=== arrrghhh is now known as arrrghhhAWAY [00:16] zul: no [00:16] adam_g: lemme just finish dinner and ill do it [00:18] zul: what needs to be done do it, exactly? [00:18] adam_g: add it to the lp:~ubuntu-core-dev/quantal/seeds basically [00:18] then magic [00:19] zul: oh, i guess i can do it and send a merge proposal if you don't want to [00:19] adam_g: probably faster if i just did it [00:20] cool go for it. dont forget python-cliff-tablib too [00:22] you mean cliff-tablib [00:24] actually i think cliff-tablib will be picked up automatically as well, anyways done === cpg is now known as cpg|away [00:33] zul: nice, thanks [00:39] hallyn, that was quantal === cpg|away is now known as cpg === dendro-afk is now known as dendrobates [01:53] smoser: can you file a bug? i'ts supposed to do that (in the block starting at line 180) === 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 === dendrobates is now known as dendro-afk === n0ts_off is now known as n0ts [05:44] hello [05:45] hi... === n0ts is now known as n0ts_off === n0ts_off is now known as n0ts === Kiall is now known as zz_Kiall [07:19] morning all === zz_Kiall is now known as Kiall [07:26] morn' [07:30] So I enabled logging in debug mode and CoreDumpDirectory, apache client process segfaulted about an hour ago and I have nothing useful in my log and no core dump. [07:36] New bug: #989841 in mysql-5.5 (main) "package mysql-server-5.5 5.5.22-0ubuntu1 failed to install/upgrade: ErrorMessage: subprocess installed post-installation script returned error exit status 1" [Undecided,Expired] https://launchpad.net/bugs/989841 [07:36] New bug: #1021573 in qemu-kvm (main) "package qemu-kvm 1.1~rc+dfsg-1ubuntu5 failed to install/upgrade" [High,Expired] https://launchpad.net/bugs/1021573 [07:36] New bug: #1045342 in mysql-5.5 (main) "partition by range on multiple col. primary key w/ no maxvalue restarts server (dup-of: 1033724)" [Undecided,New] https://launchpad.net/bugs/1045342 [07:36] New bug: #1046115 in cloud-init (main) "package lists are broken after first boot" [Undecided,New] https://launchpad.net/bugs/1046115 [07:36] New bug: #1046117 in lxc (universe) "btrfs via symlink not working" [Undecided,New] https://launchpad.net/bugs/1046117 [08:37] hi guys. how to add an existing user to a group? [08:38] gpasswd -a user group [08:57] hello [08:57] I've found something in my /var/log/auth.log === n0ts is now known as n0ts_off [08:58] and I would like to ask for help [08:58] "Failed password for invalid user root from 202.138.126.128 port 39399 ssh2" [08:59] and that ip has failed more attempts every day === n0ts_off is now known as n0ts [09:08] lynxman, a new snapshot of ipxe fixed my issue [09:08] ffunenga: block that ip if it is from the same ip [09:08] lynxman, wish I had spotted this before feature freeze.... [09:10] Troy: I've user "sudo iptables -A INPUT -s 202.138.126.128 -j DROP" [09:10] is there any tool to block any IP that tries to access the root account via ssh? [09:11] !ufw [09:11] Ubuntu, like any other Linux distribution, has firewall capabilities built-in. The firewall is managed using the 'ufw' command - see https://help.ubuntu.com/community/UFW | An alternative to ufw is the 'iptables' command - See https://help.ubuntu.com/community/IptablesHowTo | GUI frontends such as Gufw (GNOME) and Guarddog (KDE from Lucid onwards) also exist. [09:11] !denyhosts [09:12] ffunenga: denyhosts will block ip's from which illegal login attempts come [09:12] ffunenga: I use it on my servers [09:12] ffunenga: it's rather triggerhappy on failed root login attempts ;) [09:16] I didnt know ufw, it seems great. Ok I will now test denyhosts [09:16] thanks [09:16] ffunenga, root is disable by default for ssh anyway [09:16] ufw is neat [09:16] so there is no risk [09:16] but doesn§t support all of iptables' stuff [09:17] Tellmarch: it's not disabled by default, but by default, there's no root password, so ssh as root won't work. default ssh settings is to *allow* root login [09:18] really? [09:18] well, it's the same, no matter how long he tries, he won't find the password [09:18] sshd will, by default, not allow login to an account without a password [09:18] you're right, PermitRootLogin yes.... I thought this was "no" in the past... [09:18] but if you set a root password, you will be allowed ssh in, unless you reconfigure sshd [09:19] it's