vlad_starkov> I have to go. Will continue tomorrow I guess...
[19:46] <vlad_starkov> Bye!
[20:16] <zul> coreycb:  nova rc1 is out
[20:16] <coreycb> zul, cool thx
[20:47] <coreycb> zul, jamespage : https://code.launchpad.net/~corey.bryant/nova/2014.1.rc1/+merge/213291
[20:58] <zul> +1
[22:03] <rostam> Hi when 14.04 will be released? thx
[22:03] <thumper> rostam: I think the schedule says 17th
[22:03] <rostam> thumper, thx
[22:15] <Daviey> mdeslaur: Hey, looking at CVE-2013-4311 it was fixed in libvirt (0.9.13-0ubuntu12.5) quantal .. but MITRE don't have it listed as affecting that upstream version.  Which one is incorrect ?
[22:17] <[styx]> Im getting an error stating "Connection closed by [my local host]" this is weird because I was just able to connect last night and nothing has changed since. Anyhelp?
[22:17] <[styx]> Ive tried google but nothing relevant came up. Most was about connecting remotely. Im on my local network still
[22:24] <antarus> utlemming: ping
[22:25] <antarus> utlemming: you notice anything weird with the udev 9.5 bump?
[22:29] <sarnold> Daviey: the debdiff makes it look like the 0.9.13.x branch we had had problems: http://launchpadlibrarian.net/150603203/libvirt_0.9.13-0ubuntu12.3_0.9.13-0ubuntu12.5.diff.gz
[22:36] <Daviey> sarnold: Yeah, i'm trying to work out why it's not listed on the CVE that this version is vuln.
[22:36] <Daviey> (or was)
[22:39] <xperia> hi to all. i created in ubuntu a btrfs partition on a SSD Drive finally for the first time and i would like to know what are the best mount option for this btrfs partition if Read and Write Speed are very Important for working with SSDs ? what for option do you use and where do you change it ?
[22:41] <antarus> ahh nevermind
[22:41] <antarus> unshockingly it was a bug on our end ;p
[22:46] <adam_g> jamespage, any chance of pushing libvirt 1.2.2-0ubuntu7~cloud0 out to icehouse-updates?
[23:42] <Guegs> Anybody know of a good way to schedule FTP downloads from my Offshore server (ubuntu 12.04) to my home server (12.04)? to occur every ~20 minutes or something
[23:42] <sarnold> Guegs: man 5 crontab  :)
[23:43] <Guegs> groovy. thanks.
[23:44] <Guegs> I was considering setting up bittorrent sync, but I need to do some more reading on groups. Just started using any sort of linux distro about a week ok.
[23:47] <Guegs> Wow. That is quite simple. :-)
[23:48] <sarnold> Guegs: to forestall the first question -- it is best to give full pathnames to scripts in cron jobs :)
[23:48] <Guegs> heh, good to know. :-P
[23:49] <Guegs> I don't have the OS installed on my home server yet. I want to get stuff tested in a VM before I do anything too crazy.
[23:50] <sarnold> nice, VM testing is a good idea :)