/srv/irclogs.ubuntu.com/2013/01/09/#ubuntu-server.txt

=== slank is now known as slank_away
Sprocksanyone able to help with an apparent grub issue? when I reboot/boot my ubuntu server the grub menu just sits there until I go to it and hit enter00:54
patdk-lapSprocks, one of two issues01:08
patdk-lapeither grub yelled that something wasn't supported01:08
patdk-lapor grub didn't boot right last time, so it's waiting for manual confirmation01:08
Sprockshow will I know if something isn't supported?01:09
patdk-lapstuff like01:10
patdk-lapWarning: /dev/mapper/root is a crypto device, which GRUB cannot read directly.  Some necessary modules may be missing from /boot/grub/grub.cfg.  You may need to list them in GRUB_PRELOAD_MODULES in /etc/default/grub.  See http://bugs.debian.org/542165 for details.01:10
ninjixdoes anyone have a example of a libvirt xml for direct kernel booting 12.04?01:10
Sprockswould that show on the screen?01:11
ninjixhere's an example of the libivrt config I'm trying to use http://paste.ubuntu.com/1511459/01:12
ninjixit just comes up with a black screen01:13
ninjixdon't see anything when I 'virsh console test1'01:14
Sprockspatdk-lap how/where would I see this warning?01:22
SpankyAny of the network geniuses here have an idea about the STP (Spanning Tree Protocol) setting in DD-WRT?01:25
SpankyTo enable or disable?01:26
sarnoldSpanky: funny enough I just read about STP in the brctl manpage...01:26
SpankyMy old Cisco / Linksys WRVS 4400N at the home office crapped the bed entirely this morning and gave up the ghost.01:27
sarnoldSpanky: the gist is, it is a standardized way for you to have loops in your ethernet and the switches can figure out the correct paths between themselves without going nuts01:27
sarnold(in the past, loops were a definite no-no)01:27
lifelessstill are :P01:28
sarnoldso, if you've got loops, _and_ all your switches do STP, you'll definitely want it.01:28
sarnoldif you've got loops and at least one switch doesn't do STP, you'll be in a world of hurt01:28
lifelessSpanky: turn STP on unless you have no other switches (including any virtualised inside linux servers) on the same l2 network.01:28
SpankyIs a looped switch like an ethernet cable going from Switch A to B and then another cable going from Switch B to A?01:28
lifelessSpanky: yes01:29
sarnoldif you don't have loops, it won't matter much.01:29
SpankyKind of like a snake eating its own tail?01:29
SpankyWho the heck would do that - "looping" that is?  I have some switches that "cascade" from other switches but that's about it...01:30
SpankySo it's not a real risky setting that will bring my network to its knees or anything?01:31
sarnoldright01:32
sarnoldI'd turn it on01:32
SpankyThanks for the advice guys.01:33
sarnoldoh, wow, STP apparently can take 30-50 seconds to respond to topology changes: http://en.wikipedia.org/wiki/Spanning_tree_protocol01:34
sarnoldI had assumed it'd be quicker :)01:34
lifelessnah01:34
lifelessthe timers are there for propogation and retransmits01:34
lifelessits -super- noticable when it kicks in01:35
SpankyBut a topology change would be like a new switch or new wireless access point or something like that right?01:35
lifelessright01:35
lifelessplug a new switch into a production network with STP on all its ports and everything stops for 30 seconds while it learns.01:35
SpankyWould a wireless device like a laptop, android device, or a wired device like a new thin web server or something cause a 30-50 second delay, too?01:36
lifelessno to all three01:37
lifelessunless they start emitting STP frames01:37
lifelesseasy way to cause havoc if you are a troublemaker :)01:37
sarnoldhehe01:38
SpankyThanks for the response.  Sorry, got an important phone call...02:09