usually 'no' on my machines ;) [09:22] Tellmarch: I've inserted a rule in /etc/ssh/sshd_config that only lets my user access the server: "AllowUser ffunenga" === AaronMickDee_ is now known as AaronMickDee [09:29] Trying to uninstall MySQL so I can reinstall but I keep getting this error Processing was halted because there were too many errors. [09:29] E: Sub-process /usr/bin/dpkg returned an error code (1) === railsraider_ is now known as railsraider === n0ts_ is now known as n0ts__off === railsraider___ is now known as railsraider === n0ts__off is now known as n0ts === n0ts is now known as n0ts_off === n0ts_off is now known as n0ts [10:32] does flock() work at thread level or process level/ [10:34] does flock() work at thread level or process level? === n0ts is now known as n0ts_off === n0ts_off is now known as n0ts === cpg is now known as cpg|away [11:17] DavidLevin: around [11:17] Daviey: [11:20] koolhead17: [11:28] smb, your cirrus driver symptoms look just like mine BTW - it certainly does not fail all of the time [11:29] jamespage, Yeah, it fails when the cirrus module fails the VRAM setup. Just don't know why, yet. [11:32] smb, well if you need me to poke anything to help out just let me know [11:33] jamespage, Sure, at least I know understand the random behaviour. Does not help when one had also wiggled around with boot options at the same time... [11:34] well it's always reproducible if I have full disk encryption. [11:35] e.g server or desktop installer choose full-disk encryption. [11:36] every reboot. and me and slangasek did many reboots while trying to triange if it was grub/plymouth or kernel issue. and we deducted it to be kernel. [11:36] every boot fails. [11:36] plus note that every other boot is possibly in grub's last-boot-failed code path so boot experience is the same. [11:37] s/is the same/slightly different/ =) [11:51] New bug: #1046278 in ceph (main) "rest-bench always fails on initial execution" [Undecided,New] https://launchpad.net/bugs/1046278 [12:19] hallyn, [12:19] bug 1046117 [12:19] Launchpad bug 1046117 in lxc "btrfs via symlink not working" [Undecided,New] https://launchpad.net/bugs/1046117 === dendro-afk is now known as dendrobates === n0ts is now known as n0ts_off [13:00] hallyn, ping [13:01] i'm wondering how i get output of: [13:01] sudo lxc-start --name=$R-amd64 -- /sbin/init --verbose [13:04] rbasak, so i'm pretty sure that your issue is different than the lxc wone we were seeing [13:04] as in your case eth0 is comming up but events are somewhere etting lost [13:04] It seems that eth0 is being brought up after cloud-init-nonet has timed out [13:04] but in the lxc case (that forced us to add 'start networking') it seems that the networking for the given nic never cam eup [13:04] And that bringing eth0 up (or at least running ifup.d) is being blocked when cloud-init-nonet blocks. [13:05] rbasak, that makes sense. [13:05] I wonder can anyone help me, i am installing from usb on a server i pick install ubuntu then i get a blank screen [13:05] i know that at one point we had a better undersanding of this problem (at least the lxc ase) === n0ts_off is now known as n0ts [13:14] smoser: one thing that concerns me is that I'm using hackery to generate the image in the first place, and I wonder if this is introducing a problem. How soon could you publish official armhf/highbank dailies to eliminate this? [13:15] what hackery are you using ? [13:15] nothign you've shown me would cuase this. [13:15] so unless you've been lying to me. [13:15] hm.. [13:15] I'm using dannf's Makefile [13:15] That's it really [13:15] but i should look again at dannf's make file i guess. [13:15] link ? [13:15] I modified it to use -proposed (for a highbank kernel fix I need) [13:16] smoser: https://code.launchpad.net/~racb/maas/maas.ubuntu.com.images-ephemeral-arm [13:18] dannf sure loves 'cat' [13:18] Yeah I did see a few extra ones in there [13:19] i really dont see anythign there that would cuase this [13:21] stgraber, sorry to always use you as my "hallyn isn't here fallback" [13:22] but do you know how i can get output of init from 'lxc-start .... -- /sbin/init --verbose' ? [13:22] i'm basically getting nothing. === n0ts is now known as n0ts_off [13:31] smoser: hmm, that should work... I'd suggest pinging jodh as he's been using lxc quite a bit for upstart development and