=== FreezingAlt is now known as FreezingCold === markthomas is now known as markthomas|away [01:47] hey anybody up for helping a noob learn how to set up a server to host a website I want to make from home? [01:47] I really don't know where to start === havanami_ is now known as havanamint === bilde2910|away is now known as bilde2910 === kickinz1|afk is now known as kickinz1 [06:29] hilo ubuntu peeples [07:25] Good morning. [07:29] guten tag lordievader [07:30] Hey FilthyMacNasty, how are you? [07:30] having a bad computer day [07:30] Do such things exist? [07:31] yes, I had the sewer manager touch today [07:32] turns out my partner downloaded our image of 12.04 server from a torrent, the entire sources list said trukty not trusty [07:33] Hehe, that's fun ;) [07:33] 4 installs and 4 epic failures [07:34] so now I'm at the brainbleach [07:35] FilthyMacNasty: Download an image from Ubuntu itself. [07:35] .. or at least check the signatures :) [07:36] like I said, my partner did it while I was onsite [07:37] he's in a roll of carpet with some quicklime in a hole outside town [07:37] Was it a reinstall or a new deployment? [07:37] poor carpet [07:37] fortunately 2 were experimental 1 was a laptop for a user and 1 windows box was being contrary [07:39] the windows box has slic 2.1 in the bios, but its hidden by dell and will not authenticate [07:40] so I been playing the "remember hex" dance all day [07:41] turns out its windows 7 pro slic [08:29] hi - just seen 14.04 has another libssl update -> 1.0.1f-1ubuntu2.7 is this just related to the poodle attack or are there other issues fixed also ? [08:30] Perhaps you should read the package changelog? [08:30] Yossarianuk: it also has several DoS fixes: http://www.ubuntu.com/usn/usn-2385-1/ === markour is now known as suspicious_eek [08:35] sarnold_: thanks === zz_satyag is now known as satyag [09:27] <_ruben> what is the "recommended" HA cluster suite these days? all pacemaker/corosync stuff seems to end with howtos for 12.04 === nath|off is now known as nathema === kickinz1 is now known as kickinz1|afk [09:45] _ruben, pacemaker/corosync is still good [09:45] rbasak, re https://bugs.launchpad.net/ubuntu/+source/websocket-client/+bug/1374335 [09:45] are you going to chase that? [09:46] Launchpad bug 1374335 in websocket-client "FFe: Sync websocket-client 0.18.0-1 (universe) from Debian unstable (main), juju-deployer 0.4.2, python-jujuclient 0.18.4" [Medium,New] [10:07] <_ruben> jamespage: most howtos mention the cman package, but that's no longer present in trusty. not sure how critical that is tho [11:04] /w/win 17 [11:31] does /etc/network/interfaces.d/ work yet? [11:32] when i put an interface definition in there it doesnt seem to get picked up [11:32] dine909: I think you need the source-directory interfaces.d line from interfaces(5) [11:39] lordievader: Hey ... u remember the issue i have? ... it was that misterious SHH login problem ... [11:40] lordievader: as mentioned last time i have 3 Servers running on ubuntu 14.04. Only one had this issue. I just installed TYPO3 (cms, php and mysql) on one of the "as working" considered servers and guess what? [11:41] next morning that server had the same behavior: No ssh login possible ... [11:42] and the same messages in console ... [11:44] o_O [11:44] what messages i nthe console? [11:45] one sec ... [11:45] thanks sarnold_ did the trick [11:47] sarnold_: http://picpaste.com/Bildschirmfoto_2014-10-15_um_08.26.21-QunlBobe.1413359450-tQYnSQ0b.png [11:50] sarnold_: all 3 are virtual Maschines running on HyperV .... [11:50] dine9091: nice [11:50] Hi, I have used targetcli on my server to create an iSCSI target. How do I see the clients that are logged in to that target? [11:51] ochoroch: interesting. might be worth using sysctl to snyc, umout, reboot, and check the logs for the start of the issue.. [11:55] sarnold_: a little more verbose please ... :-) === neurotus is now known as Guest63693 [11:56] ochoroch: you can use the magic sysrq key to sync disks (sysrq+s), umount filesystems (sysrq+u), and reboot (sysrq+b), it's not a clean shutdown but shouldn't be too horrible [11:56] ochoroch: then you can check in /var/log/syslog for hopefully some kind of warning about what locked up [11:57] ok .... [11:58] magic-sysrq is disabled on new enough ubuntu [11:58] which is annoying [11:58] argh [11:58] I forgot :/ [11:58] ok ... [11:59] not sure who to blame for that [11:59] It's disabled? I thouhgt it was masked to 176 [11:59] What does "cat /proc/sys/kernel/sysrq" return? [12:01] disabled enough to be disruptive to me on a daily basis [12:01] dasjoe: that's the thing, he can't log in [12:02] looks like syn, umount and reboot should work though [12:02] cwillu_at_work: sub should still work [12:02] yeah, but that's not the only use of sysrq :p [12:03] all sorts of fun telling people "hit alt-sysrq-t and pastebin