adam_gjamespage: fyi--just stuff 3 reviews in the juju queue for the live migration stuff02:59
blkperltheres an interesting discussion on the mailing list that others may want to chime in on: https://lists.ubuntu.com/archives/ubuntu-server/2013-January/006459.html03:05
blkperlabout plymouth on ubuntu server installs03:05
blagwhere is the rndc.conf file supposed to be? is it at /etc/rndc.conf or /etc/bind9/rndc.conf ?05:31
=== smb` is now known as smb
negevhi, i just installed a grub update on a server with software raid and it's segfaulting trying to install to /dev/md1, do i need a special parameter or something?  it's grub-install (GRUB) 1.99-21ubuntu3.708:59
ed8Hi there, I wish to install an ubuntu server (12.04) on a HP Proliant BL460c G6, but it's not list as supported here : http://h18004.www1.hp.com/products/servers/linux/hplinuxcert-canonical.html#matrix.html nor in http://is.gd/TrOjkJ09:24
SuperMattwell, stick in a live cd or live usb and see if it works09:25
SuperMattif it does, installing it will be fine09:25
ed8SuperMatt: it's a remote server, for a planned install. I'm preparing whatever I need before doing the install09:26
=== yofel_ is now known as yofel
=== peterrus- is now known as peterrus
=== smb` is now known as smb
=== frankban_ is now known as frankban
SuperMattdear #ubuntu-server, I am writing because I would like to know if PHP5.4 will ever be in backports for 12.04. Regards, Matthew11:22
thealmandPlease forgive my ignorance. I had some questions about hosting a site. I played around a little with a second layer domain and am ready to take the plunge and buy and domain but don't want to have to "register" it I can do it myself. Know any good books lol11:27
thealmandI like to find my own answers but Google is really trying to sell me in this one....11:29
SuperMattsorry, I'm a little confused11:37
SuperMattyou want to buy a domain, right?11:37
greppythealmand: you have to pay a registrar for a domain name, personally I suggest http://hover.com.  Straight forward, don't try to upsell you on every mouse click, good prices.11:44
jamespagezul, quantum just started using alembic for db migrations - I'm taking a look now11:59
zuljamespage: wtf....why would they do it before g212:11
jamespagezul, it wants 0.4.1 - just updating the package now12:12
jamespageit landed today12:12
zulstilll..12:13
=== cpg is now known as cpg|away
=== Err404NotFound is now known as Error404NotFound
zullifeless: ping....did you cut a newer testrepository?13:45
zuljamespage: since alembic is being used now does the configuration files need to be updated?13:48
jamespagezul, hrm - maybe13:48
jamespageI'm not sure TBH13:49
jamespagezul, are you referring to the patch we carry for rootwrap and sql_connection defaults?13:49
zulyeah13:50
zulDaviey:  ping did you have a look at python-pecan and python-ceilometerclient13:50
Davieyzul: nein13:52
koolhead17jamespage: so you did testing of quantum with tunnel mode i suppose13:57
jamespagekoolhead17, yep13:57
koolhead17jamespage: am trying vlan mode let me see how far i reach :)13:58
Davieyhttp://blog.picloud.com/2013/01/08/when-ec2-hardware-changes-underneath-you/14:28
zulDaviey: see ec2 is evil14:30
=== huats_ is now known as huats
eagles0513875_hey guys how often by default does ubuntu run ntpd to sync time or does one need to set a cron job for that?14:38
Davieyeagles0513875_: by default, it happens as a hard blow on interface up.14:40
Davieyeagles0513875_: you probably want to install ntp for gradual sync14:40
eagles0513875_Daviey: i was going to setup a cronjob for ntpd as for some reason when time gets out of sync on the server postfix and dovecot throw hissy fits in sending emails it seems14:41
ikoniathat's not how you use ntpd14:41