testing [13:31] stgraber, you'd expect that to just go right to the console that i'm looking at? [13:32] yeah and I'd assume it'd also end up in some kind of log somewhere... [13:34] jodh seems not around :-( [13:34] smoser: gah, right, forgot, he's off until Monday... [13:35] i was testing quantal running precise [13:35] i will now test precise urnning precise [13:36] * hallyn reads up [13:36] but i swear that when i was debugging bug 800824 this used to work. [13:36] Launchpad bug 800824 in cloud-init "cloud-init-nonet times out in lxc" [High,Fix released] https://launchpad.net/bugs/800824 [13:37] smoser: how about 'lxc-start -c consoleoutput .... -- /sbin/init --verbose' ? [13:38] oh wait [13:38] i got the same result with --console=/tmp/out as was going to console [13:39] right [13:39] init takes its arguments though env [13:41] i.e. we pass 'container=lxc' to init by doing putenv("container=lxc") === caribou_ is now known as caribou [13:43] hallyn, what? [13:43] init says it takes arguments via cmdline [13:43] its cmdline [13:43] it takes them from the kernel's cmdline [13:43] the kernel provides that to init through env [13:45] :) [13:45] Daviey: it was my stupidity [13:45] now nova/quantum [13:46] glance okey !! [14:01] New bug: #1046330 in php5 (main) "Incorrect crypt() function behavior" [Undecided,New] https://launchpad.net/bugs/1046330 === n0ts_off is now known as n0ts [14:18] hallyn,so are you saying what i want to do is impossible? [14:18] because i swear this used to work [14:20] smoser: no. I haven't gotten it to work. BUT I think it has to do with how we're doing 'console' file now. [14:21] i swear it worked when i was debugging this bug the first time. [14:21] but it seems broken in precise also [14:21] New bug: #1046340 in ntp (main) "ntp package missing logcheck exceptions file /etc/logcheck/ignore.d.server/ntp" [Undecided,New] https://launchpad.net/bugs/1046340 [14:22] Daviey, hallyn: how would you guys feel about a new ipxe snapshot for quantal? [14:22] I hit a bug yesterday with regards to IPXE iscsi san booting - its already reported upstream and is fixed in upstream git repo [14:23] however it appears to be hard to say - "yeah - you need this commit" [14:23] jamespage: sounds good. mind pushing it? [14:23] hallyn, it will need a FFe really [14:23] why? it sounds like it's broken [14:23] not a feature [14:23] well it works for most use cases [14:24] I'm sure there are lots of other new features since the snapshot from feb we currently have [14:24] 380 commits worth of stuff [14:24] but oh i misunderstood [14:24] to get up-to-date with HEAD [14:24] if I could pick a specific commit for a patch I would had JFDI'ed it [14:24] right [14:24] (after beta-1 of course as its seeded) [14:24] have you asked in -release? [14:25] not yet [14:25] bug 1045923 for reference [14:25] Launchpad bug 1045923 in ipxe "input/output error on iscsi sanboot from linux tgt" [Undecided,New] https://launchpad.net/bugs/1045923 [14:27] smoser: yeah i can't even see the arguments in /proc/cmdline or /proc/1/cmdline in the container. [14:30] hello all, I have been trying to use nomodeset to slove my boot problems with no luck [14:30] whereas 'lxc-start -n qlvm1 -- /bin/bash -c /bin/ps -ef' does pass the arguments [14:30] so upstart seems to be (newly) dropping them [14:30] anything else i could try [14:37] jamespage: a newer ipxe sounds good imo [14:37] Daviey, is 380 commits going to cause me grief for a FFe? [14:38] I tested it locally for my problem and it works just fine [14:39] jamespage: i don't think so.. if hallyn is happy from a kvm PoV. [14:42] tdr112: ello === n0ts is now known as n0ts_off [14:49] Daviey, hallyn (+ anyone else): if I stick this in a PPA would you be OK to give it a sniff as well [14:49] ? [14:50] jamespage: sure, though i don't have any exotic tests or test recipes [14:51] hallyn, just 'it didn't break kvm for me' would be good [14:53] wonder if lynxman in his new role woudl still have use for testing it [14:56] tdr112: what issue are you having ? [14:58] on a new server i cant get ubuntu to install, i am getting a blank screen after grub , i have tried NOMODESET but that does not help === n0ts_off is now known as n0ts [14:59] most odd [14:59] Daviey: can you point tdr112 in the right direction please.... [15:00] to give more info its a dell r420 [15:01] booting from