your dmesg" when trying to troubleshoot deadlocks, only to find that sysrq-t doesn't do anything anymore :p [12:03] Though probably the most important one ;) [12:03] depends on who you're talking to [12:04] and I still think the security benefits are questionable [12:04] physical access outside a kiosk still lets you do the same things [12:04] and kiosks already require many more manual adjustments [12:16] Can anyone recommend a good SSH app for Mac to organise multiple SSH connections - something a bit more structured than iterm or standard terminal? === nathema is now known as nath|off [13:11] ochoroch: Err, please jog my memory... [13:18] lordievader: sorry for the delay ... you want the full story? [13:18] ochoroch: Sure, why not ;) [13:18] i have 3 Ubuntu VMs running on HyperV gen1. .... [13:19] Ubuntu Version is 14.04. It just worked. ... [13:20] i had trouble with the disk failing/read-only every now and then ... [13:21] i was able to fix this by installing all the required drivers and such ... doing a full fsck in recovery mode seems to fix this. [13:21] .. ok... after this i thought "yeah man, i made it" .... [13:23] but the day after i got this when i tried to login: http://picpaste.com/Bildschirmfoto_2014-10-15_um_08.26.21-QunlBobe.1413359450-tQYnSQ0b.png [13:23] thisw is the current ecisting issue, it occurs after roughly 16 hours ..... [13:23] i cant login anymore ... [13:24] ssh promts for password, but does not do a thing. [13:24] Ah, right. That screenshot reminds me. [13:24] when ic force it to turn on/off again it works .... [13:25] ochoroch: Is your machine swapping heavily at that time? [13:26] no ... its bored ... [13:26] ochoroch: Appearantly those type of messages are given when it does not respond. [13:26] no swap at all ... [13:26] Heavy io wait? [13:26] in addition: coz of the fact the other 2 machines seemed to work well for 2 weeks now i decided to put a website (TYPO3 CMS: php and mysql driven) on one, the same happend the this maschine ... [13:27] lordievader: io wait, i'll check my stats ... [13:27] hmmmm, I remember seeing some bug with vmware not writing zeros properly... [13:28] lordievader: io wait seems normal ... [13:28] a -little- iowait is normal, 120 seconds is lunacy :) [13:28] here's the bug I was thinking of, https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1371591 [13:28] Launchpad bug 1371591 in linux-lts-trusty "file not initialized to 0s under some conditions" [Undecided,Confirmed] [13:29] sarnold_: ok ... [13:30] ochoroch: just skim it, see if it looks familiar to you.. [13:30] sarnold_: yep ... [13:33] sarnold_: ok ... i "think" i need to install https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1371591 [13:33] Launchpad bug 1371591 in linux-lts-trusty "file not initialized to 0s under some conditions" [Undecided,Confirmed] [13:34] sorry install these guys; http://people.canonical.com/~arges/lp1371591/ [13:34] i'll give it a go .... [13:35] ochoroch: good luck :) [13:35] just made a "snapshot" .... [13:35] :-P [13:36] but thanks! ... [13:49] jamespage: re: bug 1374335, not currently. Sorry, I should have made it clear that from the juju-quickstart end, they aren't chasing the release team and are OK with it slipping to V. [13:49] Launchpad bug 1374335 in websocket-client "FFe: Sync websocket-client 0.18.0-1 (universe) from Debian unstable (main), juju-deployer 0.4.2, python-jujuclient 0.18.4" [Medium,New] https://launchpad.net/bugs/1374335 [14:02] how do I debug a problem with udev and my ethernet device being named /dev/p4p1 instead of /dev/eth0? /etc/udev/rules.d/70-persistent-net.rules is being ignored, maybe because I installed udevname? [14:02] it's a remote server without console access, so getting this wrong is.. awkward. already had to call in the remote admin once. [14:03] heh? it's not something you debug [14:03] that is *working* as designed :) [14:03] I'm confused as to why it's named /dev/p4p1 when the file in /etc/udev/rules.d says to name it eth0. I think the presence of biosdevname may be causing that, but I'm not sure and am confused by the docs. [14:05] well, uninstall it [14:05] update your initramfs [14:05] and reboot [14:06] patdk-wk; I'm not in a position to do experiments (no console access). is how biosdevname interacts with device naming documented somewhere you know of? [14:06] in the manual :) [14:06] http://manpages.ubuntu.com/manpages/saucy/man1/biosdevname.1.html [14:07] but it's used on startup [14:07] yes thank you, I've read that a few times. It does not describe how ubuntu's boot process uses it. If you don't know the details on how this stuff works, that's OK, was hoping someone here did. [14:08] uninstall it, update initramfs (or it will still exist) [14:08] and reboot, then it's gone, and not renaming [14:08] the issue is, udev works [14:08] device made, eth0, udev sets it to what