=== freeflying is now known as freeflying_away === freeflying_away is now known as freeflying === freeflying is now known as freeflying_away === freeflying_away is now known as freeflying [05:19] index.php pages are not loading by default. instead i am getting a directory listing [05:19] what can i do to fix this? [06:10] everything was working fine in my lampp stack, untill dpkg configure mythtv-common === holymac_ is now known as holymac [09:47] hi [09:48] is possible to "reconfigure" automatic-updates in ubu12.04server ? [10:01] sacarde: try "sudo dpkg-reconfigure unattended-upgrades" [10:03] rbasak, I try, but I view only 2 choises [y] [n] [10:04] sacarde: what do you expect? [10:04] during installation I view: [10:05] 1 no upgrade [10:05] 2 automatic upgrade [10:05] 3 using landscape [10:06] sacarde: and which of those are you trying to do now? [10:06] 1 no upd [10:06] sacarde: "sudo dpkg -P unattended-upgrades" - you don't want the package at all in that case. [10:08] ah [10:08] and if I will wont: landscape ? [10:11] sacarde: for Landscape I think you just run "landscape-config" [10:11] With sudo I presume [11:07] ps aux shows "/sbin/mdadm --monitor…" running — this emails root if a raid drive is malfunctioning, right? but I never log in as root, so how would I know that root was emailed? [11:12] jdimatteo: because you let root mail forward to a user who reads such mails [11:13] geser: how do I do that? does ubuntu automatically forward such emails to the admin user set up during install? [11:17] I am having lots of trouble getting the tftpd-hpa to work ... it installed find ... but it doesn't seem to start... netstat shows nothing, and there is nothing in syslog as to why... can anyone help? [11:34] ok, I think I can set up email forwarding with /etc/aliases — thanks for the help geser! === freeflying is now known as freeflying_away === freeflying_away is now known as freeflying === gary_poster|away is now known as gary_poster === zz_Gurkenmaster is now known as Gurkenmaster === d1b_ is now known as d1b [13:38] hi [13:38] i receive many of these [Wed Nov 13 14:29:03 2013] [notice] child pid 29367 exit signal Segmentation fault (11), possible coredump in /etc/apache2 [13:39] i just enabled ssl module [13:39] how can I debug this? [13:48] mdeslaur: libvirt 1.1.4 uploading(ed) [13:48] zul: \o/ thanks [13:49] mdeslaur: you can stop nagging me now ;)( [13:53] zul: I'll find something else to nag you about, don't worry :) [13:53] mdeslaur: huzzah [14:51] at some point last week I set a proxy for wget, but now I can't figure out where the proxy is set and how to turn it off. I typed 'set | grep -i proxy' in bash and I get nothing. [14:51] nevermind, it was in .wgetrc [14:55] hiya. how do people rsync stuff between ubuntu servers with disabled root access? in centos, I could rsync /usr/local, for example. here -- I can't. [14:56] slava_dp: this should help: http://crashingdaily.wordpress.com/2007/06/29/rsync-and-sudo-over-ssh/ [14:57] You can still use key auth with the root account. [14:58] Not sure if sudo rsync is so much better than root rsync, aside from leaving a slightly more distinct entry in the auth log. [14:59] when start vm windows (XEN) [14:59] Xenctrl.Error [ memory 6796760 KiB free; to be scrubbed 0 KiB; total 7893 MiB]: 1: Operation not permitted [15:00] ubuntu server 13.10 [15:00] http://www.linuxquestions.org/questions/linux-server-73/xen-start-vm-windows-%3D-problem-help-please-4175484442/ [15:15] help please [15:17] I'm running a webserver on ubu 12.04 server and I'm running into an error when testing imap configuration: http://i.imgur.com/zxullwr.png I understand that no firewall is enabled by default in ubu 12.04 server. Any suggestions why it is timing out? I can ping imap.gmail.com from the machine, avg 30ms [15:27] http://www.linuxquestions.org/questions/linux-server-73/xen-start-vm-windows-%3D-problem-help-please-4175484442/ [15:34] mjeanson: thank you. --rsync-path='sudo rsync' works. [15:35] hi, i got a problem on a 12.4.3 server install, how can i fix this : apt setup udeb succeeded but requested to be left unconfigured (is there a better place for my question ?) [16:08] hi, i got a problem on a 12.4.3 server install, how can i fix this : apt setup udeb succeeded but requested [16:08] to be left unconfigured (is there a better place for my question ?) [16:16] how to install module blktap (Kernel) ? [17:23] I received an email from my vps provider about weird smtp traffic coming out