usb, I have tested the usb on another server and it boots to the installer fine [15:01] smoser: do you find btrfs containers to actually be usable right now? [15:01] (gauging importance of bug 1046117) [15:01] Launchpad bug 1046117 in lxc "btrfs via symlink not working" [Medium,Triaged] https://launchpad.net/bugs/1046117 [15:02] every time i try btrfs i get corruption [15:03] hallyn, no [15:03] well, i dont know. [15:03] i opened that bug [15:03] because dielete didn't work [15:04] right. but if btrfs is unstable anyway then that bug becomes low priority :) [15:04] but all right, fix should be simple. I'l lstage it in ubuntu:lxc for now. [15:05] smoser: setting up an oneiric container right now to see if it'll work with init --verbose [15:06] hallyn, well, both host and guest oneiric would be what i would have used [15:06] as there was no quantal host at that point [15:06] anyone running a mysql-cluster with also non-cluster tables in the same sql? [15:07] smoser: yup, might nest if i have to :) [15:07] hallyn: I'd gladly :) [15:08] hallyn, i'd trust htat result about as much as btrfs (nested) [15:08] lynxman: \o/ thanks :) [15:08] smoser: stgraber does it all the time [15:08] have faith! [15:09] anyone know how to construct a menuentry for grub2 to boot an ubuntu-server iso? [15:09] tdr112, it'd be "nomodeset", not the capital [15:09] hallyn, lynxman: hold the line - just spotted another commit that might fix it - trying now... [15:09] smoser: i did try it in lowercase [15:09] http://paste.ubuntu.com/1187381/ [15:10] i add that into /etc/default/grub [15:10] and sudo update-grub [15:12] tdr112, also, you can check if you're actually booting correctly or not. if you are seeing a grub prompt, its quite possible that you're booting up fine (getting network and everything) just blindly. [15:12] then ssh in and poke around [15:17] smoser: its a blank system , i dont think the installer will have an ssh server on it [15:21] oh. i see. this is off the cd install. i thought you were saying afterwards. [15:21] how did you create the usb disk? [15:21] and from what iso? [15:21] ubuntu-12.04.1-server-amd64.iso [15:22] how did you create? [15:22] using unetbootin [15:25] i've never used that. but i would suggest trying with usb-creator-gtk before anything else. [15:26] ok thanks smoser will do [15:33] hallyn, lynxman, Daviey: OK - so that second patch fixed my bug; I'll defer the new snapshot until next release... [15:34] jamespage: great \o/ === mcclurmc_away is now known as mcclurmc [15:41] smoser: bad live usb , working now thanks === n0ts is now known as n0ts_off [15:48] jamespage: cool, thanks much [15:48] tdr112, good deal. [16:01] New bug: #1046397 in maas (main) "The DHCP config file does not get written." [Undecided,New] https://launchpad.net/bugs/1046397 === matsubara is now known as matsubara-lunch === matsubara-lunch is now known as matsubara [16:50] soren: ping === Ursinha_ is now known as Guest65952 [17:12] stgraber: hm. i didn't do it. But I assume you'd want clear_config() to be exported through the api? [17:12] (note, a package fixing all your woes should be built in ppa:serge-hallyn/virt. i'm still testing) [17:14] yup, test.create("ubuntu") ends up with a nicely reloaded config === n0ts_off is now known as n0ts === Ursinha-afk is now known as Ursula [17:35] hallyn: exporting clear_config() would be nice indeed. I don't have an immediate need for it, but that might change in the future. === mcclurmc is now known as mcclurmc_away [17:38] stgraber: I assume that would be deemed a new feature, alas [17:38] simple enough to do though. [17:39] let's let the current pending stuff settle a bit and then i'll export clear_config in my github tree [17:40] hallyn: yeah, that'd be a new feature, so will just keep it in git for now === dendrobates is now known as dendro-afk [17:46] gents, installed /boot on sda. i have an sdb, sdc and sdd available that i have mounted in fstab and shared out through nfs. problem is....sometimes the mounts don't take because /boot randomly jumps to sdb. can i make the system always use sda for the OS? === n0ts is now known as n0ts_off === koolhead17|afk is now known as koolhead17 [17:48] lickalott, you can give the mount throught the UUID instead of sda-d [17:48] this way it's always the same thing [17:50] i just learned about blkid 10 mins ago, so i'm