ikoniacron depends on time......ntpd manages time....do you see the problem ?14:41
eagles0513875_actually yes i do14:42
ikoniaand of course mail may go mad if time is out of sync.....time stamps /future/past can cause problems for the target mail serers14:42
Davieyeagles0513875_: just sudo apt-get install ntp .. then go home :)14:42
eagles0513875_ikonia: i have noticed that and running sudo ntpd seems to have solved the problem. does ntp function with a configuration file that has it run every so often?14:42
eagles0513875_Daviey: its already installed just not sure how often it is run and this server doesnt get rebooted too often14:43
ikoniaeagles0513875_: I suggest researching ntp - the ntp daemon should be running 24x7 and sync within a certain drift range14:43
patdk-wkin my case, removing ntpdate from ifup helps a crapload too14:43
Davieyeagles0513875_: Are you /certain/ it is installed?  by default, we only install ntpdate14:43
patdk-wknothing like 30+ copies of ntpdate running at boot :(14:43
ikoniaif a drift is greater than 5 minuts, ntpd will not sync time14:43
maxbNTP is designed to run continuously, *not* just occasionally re-step the clock14:43
ikoniaso you need to manually sync the clock to less than 5 minute drift and then maintain ntp14:43
eagles0513875_Daviey: yes according to apt-cache policy its showing installed14:43
eagles0513875_ok14:44
eagles0513875_will do some reading on ntp14:45
ikoniamaybe doing research before trying to run these things is a better step14:45
jamespagezul, Daviey: any reason why I should not bump python-mock up to 1.0.1?  who would know best?14:46
zuljamespage: no reason for me14:47
zuldoko maybe?14:47
hallynwell that was weird.  ssh agent seemed (on raring) in a funky state.  only had one key in it, but had to ssh-add -D and reload it14:50
Davieyjamespage: Seems wise to me TBH14:51
=== kentb-out is now known as kentb
tdelamI am having an issue I can't figure out why it is happening and what I can do to figure it out. My Ubuntu Server 12.04 LTS keeps dropping connection sporadically, it has 2 networks cards in it so I configured the second one and changed to that one and it is still happening, I am using aa static internal address. What could it be or how could I pinpoint this issue?15:00
tdelamThe servers purpose is simply a web proxy for a web server so it's not overloaded or anything remotely close to that, infact only 3% of it's available resources are used up :)15:01
jamespageDaviey, zul: that fixes the quantum build failures; I'm just testing the rbd's15:09
zuljamespage:  ack...can you find out the reverse depends on python-mock and see what it effects15:09
jamespagezul, testing those right now15:09
lifelesszul: slipped, sorry. Near top of pile now15:15
zullifeless:  cool...i got the testsuite running now thanks for the help15:15
=== slank_away is now known as slank
=== wizonesolutions_ is now known as wizonesolutions
=== Err404NotFound is now known as Error404NotFound
Davieyadam_g: i added a cronjob for the grizzly report, is it correct? http://status.qa.ubuntu.com/reports/ubuntu-server/cloud-archive/grizzly_versions.html15:50
Myrttirobbiew: https://developer.pidgin.im/ticket/1463515:55
robbiewhuh?15:56
blagI think my bind server (recompiled from apt-get source with --with-dlz-postgres option) is crashing when I run a "dig @localhost" command - how do I set the debug level for named and where to log its output?16:04
blag...sorry, that should be "dig @localhost example.com" command16:04
=== koolhead11|away is now known as koolhead|H
hallynzul: d'oh!  the qemu package for raring depends on an ipxe package which i forgot to ask you to push.16:51
hallynzul: (which i don't have permission to push)16:52