it should [14:08] then biosdevname changes it again [14:10] is the way biosdevname interacts with udev documented somewhere? it's not the man page you linked. [14:10] I don't know that it even interacts [14:11] Theres a mention of a udev rule in the biosdevname package description. [14:11] hmm, don't have biosdevname installed anywhere [14:11] Pici: thanks! that's helpful. it installs /lib/udev/rules.d/71-biosdevname.rules [14:12] so it's 71 [14:12] so change it to be a lower number? [14:12] since the persistant net naming is 70 [14:14] lordievader: sarnold_: installed kernels and friends as suggested in the ticket. straight away notices ssh login is faster and smoother .... [14:14] i gotta wait till monday and report back! .... [14:15] lordievader: sarnold_: thanks ! ... really appreciate your help.... [14:19] heh, here's a bug saying the Ubuntu Server guide has not been updated for 14.04, with how ethernet device naming works. https://bugs.launchpad.net/serverguide/+bug/1312785 [14:19] Launchpad bug 1312785 in serverguide "ethernet Interface naming has changed" [Undecided,Confirmed] [14:26] read more; I think installing biosdevname caused my device to be renamed in udev and now the old /etc/udev/rules.d/70-persistent-net.rules is being ignored. gonna do some more tinkering and then reboot, will update that Ubuntu bug for the record. I think new 14.04 installs may work differently from do-release-upgrade systems. [14:56] I have `CustomLog ${APACHE_LOG_DIR}/example.com_access.log combined` in an Apache vhost config, however I am only getting the string "combined" logged to the log file on each request. Anyone have any ideas why this would happen? [14:58] is "combined" defined as a log format? Relevant docs: http://httpd.apache.org/docs/2.2/mod/mod_log_config.html#customlog [14:58] MasterRoot24: have you defined what "combined" means in a LogFormat directive? [14:58] Yeah, what NelsonMinar said :) [15:00] rbasak: Possibly not, I'll check! [15:08] NelsonMinar: rbasak: Thanks guys - turns out someone had commented out the respective LogFormat lines. Uncommenting those lines fixed the problem. [15:08] \o/ hooray [15:35] last comment about my udev drama: system rebooted fine. I removed the 70-persistent-net.rules that was being ignored. it appears simply installing biosdevname is enough to cause devices to be renamed next time you boot. [15:39] that was a given :) [15:40] and yet it's not documented anywhere. I updated the bug. === Lcawte|Away is now known as Lcawte === kickinz1|afk is now known as kickinz1 === kickinz1 is now known as kickinz1|afk [17:01] ciao non vedo un disco SCSI su un server. Ho lanciato il comando parted -l e mi ha restituito questo -> http://paste.ubuntu.com/8580059/ [17:02] !it | Guest83311 [17:02] Guest83311: Vai su #ubuntu-it se vuoi parlare in italiano, in questo canale usiamo solo l'inglese. Grazie! (per entrare, scrivi « /join #ubuntu-it » senza virgolette) === havanami_ is now known as havanascode === markthomas is now known as markthomas|away [18:20] Hi, any ubuntu-server admins about? F.Y.I. bug 1380774 has a fix, but note 20141017 Ubuntu Server does NOT have the right version of apt. A release manager needs to check that apt is version 1.09-2ubuntu2 and then ask for a respin. [18:20] Launchpad bug 1380774 in apt "debian-installer does not find kernel" [Unknown,Fix released] https://launchpad.net/bugs/1380774 [21:01] * donspaulding prepares to quote himself from the main #ubuntu channel [21:01] Hi there, I'm using n2n to setup multiple VPN interfaces on a single host. I would like to use avahi/mDNS to broadcast a different hostname tied to a different IP address on each device. [21:01] For example, if I have a 'prodvpn0' interface and a 'testvpn0' interface, each with their own IP (assigned by avahi-autoipd), how can I have avahi-daemon broadcast the mDNS hostname as mybox.testvpn.local on the testvpn0 interface and mybox.prodvpn.local on the prodvpn0 interface? [21:01] I tried creating individual upstart scripts pointing to different avahi-daemon.conf files, but avahi-daemon refuses to start when it detects a running daemon. [21:25] donspaulding: are you using an 'instance' stanza in your upstart jobs? http://upstart.ubuntu.com/cookbook/#instance [21:26] donspaulding: one upstart job, using an env var for the config file & the instance line [21:26] might do the trick [21:26] if it's just an upstart issue [21:26] sounds like it's not though [21:30] semiosis: I hadn't seen instance before, thanks for pointing that out. [21:31] semiosis: but your guess is correct, it's not an upstart thing. In this case, the avahi-daemon process is looking around for other running processes after upstart spawns it, but before it begins functioning. If it finds another avahi-daemon process, it bails out. === xtriz__ is now known as IIT === bilde2910 is now known as bilde2910|away