sure I could but I don't really understand how to do it. [17:51] same same? mount /dev/sdc1: UUID="81204814-775a-4ac8-923b-a6690381fe9b" /media/WHATEVER ? [17:52] lemme google first [17:55] arosales, Daviey, utlemming, could you take a pass and update what's making it in to Beta 1 (and noting the bits about to land right after), and any key bugs you want to warn folks about. https://wiki.ubuntu.com/QuantalQuetzal/TechnicalOverview/Beta1 [18:03] Tellmarch, quick question.... right now sdb is the drive used for boot, swap, etc.... but it was installed against sda. when I'm setting up fstab, do i want to use the UUID for sdb to mount 1 of my non-os drive? [18:03] *drives [18:03] sda, etc. can change [18:03] while the UUID doesn't change [18:04] so you make all the mounting without talking about sda, sdb, etc. at all [18:04] this way you have no problem [18:04] you say for instance, UUID 81204814-775a-4ac8-923b-a6690381fe9b will be in /boot, UUID (another) will be in /mnt/disk1, etc. [18:05] then no matter if it's called sda or sdb, it will be in the same place in your filesystem [18:05] okay...i got ya now. [18:05] thanks! === dendro-afk is now known as dendrobates [18:10] can I paste my fstab for you to look at? i'm getting an error when i mount -a (mount.nfs: remote share not in 'host:dir' format) [18:12] http://paste2.org/p/2194616 === railsraider_ is now known as railsraider === cpg|away is now known as cpg [18:48] skaet: hello, I'll take a look and add any relevant bits [18:49] thanks arosales [19:02] New bug: #1045616 in bind9 (main) "DHCP server handing out DNS server addresses but DHCP client does not write these to resolv.conf" [Undecided,New] https://launchpad.net/bugs/1045616 [19:34] Tellmarch didn't work. modified teh fstab, rebooted....no mounts [19:45] hallyn, around? [19:45] i've a questoin on lxc.conf [19:47] http://manpages.ubuntu.com/manpages/precise/man5/lxc.conf.5.html [19:47] the moutn point section says: [19:48] the path of the rootfs mount point should be prefixed with the /usr/lib/lxc/root default path or the value of lxc.rootfs.mount if specified. [19:48] smoser: yes, and while i'm here note that ppa:serge-hallyn/virt should have the fix for your btrfs lxc-destroysymlink woes [19:49] what does that mean? [19:49] /usr/lib/lxc/root seems wrong [19:49] smoser: i think that's outdated [19:49] it's outdated in more ways than one [19:49] /usr/lib/lxc/roto is wrong, and it is actually preferred to not prefix the path, rather make it relative [19:49] i.e. just 'dev/pts' [19:50] can i put an entry for / in there? [19:50] i want to mount / ro [19:50] hm. no [19:50] if you're on quantal i'd recommend using a hook. but you're probably not [19:50] i could be. [19:50] you have documentation on "hook" ? [19:51] no manpage update for it yet, but documentation is in with the example in the package for now [19:52] i.e. /usr/share/lxc/hooks/mountcgroups === nxvl_ is now known as nxvl [19:52] smoser: can you just specify /var/lib/lxc/q1/rootfs/ as the target? [19:52] or do you not know the container name at the place where you'd need it? [19:52] actually, [19:53] i could specify that as the target, sure. it'd be hacky, but i could do that. [19:54] that, or you have to use /usr/lib/x86_64-linux-gnu/lxc === dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates === glebihan_ is now known as glebihan === cpg is now known as cpg|away === AaronMickDee_ is now known as AaronMickDee [20:58] i tried to find the virtual package "apache2-dev" on quantal but it's not available. it's available on precise and on debian. why is it not available on quantal? is this a bug? [20:58] Hey guys. New to Ubuuntu Server. Trying to install it on our server at work due to us getting frustrated with FreeNAS constantly failing. However I'm stuck at the install process. For some reason network autoconfiguration has failed, even though it's connected directly to our router. I have the option of doing it manually or continuing without it. What should I do? === cpg|away is now known as cpg [20:59] the router has a DHCP enabled? [20:59] it might be a hardware problem... [21:00] i mean, it might be that the network card isn't handled by ubuntu correctly [21:02] Tellmarch: Well I've tried both the onboard ethernet port, and the PCIe ethernet port in the server and I get the same thing. However if I attempt doing