hallynzul: it's https://launchpad.net/~serge-hallyn/+archive/crossc/+files/ipxe_1.0.0%2Bgit-4.d6b0b76-0ubuntu1ppa1.dsc (substituting 1ppa in the version in changelog for 2)16:52
hallyn((that is, 0ubuntu1ppa1 becomes 0ubuntu2, of course)16:53
hallynsigh, wish i had held off on the changes to /etc/init/qemu-kvm.conf17:03
TheLordOfTimehow can i create a local apt mirror, and what spec(s) am i going to need for it?17:05
jpdsTheLordOfTime: https://wiki.ubuntu.com/Mirrors17:05
tgm4883TheLordOfTime, depends on what you want to use it for17:05
tgm4883eg. public vs private mirror17:06
TheLordOfTimetgm4883, private mirror17:06
tgm4883TheLordOfTime, full mirror or partial?17:06
TheLordOfTimetgm4883, difference?17:06
tgm4883eg. all ubuntu releases, or only some17:06
tgm4883For our private mirror at work, I only mirror 12.0417:07
TheLordOfTimetgm4883, precise and later.17:07
TheLordOfTimetgm4883, needs precise, quantal, raring, and likely later17:07
TheLordOfTimealthoug hcome April a full mirror may be more prudent.17:07
TheLordOfTimetgm4883, short term i need precise and quantal17:07
tgm4883well for a full mirror (with I assume less maintenance), you would just use rsync17:08
tgm4883here, I use apt-mirror17:08
tgm4883I say the other is less maintenance because you are just doing a full sync, where with apt-mirror, I have to tell it what releases I want17:09
tgm4883TheLordOfTime, How much space do you have?17:09
TheLordOfTimetgm4883, how much'll i need minimum?17:10
tgm4883550GB for a full mirror according to the wiki17:10
TheLordOfTimeat least, just for precise17:10
TheLordOfTime(i'd like to get Quantal, but...)17:10
tgm4883looks like i'm using 88GB for precise17:11
TheLordOfTime88GB is no issue, so assuming twice, 176GB maybe?17:11
tgm4883and that is also i386 and amd64 only17:11
TheLordOfTimewhich is all I need17:11
tgm4883TheLordOfTime, I'd say 200GB should be fine17:11
tgm4883Assuming that Quantal has more packages than Precise17:12
TheLordOfTimemhm17:12
TheLordOfTimetgm4883, how do i configure the apt-mirror (for precise at least)?17:13
tgm4883TheLordOfTime, so once apt-mirror is installed, configure /etc/apt/mirror.list17:13
tgm4883ours looks like http://pastebin.ubuntu.com/1513561/17:14
tgm4883set the base path to where you want to store the stuff17:14
TheLordOfTimeroot-writable?17:14
TheLordOfTimeor can it run in userspace17:14
tgm4883it runs as the user apt-mirror17:14
tgm4883oh, you'll need a webserver too17:16
TheLordOfTimeoops i have to update my local server17:16
TheLordOfTimetgm4883, does nginx work?17:16
TheLordOfTime:P17:16
tgm4883TheLordOfTime, I'd assume anything that can serve up files would work17:16
tgm4883I just use apache and symlink the mirror to /var/www/ubuntu17:17
TheLordOfTimei can configure where nginx listens which makes life easy17:17
TheLordOfTimeshould probably update the system first :P17:18
TheLordOfTime(96 updates)17:18
tgm4883Not 100% sure that matters, but ok17:19
TheLordOfTimetgm4883, a lot of it is security updates :P17:19
TheLordOfTimewhich means i should update anyways17:19
tgm4883yes17:19
tgm4883Once you've mirrored everything, you'll need to point your clients to this server (you should probably point this server at itself too)17:20
tgm4883Now, depending on how large environment and what exactly you are trying to accomplish, squid-deb-proxy might work better17:20
TheLordOfTimeindeed.17:20
tgm4883which just caches packages that are requested17:21
TheLordOfTimetgm4883, small environment - 6 systems, plus 2 VMs.17:21
tgm4883hmm17:21
RoyKanyone that knows a good VoD solution that can work well in a large environment and has good content management? testing mediamosa now with sitebuilder, and it's got a few bad sides17:22
tgm4883what is the need for the mirror?17:22
tgm4883slow internet connection? Don't want to redownload the same packages repeatedly?17:22