of my container. Something like 100 attempts to send per 10min. I'd like to find out where those emails or attempts to email are coming from. I checked /var/log/mail.log and the most recent event was Oct 2. :/ === Artpicre is now known as Artpicre__ [18:13] Can anybody recommend some good server/client backup software that's cross-platform? Was hoping to find a simple client where all the config stuff comes from the server. === Artpicre_ is now known as Artpicre [18:15] kpettit: investigate bacula, amanda, rsnapshot, rdiff-backup [18:16] I don't even care if it's commercial or not. Just want something easy to use. [18:16] Have you use bacula or amanda before? [18:17] I just noticed the saucy kernel/headers are in the repo. anyone install them on a 12.04 server yet? any issues? [18:18] kpettit: I've never had enough systems to see the benefits of the more complicated systems compared to rsnapshot [18:20] I'm with ya. I'm getting enough customers that don't have backup's and I have to figure a linux/windows solution that will work on desktop and servers. Normally I just do rsync/tar/etc, but I need something I can configure on a server rather than each individual client. I'm reading into amanda/bacual but I remember trying them before and had problems. Hopefully I'll have better luck or find something better [18:26] I don't like bacula setup [18:26] I want to try amanda though [18:27] I'm looking at it now. [18:27] It's sad how complex they are to setup [18:27] yep [18:27] personally, I just want to offload my backups onto tape [18:28] have a huge 48tape lib to use [18:28] nice! [18:28] I've got a Drobo I'll be using i think. [18:56] http://www.linuxquestions.org/questions/linux-server-73/xen-start-vm-windows-%3D-problem-help-please-4175484442/ [18:56] sarnold: how does rsnapshot relate to rdiff-backup? (I've used the latter - though these days I just forget about incremental and do rsync) [18:57] hallyn_: funny enough, I've only used rsnapshot and not rdiff-backup, and I can't recall the reasons I had when deciding.. [18:58] sarnold: ok :) [18:59] hallyn_: I learned to love incremental when I accidentally screwed up something when releasing an update... [18:59] hallyn_: I needed to know some details of how the USN database management worked with multiple CVEs, and I got lucky, an older temporary shell script with multiple CVEs was still in my incrementals.. [19:00] it saved probably an hour of reading source code or worse yet, waiting until the next day for others to come online to help me fix my mistake ;) hehe [19:00] sarnold: if i keep local backups i'll go back to incremental, but right now my backups are only for disaster recovery (fire, flood, earthquake, war...) [19:01] i actually started my newest script using rdiff-backup, but for some reason - forget what - changed my mind [19:02] hallyn_: ah, see, I need to start doing some off-sites. I'm contemplating amazon glacier for large and low-priority, and tarsnap for small and important.. [19:03] Damn, I just looked at a few commercial solutions. They are freaking expensive [19:03] "glacier" sounds liek it will take a thousand years to reach its destination :P [19:03] like* [19:03] http://www.linuxquestions.org/questions/linux-server-73/xen-start-vm-windows-%3D-problem-help-please-4175484442/ [19:04] JanC: yes; restores can take three to five hours before your data can be downloaded :) [19:04] sarnold: hm, they use a tape robot, or what? [19:04] Got an email from vps provider saying vps was attempting to send smtp mail at ~100+ times per ten minutes. However I'm not aware of any services that would be sending /that much/ mail in /that little/ timeframe. I've checked my mail log and the most recent entry is Oct 2. How can i find out what is trying to send smtp mail? [19:05] sarnold: Asisde from slow restores I'm really really happy with tarsnap. [19:05] JanC: I _hope_ it's a robot.. :D [19:05] or sjust dudes runnign aroun with hard disks? :) [19:06] sarnold: I'm using rsync.net for all offsites, for two reasons - one they have georedundancy built-in (if you pay for it :), and two - if I do something that I'm likely to have to change (i.e. when s3 changes somehow) i'm 100% certain to lose something during the transition :) [19:06] esde: check netstat -antp for connections that look like outgoing mail [19:07] so i put everything in one place (separately encrypted) and won't hopefully ever move it. [19:07] andol: cool, thanks :) I wouldn't mind kicking cperciva