it on one of our workstations that's connected to the same router, it does the DHCP dance just fine and continues on. And our workstations are connected via onboard ethernet ports. [21:02] And might I add that our servers and workstations have the same hardware save for the servers having a crap ton of 2TB HDDs in them [21:03] psionicsin, have you attempted giving it a static IP? [21:04] you can try the manual configuration and see if you get network access this way [21:04] but if not, you might have to add a driver or something to recognize your hardware properly [21:04] tgm4883: I'm currently in the very bieginning on installation stages. I still have an 8-16bit screen right now. [21:05] Tellmarch: how would I go about doing it manually. i'm not sure what to look for. [21:05] psionicsin, so? it gave you the option to set it manually [21:05] IMO, all servers should have a static IP anyway [21:05] maybe you can switch to another console, and see the output of lspci (the line about the network adapter), and maybe dmesg | grep -i eth0 or something... [21:06] see if everything looks all right there [21:08] tgm4883: Yes. Although I haven't selected it yet, the server always used the IP "192.168.1.68". It then asks for my router gateway. Since we use windows here I did ipconfig in my prompt to find the router gateway and subnet mask. However after all of that it asks for name server address. THIS is what I have no clue about,. [21:08] DNS? [21:09] ipconfig /all in windows will tell you the DNS used [21:09] psionicsin, it's asking about the DNS server, do 'ipconfig /all' in windows as Tellmarch has suggested [21:09] likely though, it's the same as your router [21:10] that depends, it's not the same for my server [21:11] You guys are both right. I'm generally a Mac guy so I'm so lost lol. The DHCP server is also he router so I'm finding it hard to understand why it can't find something it's directly connected to [21:12] Ok now I've made it to where it's asking me to enter the hostname for the system. freenas has been automatically filled in. 0_o [21:12] Tellmarch, while that is true, I sometimes make assumptions based on the technical level of who I'm talking to [21:13] tgm4883, Tellmarch: Is it safe to enter anything into the hostname field? This is just what the server will be called over the network correct? [21:14] yes, you can choose your name [21:14] psionicsin, my feeling is it is either A) bad cabling/port between the server and the router or B) the NIC in the server isn't recognized by Ubuntu [21:14] psionicsin, yes, but don't use underscore [21:14] Tellmarch: why do you think that the name freenas populated that field before i could enter anything? [21:14] that is the most common mistake I see people do [21:15] tgm4883: do you think cable length has to do with it having a hard time? [21:15] psionicsin, probably not [21:15] unless it is really long [21:15] 100ft [21:15] eg. I think the limit is ~300 ft [21:15] ok [21:16] tgm4883: ok now it's asking for a domain name.Never encountered anything like this with freenas. What exactly is this asking of me? [21:17] psionicsin, in windows, when you do an 'ipconfig /all' do you see anything listed for "Primary DNS suffix"? [21:18] tgm4883, no nothing is listed [21:18] then you can leave that blank [21:19] tmg4883, now I just have a blank purple screen with a white bar at the bottom. no indication of anything going on. [21:20] well, the fact that the DHCP didn't work was a sign of trouble... [21:20] can you switch to another console (ctrl alt F2) ? [21:21] and maybe start by ping www.google.com ? [21:22] Tellmarch, did that. Says www.google.com is alive! [21:22] so the network is working, at least... [21:23] I'm not sure what could block the install process then... [21:24] I don't recall what it does after network [21:24] Tellmarch, right. I'm not too savvy on servers and unix/linux systems...but I'm sure a network problem wouldn't stop software from being installed. Seems weird. [21:24] which is odd, as I've installed 3 servers today [21:24] psionicsin, did you verify the checksum of the ISO you downloaded and burnt? [21:24] Is there anything I can type into this console to find out something> [21:25] tgm4883 no...care to inform me how I do that lol? [21:25] psionicsin, well, the end of dmesg might tell something, sometimes... [21:25] psionicsin, are