TheLordOfTimetgm4883, the second one, as well as intermittent external connectivity17:22
TheLordOfTimeat least, from the systems on that subnet17:23
tgm4883TheLordOfTime, ah ok, so yea a mirror is probably better then17:23
tgm4883You'll also want to edit /etc/cron.d/apt-mirror for when you want it to sync17:23
TheLordOfTimewhich sync time would you suggest?17:24
RoyKonce an hour?17:24
tgm4883Depends on how up to date you want to be?17:24
tgm4883I do it every 4 hours17:24
tgm4883RoyK, once an hour is probably overkill17:24
RoyKshould suffice17:24
TheLordOfTimeuptodate daily may be more prudent - unattended-upgrades runs every 24 hours.17:24
TheLordOfTimeand pulls security updates17:24
samba35hi RoyK how are you17:24
TheLordOfTime(I got to the machine before unattended-upgrades ran today)17:24
RoyKsamba35: fine, thanks17:25
samba35was playing with openvswitch17:25
samba35understanding process17:25
TheLordOfTimetgm4883, where should i pull data from for the mirror, any of the regional archives (such as one of the us archive mirrors)17:30
tgm4883TheLordOfTime, Unless you know of a fast local mirror, probably us.archive.u.c17:31
hallynjamespage: hey - are you around?17:34
jamespagehallyn, yep17:34
adam_gDaviey: looks like its still pointed at the folsom repos?17:35
TheLordOfTimetgm4883, i may use the us.archive.u.c mirror, and i'll do every 6 hours for an update.17:36
hallynjamespage: could I get you to push http://people.canonical.com/~serge/ipxe_1.0.0+git-4.d6b0b76-0ubuntu2.dsc ?17:36
hallynjamespage: it just adds a qemu-ipxe package which the new qemu depends on17:36
hallyn(which it does as part of an attempt to minimize delta from debian)17:37
hallyn(much more to be done for that, in progress..)17:37
=== ragnar0k_ is now known as ragnar0k
jamespagehallyn, dget fails to download it17:38
hallynhm?17:38
jamespagehallyn, dget -u http://people.canonical.com/~serge/ipxe_1.0.0+git-4.d6b0b76-0ubuntu2.dsc17:38
jamespagecan't get the .orig.tar.gz so bails17:38
hallynjamespage: oh the orig.tar.gz from raring17:39
hallynbut i can push it if you prefer, one sec17:39
hallynjamespage: pushed17:39
jamespageta17:39
hallynwell... pushed, but rsync still hanging, hold on,17:39
hallyni don't get why it goes to 100% then hangs for 2 mins17:40
hallyn(while sha1sums are still different, so it seems to still be writing)17:40
hallynjamespage: ok, there, NOw done ;)17:40
TheLordOfTimetgm4883, so i assume i point the systems on the subnet to the system that has apt-mirror on it.  where do i need to symlink for the http system to serve it right?17:49
tgm4883TheLordOfTime, IIRC, it's <basedir>/mirror/<mirror>/ubuntu17:50
TheLordOfTimewhat's its defaults?17:50
TheLordOfTime... oop?17:51
tgm4883/var/lib/apt-mirror/mirror/us.archive.ubuntu.com/mirror I think17:51
TheLordOfTimesegfault?17:51
TheLordOfTimeo.O17:51
TheLordOfTimeit says "Segmentation Fault", then dropped me to the shell o.O17:51
tgm4883hmm17:51
* TheLordOfTime checks kernel logs17:51
TheLordOfTime... again, memory errors?17:52
* TheLordOfTime goes to order replacement RAM chips17:53
TheLordOfTimes/chips/modules/17:53
=== Ursinha is now known as Ursinha-afk
TheBronxhi all18:14
TheBronxis it normal to have thousands of messages like "r8169: eth0: link up" during a SYN flood attack?18:15
TheBronxwhat does "link up" means?18:15
TheBronxmean*18:15
RoyKTheBronx: it means the realtek driver detected the network link went up (probably just after it went down)18:17
TheBronxi have this message repeated a lot of times18:18
TheBronxeven 20 or 30 per SECOND18:18
RoyKrealtek NICs are rather infamous of small buffers, and rather bad hardware design in general18:18
TheBronxso, that's not normal, even under a flood attack of about 50Mbps18:18