a few bucks every now and then.. hehe [19:07] hallyn_: I'm checking them out. Never looked at them before [19:08] stgraber: lxc-info -p is broken, spits out name and well as pid. [19:09] thanks sarnold im seeing localhost listening on smtp ports 587 and 25, going to try and find the processes listening now [19:09] you need to be root for the 'p' in 'netstat -antp' to work, I think [19:09] ahh [19:09] ty [19:09] esde: 'p' should show you the processes [19:10] esde: note that the processes listening on ports may not be the ones sending mail.. [19:11] sarnold: thanks for the prod, fixed my backup script (which hadn't worked since oct 28 due to lxc-info -p break :) [19:11] hallyn_: \o/ yay! :) [19:11] another cool thing about rsync.net which i haven't blogged about yet - they have subaccounts, [19:12] so i have less-trusted ssh keys on remote machines which can only push to subaccounts [19:12] hallyn_: hey, error27 pointed out earlier that having network admin capabilities allows setting up iscsi, since it apparently just requires netlink ... [19:12] stgraber: oh, sorry, i thought lxc-info was a python script [19:12] hallyn_: oh very cool [19:12] and of course in case somebody broke in, they can also "fix" netstat or the kernel so that it doesn't show their malware... [19:13] sarnold: not in a container. netlink to kernel doesn't work from child netns [19:13] Hey guys I'm having issues with pled media server indexing any you guys had any experience in pled media server ? [19:13] hallyn_: sounds like a regression from Dwight's recent change [19:14] hallyn_: Oh? Nice. He thought for sure network-admin-rights in a namespace would be sufficient.. [19:14] hallyn_: are you working on a fix already or should I do that? [19:14] hallyn_: how do you like rsync.net? It looks like they charge per GB rather than by how many clients you use, is that right? [19:14] Hi, where can i find hd media for ubuntu server 12.04.03 install ? [19:14] stgraber: i'm not right now. if you have a chance that'd be great, else i'll get to it [19:15] i can't find it on the website [19:15] esde: if you're having trouble spotting it try using tcpdump -- it might be a flood of data, and you might want to make sure you don't dump your ssh connection packets, since that'll just be viewing tcpdump output... :) [19:15] kpettit: i love it. yes, by Gb [19:15] kpettit: that's why i had to sit and rationalize whether/why to stick with them. i've decided it's worth it to me. [19:16] of course, keep in mind, when i talk about fire/floods, i'm not talking hypotehticals, both have destroyed some hardware for me and made me move :) so it *is* worth it to me. [19:16] thank you sarnold trying that now [19:16] ok i need to change locales - biab [19:16] hallyn_: thanks. I've been checking on various ones out there. I hate the ones that charge by client. Some of my customers only have 10-100MB of data so it sucks to pay for that [19:16] http://www.linuxquestions.org/questions/linux-server-73/xen-start-vm-windows-%3D-problem-help-please-4175484442/ [19:16] kpettit: they also have great customer support, and are great for diy unix users. rsync and ssh. you can ask them to disable your password once you set up a ssh key. [19:17] ok, out [19:17] hallyn_: Can you centrall manage it, or do you have to configure each install on the client individually? [19:17] oh cool. I use nothing but keys on all my linux servers and have cygwin/ssh/keys on 1/2 of my windows ones as well [19:19] hallyn_: hmm, so I think this is a voluntary change by Dwight as -n is now optional and accepts regexps [19:20] hallyn_: hrm, re: netlink and namespaces.. df008c91f835 [19:22] kpettit: there's no client configuration [19:22] kpettit: I keep a .ssh/config file entry called 'rsync', i guess [19:23] so i can just 'rsync -va $dir rsync:' [19:23] cool [19:23] stgraber: hm. well it messes up scripts [19:23] trivially fixed of course... on a case by case basis though [19:23] hallyn_: yeah, I know... I think I'm going to send an e-mail to lxc-devel suggesting we revert that particular feature [19:24] hallyn_: Ah ok. thanks [19:24] hallyn_: I don't see a good reason for that specific command to support regexps when we already have lxc-list that does a similar job [19:24] stgraber: well, still, long as it's changing, can we have lxc-info -n -p not print out "Pid:' either? :) [19:24] stgraber: or, we should just have it output "name: pid" for each entry [19:24] that'd make the most sense [19:24] the previous