RoyKhaven't seen that18:19
RoyKbut could you try with another NIC?18:19
TheBronxwell, Im from spain, the server is in germany18:19
TheBronx:P18:19
RoyKvirtual or physical?18:20
TheBronxphysical, it is a dedicated server, from hetzner18:20
hallynjamespage: did that dget all right?18:21
TheBronxI've sent a ticket to them explaining the problem. Lets see what they answer18:29
TheBronxafter every attack I can't access the server until I manually restart it remotely18:30
TheBronxit seems the NIC doesn't "wake up" after the attack18:31
TheBronxAnswer: we have many servers with that network chip and I am not aware of any18:36
TheBronxproblems. However, most of our customers use CentOS Linux, so there18:36
TheBronxmight be an issue with Ubuntu.18:36
TheBronxwell, their customers are not being attacked probably xD18:36
=== slank is now known as slank_away
adam_gwin 2218:42
TheBronxis it possible that the link up error is caused by a wrong configuration in /etc/sysctl.conf??18:47
sarnoldTheBronx: in my experience, spurious link up / link down messages point to failing cables, failing switches, etc.18:49
TheBronxthis is my sysctl http://pastebin.com/HUk0RVip18:49
sarnoldTheBronx: though you've got a unique experience, hehe18:49
TheBronxI will update my ubuntu 10.04 server18:50
TheBronxhttp://www.twm-kd.com/linux/realtek-rtl81688111e-and-ubuntu-linux/18:50
TheBronxit seems there is a bug in the driver18:50
TheBronxsolved in ubuntu 1218:50
sarnoldTheBronx: oh, wow.18:51
TheBronxis it a good idea? what do you think?18:51
TheBronxif it doesn't work I will move to a different server, I just can't do more18:53
jamespagehallyn: it did - sorry I've managed to crash into parenting duties for the evening; is tomorrow AM my time early enough?18:54
TheBronxis it a good idea? what do you think? sarnold?18:54
sarnoldTheBronx: it seems like a good idea; even if it doesn't _work_, you'll at least be on newer software, with longer support.18:55
hallynjamespage: ok, thx, ttyl18:55
TheBronxwell, thank you guys, sarnold and RoyK, I will update everything and see what happens (and what crashes hehehe)18:57
sarnoldTheBronx: hehe :) have fun, good luck18:59
=== ragnar0k_ is now known as ragnar0k
=== slank_away is now known as slank
zulhallyn: ixpe?19:30
hallynzul: ipxe19:30
zulhallyn: what about it again?19:30
hallynzul: i've asked jamespage to push it19:30
zulhallyn: oh sorry19:31
hallyn(he'll do it tomorrow - rugrat issues :)19:31
hallynzul: np!  thanks - ttyl19:31
keithzgAny chance of newer qemu-kvm packages making it into 12.04? I'm desperate for the ability to use blockpull to merge back external snapshots, as one can do in the version in 12.10.19:55
ikoniadoubtful19:56
paco1hello masters!19:58
paco1Anybody knows a good documentation to implement TSIG in my DNS in ubuntu 12.04?19:58
paco1thanks in advance!19:58
keithzgikonia: that's what I thought . . . but that leaves me in a pretty tough place. The version of qemu-kvm in 12.04 can split VMs into external snapshots, but not collapse the chain, so incremental backup snapshots result in an ever-growing backing chain . . .20:00
* keithzg is wary of upgrading important hosting a dozen VMs to a non-LTS release, though, so . . . sigh.20:00
RoyKkeithzg: eh - no systems I know can combine snapshots20:03
sarnoldkeithzg: perhaps you could ask for a backport? https://wiki.ubuntu.com/UbuntuBackports20:03
keithzgsarnold: not a bad plan, not a bad plan a'tall20:04
keithzgRoyK: maybe not "combine" per se, but at least collapse the backing chain of external snapshot .qcow2 images. It's what the libvirt and qemu folks suggest, since over time an extremely long external chain will start showing performance issues for obvious reasons20:05
tomVovohello everyone..20:06
tomVovoI need help20:06
tomVovoanyone help me?20:06
keithzgRoyK: AFAIK there's a nicer, fairly different method of snapshotting coming down the pipeline in qemu that'll do an end-run around these sorts of issues, but *that's* certainly never going to make it into 12.04, heh.20:06
tomVovohello20:07
tomVovoI run my Ubuntu server apache 12.04 ...I have domain with godaddy.com...iT'S WONT CONNECT20:07
tomVovoHELP!20:07
RoyKkeithzg: the good way is to snapshot at the lvm level, when that gets good - some patches are posted, but haven't been accepted yet20:08
tomVovohelp!!!..20:08
RoyKthe current lvm snapshot model isn't too scalable to say the least :P20:08
RoyK!ask | tomVovo20:08
ubottutomVovo: Please don't ask to ask a question, simply ask the question (all on ONE line and in the channel, so that others can read and follow it easily). If anyone knows the answer they will most likely reply. :-) See also !patience20:08
tomVovook20:09
ikoniahow do you expect people if you just join a channel saying "help, anyone help, help"20:09
keithzgRoyK: Heh. I'll admit, I have an irrational aversion to LVM---and at least in my case, the setup I've inherited is on some iSCSI shares where LVM wouldn't be an option anyways---but I don't doubt you're right20:11
ikoniatomVovo: ok, so the help isn't that urgent then as you've not said anything since being asked to provide information on the problem you have20:16
TheBronxhey, one question20:43
TheBronxif I upgrade ubuntu server from 10.04 to 1220:43
TheBronxvia SSH20:43
TheBronxand the network driver updates20:43
TheBronxdid the SSH connection crash?20:44
TheBronxwill*20:44
RoyKyou'll have to reboot, and after the reboot, it'll probably work20:44
RoyK*probably*20:44
RoyKbecause some issues may come up, some driver may not work etc20:44
TheBronxwill the upgrade process cancel?20:45
RoyKno, the upgrade process will continue and probably finish until you reboot20:45
RoyKafter the reboot, there's always a chance of something going wrong20:45
RoyKnot a big chance, but still20:45
TheBronxok, that's not a problem, I will backup the data20:45
TheBronxI just was worried about the upgrade process cancelling20:46
RoyKif you have a good backup, and can live with some downtime, no problem20:46
RoyKit'll open up a secondary port for ssh for the upgrade20:46
TheBronxwell, after 5 days of continued DDoS downtime is not going to be a problem20:46
TheBronxxDDDDDD20:46
RoyKusually it won't cut that port20:46
RoyKnever seen that happen20:46
TheBronxok, thank you again RoyK!20:46
RoyKnever seen 22/tcp cut either20:47
RoyKthat is, never used the secondary port20:47
shaunoI like to use screen and plan on it kicking me off.  it never seems to do it if I'm ready for it20:47
RoyKshauno: did screen kick you out, or did the ssh session die?20:49
RoyKshauno: tmux is a newer thing simiar to screen, some say better, but it won't help much if the ssh session times out - then you need just ssh keepalives20:50
shaunoneither; I mean if I'm prepared for it to go wrong, it doesn't break.  if I'm not prepared, it's inevitable20:51
RoyKhehe20:52
phillwHi guys, is my theory correct in using the server iso (13.04) about the nearest we can get to the mini-iso that produces a ubuntu type system upon which a DE can be further installed if we select no additional tools?21:55
justizinhey guys i have a machine whose grub menu i need to get into via the serial console, i don't think shift-down works, i have done it before but my google fu is failing.  anyone got a notion handy of how i can get in?21:59
sw0rdfishuhh, how do I start znc in screen mode, and detach and re-attach at will?22:02
sarnoldjustizin: wild-guess-land: send a break?22:04