behaviro was redundant [19:25] hallyn_: yeah, I was kind of hoping that the new -H would do that, but apparently dwight only used it to change the notation for network traffic... [19:25] sarnold: ah, yes. i'd forgotten that, and that's good - though it won't affect iscsi netlink msgs, [19:26] note that all the msgs allowed are relating to your own private copy of the l2 net stack [19:28] hallyn_: that is so hard for me to wrap my head around. I think my brain stops at 4.4 BSD :) [19:31] sarnold: just think, anything you create in your namespace, is yours. [19:32] when you create a netns, it creates a new network stack, and you own it. [19:32] you can create devices in there, but you can't pass them back to the host one, bc you dont' own that one. the parent one does own your stack, bc it is the creator of the ns which created the stack, and so it can pass devices to you. [19:33] sarnold: i'm elaborating since it relates to the lxc_user_nic.c i asked you to review :) [19:34] hallyn_: :) I'm roughly half-way through the lxc_user_nic.c from friday... I'm hoping to finish it off today after I finish the day's cve triage :D [19:35] sarnold: awesome, thanks [19:43] hallyn_: sent an e-mail to lxc-devel about lxc-info and what I think we should change [19:45] stgraber: thanks. about to test my workaround for libvirt+qemu bug in trusty that's been playing with my head [19:45] if i can lick that, then i can finish up qemu and then finish up lxc-destroy for unprivileged users [19:47] what are the odds of this working though [19:47] where's my magic eightball [19:48] woohoo, outlook good it says [19:54] magic eithballs are always accurate if they give good news [19:55] i used http://8ball.tridelphia.net/ :) [20:05] http://www.linuxquestions.org/questions/linux-server-73/xen-start-vm-windows-%3D-problem-help-please-4175484442/ [20:16] hi, any way to determine if my server was under a ddos ? mysql crashed, I think because apache took out all the memory so kernel shutdown mysql.. I tried looking at apache access log but its not helping me [20:19] jnor, did you already tried the mysql error log? [20:19] to see the reason of the crash [20:19] yes, I think it crash because lack of memory [20:20] hm what your free -m output says? [20:23] sorry network got cut off [20:24] Im thinking mysql crashed duo lack of memory, but I would like to determine if someone was using ddos on http or if my server just got overloaded [20:38] you think it crashed due to lack of memory? [20:38] it will be in the logs [20:45] Hi Guys, i just rebooted my ubuntu12.04 LTS server for a firmware upgrade, and now apt-get is slow as hell while reading the package list. Any clues of how to solve this ? [21:06] zul: http://people.canonical.com/~serge/0001-accomodate-new-qemu-migration-status-setup.patch that seems to work, leaving me with only 5 unrelated qrt failures [21:06] may send it upstream, if i get over my shyness [21:07] hallyn_: should we sit on it until you look at the other onces? [21:11] zul: maybe. i'mjust donig a fresh build to make sure i didn't drop a change trhough blatent quilt mis-use [21:11] ack [21:12] looks like they're all attach/detach_device errors, plus test_CVE_2010_2237_2238 [21:12] * hallyn_ rolls up his sleeves [21:27] http://www.linuxquestions.org/questions/linux-server-73/xen-start-vm-windows-%3D-problem-help-please-4175484442/ [21:30] zul: oh! well 4 of the faiulres i think are still due to the apparmor bug since saucy... lemme ping jjohansen [21:30] hey jjohansen :) === MalcontentMatt is now known as mjohnson15 [21:33] jjohansen: the fix for apparmor policy updates not applying to running tasks, do you expect that to hit trusty's kernel soon? [21:34] zul: so i have one fix to push to qrt itself, other than that, i think that with http://people.canonical.com/~serge/0001-accomodate-new-qemu-migration-status-setup.patch pushed, libvirt 1.1.4 is ready. do you want to push? [21:43] (jjohansen: that's bug 1236455 fwiw) === MalcontentMatt is now known as mjohnson15 [21:55] hallyn_: yeah, I did an audit after my first pass of the fix and found a couple more spots that needed updating. I've done that and am submitting the patch today [21:56] jjohansen: cool, thanks. How long after that before it hits trusty archive? (I've got no clue how often kernel gets published now) [21:56] i'm just wondering when to schedule my next qrt run :) [21:57] saucy: will be 2-4 weeks just depending on where things are in the kernel update cadence (which is 3 weeks) [21:57] trusty: should only be a day or two from when the patch rolls in, there is no fixed schedule but they try to keep it fairly frequent [21:57] hallyn_: ^ [21:58] jjohansen: thanks === MalcontentMatt is now known as mjohnson15 === Gurkenmaster is now known as zz_Gurkenmaster [22:09] Hi good people, with a clean install of 12.04 autocomplete is not working for the users, any ideas? I've looked into a couple of things via forum links (e.g. http://ubuntuforums.org/showthread.php?t=1949279&p=11804219#post11804219 etc.) [22:14] i create vm windows (XEN) [22:14] Internal error: Xenctrl.Error [ memory 7225148 KiB free; to be scrubbed 0 KiB; total 8175 MiB]: 1: Operation not permitted [22:14] ubuntu server [22:14] 13.10 [22:44] do you guys alway set your web-server to UTC by default? [22:45] hey guys trying to run "sudo virt-install -n taylor -r 4096 --disk path=/home/administrator/Images/win2003.img,bus=virtio,size=80 -c /home/administrator/Images/winsrv200364R2-1.iso --accelerate --network network=default,model=virtio --connect=qemu:///system -v" and I get a permissions error after the file is created? [22:49] the error i get is here: http://paste.ubuntu.com/6413130/ [23:25] anyone know what the adduser default settings are for the new user? [23:26] because apparently in 12.04.3 it doesn't set the login shell to bash, just uses dash or whatever sh is symlinked to... === freeflying is now known as freeflying_away [23:28] TheLordOfTime: grep DSHELL /etc/adduser.conf [23:29] sarnold: i think then that adduser is broken [23:30] sarnold: because when using adduser it failed to set bash as the login shell [23:30] in 12.04.3 [23:30] although the adduser config *is* set to use /bin/bash [23:30] it's not setting that [23:31] (double checking on my testing VM) [23:31] TheLordOfTime: worked fine for me with adduser 3.113ubuntu2 [23:33] sarnold: huh, works fine here on my testing VM I use for triaging stuff on... [23:33] * TheLordOfTime shrugs [23:33] cosmic rays, then? [23:34] TheLordOfTime: ltrace or latrace or strace the adduser? === freeflying_away is now known as freeflying [23:40] sarnold: tested with the same adduser on the system (with permission from the system owner) [23:40] and it correctly sets /bin/bash [23:41] so i'm gonna blame cosmic rays messing up the command or something [23:41] TheLordOfTime: those rays get up to some crazy things when we're not looking. [23:42] indeed [23:42] sarnold: a week ago, the mailing lists bounced a message from my subscribed email [23:43] with not modqueue tick on the email or anything [23:43] so the cosmic rays broke that too for a moment xD [23:43] and it was on a list which the email was a subscribed member of, so... heh [23:50] TheLordOfTime: sheesh. You attract the strangest cosmic rays :) [23:50] sarnold: i know, right? [23:50] sarnold: actually, the system that got the rays interfering with adduser isn't my system [23:51] I was just poked as a server guy to see whether it was a bug in bash tabcompletion but noticed it was actually the users not being given bash as their set login shell [23:51] so it ended up dropping them to sh / dash [23:51] (with permission from the owner of course :) ) [23:51] sheesh, that'd be a frustrating one to debug. :) [23:51] sarnold: heheheheheheheheh... [23:51] sarnold: 'cept I have run into that all the time [23:51] ha! [23:52] sarnold: i never use adduser, I use `useradd --create-home --user-group --shell /bin/bash USERNAME` which creates the user and gives them Bash shell... [23:52] sarnold: but... i occasionally forget to add the --shell argument... [23:52] so... it drops them to dash instead :/ [23:52] (`usermod --shell /bin/bash USERNAME` fixes that quickly) [23:53] so... (1) Not A Bug, its the fault of cosmic rays, and (2) issue resolved for the server [23:53] * TheLordOfTime smiles [23:53] TheLordOfTime: and once you fix the tab completion do they then ask you why it takes longer to do things? :) [23:53] "I used to get my shell prompt so quickly... it's not prompt anymore, it's a slompt!" [23:53] sarnold: lol [23:55] sarnold: actually the users of the system I was diagnosing are all "OMG THANK YOU SO MUCH <3 <3 <3" [23:55] :) [23:55] and i'm like "IT'S NOTHING! LEAVE ME ALONE! D:" [23:55] xD [23:55] sarnold: the server guide is more correct saying to use adduser though [23:55] useradd is for power users like myself... not for the standard sysadmin [23:56] sarnold: but yeah, cosmic rays love to interfere with things :/