sarnoldsw0rdfish: check the screen manpage, look for ^A d (to detach) and screen -RAD (to reattach) -- no idea about znc specifically22:05
justizinworth a shot. :)22:05
justizinmaybe i fixed this last time by mounting the damned thing to a netbooted nfsroot.22:06
sw0rdfishhmmm22:07
keithzgphilw: does the net install image not still exist? I'd imagine that'd be the closest, actually.22:14
keithzgphillw, I mean22:15
jcastrohttps://help.ubuntu.com/community/Installation/MinimalCD22:15
jcastroit exists22:15
phillwkeithzg: at the moment, there is https://launchpad.net/bugs/107715822:16
uvirtbotLaunchpad bug 1077158 in testdrive "Testdrive should have a netboot repository" [Undecided,New]22:16
phillwwhich seems a lonely bug, but blocking22:16
linuxmanIs running a Ubuntu server with Desktop GUI good?22:18
linuxmanI am used to gui..22:18
TheLordOfTimelinuxman, i run a netbook that runs as a fileserver for my network via a GUI22:19
linuxmani was expecting people to laugh at me22:19
TheLordOfTimeinstalled Lubuntu to start with, then installed the server packages.  so sort of the opposite of what you're at, but...22:19
linuxmanhehe22:19
TheLordOfTimelinuxman, BUT22:19
TheLordOfTimeyou need to be aware of something: ssh != remote desktop enviromnet.22:20
linuxmanyes22:20
TheLordOfTimeas well, the GUI'll tax your RAM/system, so if you don't have a lot of RAM, you'll need to reconsider installation of the gui22:20
linuxman10-18 CPU usage, 255MB RAM used out of 512MB22:21
linuxmanmy PC is OK about this, I run it in VM22:21
keithzgphillw: ah, I see I see, no netboot for Raring.22:22
linuxmani got OpenSSH, LAMP install, how do i manage them?22:22
keithzgphillw: you could always just use the 12.10 netbook, manually upgrade to raring, and go from there22:22
linuxmancan i access them via GUI, as well as command line?22:22
keithzgphillw: err, 12.10 netboot I mean22:23
TheLordOfTimelinuxman, unfortunately for you for those... no real gui configuration tools22:23
TheLordOfTimeexcept for webmin which i strongly DO NOT advise you use.22:23
linuxmanwhy not?22:24
TheLordOfTimebecause i hate guis for server management22:24
* justizin hangs head in shame over defeatist "just mount with netboot nfsroot" solution22:24
linuxmanok but how do i navigate with command line? i mean how do i configure OpenSSH, LAMP etc.. I am used into having GUI and command line..22:25
keithzgjustizin: what works, works22:26
* TheLordOfTime has to disappear to fix a misbehaving server at a datacenter down the road22:26
TheLordOfTimeso i can't help much more there linuxman, since this is an urgent-attention fix.22:26
linuxmanok, good luck22:27
sarnoldlinuxman: to a certain extent, some packages are configurable via debconf; dpkg-reconfigure <packagename> will prompt you a few times to set things up.22:27
sarnoldlinuxman: but that only gets you so far. at some point, things are just easier to configure with a text editor.22:28
=== cpg|away is now known as cpg
linuxmanhow do i SSH from outside world into my own Server at home that is running in VM? I have a router and several other people have been assigned internal ips, so as me.22:38
sarnoldlinuxman: does your VM have a publicly routable IP address assigned? or does it have an RFC1918 private IP address?22:39
linuxmani did ifconfig on it, it has a private IP address22:39
linuxmanthe router has only ip , its like a NAT22:39
linuxmanit acts*22:40
phillwkeithzg: you have mis understood my request..22:41
sarnoldlinuxman: then you'll probably need to portforward a port on your machine to the VM, see e.g. http://www.debian-administration.org/articles/7322:42
linuxmanok thanks22:45
paco1Anybody knows TSIG on DNS?23:38
paco1i want to configure TSIG on my DNS, someone knows good documentation?23:41

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!