you running 12.04? [21:25] like if you see a big ERROR message... but it's not very likely [21:26] yea I'd probably just check dmesg or /var/log/syslog [21:26] see if it's trying to do something [21:26] i'm getting 403 forbidden on my web server...does this look correct? [21:26] -rw-r--r-- 1 www-data www-data 49 Sep 5 16:27 index.html [21:26] I also seem to remember there was something in /var/log/installer or something like that [21:27] ehnde, yes, this should work... what are the permissions on the folder? [21:27] Tellmarch: folder permissions are drwxr-xr-x 2 www-data www-data [21:27] ehnde, is that still at /var/www/? [21:28] tgm4883: /srv/www [21:28] what are the permissions on /srv [21:28] err.../srv/ [21:28] tgm4883: permissions on /srv are drwxr-xr-x [21:28] that doesn't look good :| [21:29] if all the permissions in the tree are like this, it should work... [21:29] yea that looks like it should work [21:29] tgm4883, Tellmarch: no big error messages at all from what I see. Near the end it's talking about eth0 and eth1 (0 is the PCIe card, 1 is onboard). A bunch of "link not ready"s and "link down"s. Ends with eth0 saying "link becomes ready" and then saying no IPV6 routers present. [21:29] maybe i still see the 403 because of browser caching [21:29] we're assuming you pointed the apache site at /srv/www/sitename [21:30] psionicsin, so nothing wrong in dmesg... have you tried the installer log? less /var/log/installer i think [21:30] psionicsin, that all looks fine [21:30] Tellmarch, says les not found [21:30] less* [21:30] hmm nevermind i remembered wrong [21:30] maybe the problem is group ownership [21:30] during the install it's in the syslog [21:31] nginx is running as 'nobody' [21:32] but everybody can read these files, including "nobody" [21:32] tgm4883, I flipped back to the install screen right quick and still a blank purple screen. [21:32] try hitting enter? [21:32] psionicsin, you're not running the graphical installer, right? [21:32] Tellmarch, graphical installer doesn't prompt for IP information [21:33] [21:33] Tellmarch I'm running the first option from the moment it booted the CD that said "Install Ubuntu Server". And yes version 12.04 [21:34] what was the last working step? [21:35] it seems it should go hostname -> timezone -> disk.... [21:35] hostname was the last step I beileve [21:35] tgm4883, is regular Ubuntu capable of running as a ROCKSOLID NAS? [21:35] psionicsin, IDK, I don't do that [21:35] I would assume so [21:36] because freenas on this server has always been buggy. You look at it wrong and it drops the mount. [21:36] regular ubuntu is capable of doing everything that ubuntu server does [21:37] Tellmarch: Would you be able to guide me through installing regular ubuntu to a usb key and using the 4 2TB internals in a RAID5/Z@ config to share with everyone on the network? [21:38] Z2* [21:38] no, I've never done that :-) [21:39] but all the packages available in ubuntu server are also in the regular ubuntu distribution [21:39] the difference is really that ubuntu server install less things by default, and the installer is different (not graphical for the server) [21:39] Tellmarch, ahh ok...this is all so confusing for no reason lol [21:40] software raid? [21:40] https://help.ubuntu.com/community/ServerFaq#What.27s_the_difference_between_desktop_and_server.3F [21:40] the difference is explained here [21:40] tgm4883, yesh [21:40] I'm assuming so, since hardware raid would be configured before the OS [21:41] tgm4883: short answer yes, long answer starts with 'it depends.....' [21:42] xnox, it depends? [21:42] regarding RAID or something else? [21:43] tgm4883: depending what you count as OS: UEFI/bios, grub, initramfs or real root. Whether the "hardware raid" is actually external hardware raid, or something activated by dmraid/mdadm (e.g. Intel Rapid Storage, DDF, etc). [21:43] tgm4883: and if your $rootfs is not on raid, you can for example spin it down and shutoff [21:43] xnox, does anyone actually consider dmraid/mdadm hardware raid? [21:44] if dmraid/mdadm is using external metadata formats (as in _not_ LINUX_RAID_MEMBERS, then why not?!) how is Intel Rapid storage or the industry DDF standard any different from other hardware raids? [21:45] * tgm4883 shrugs [21:45] you can setup Intel / DDF outside the OS (e.g. on the hardware controllers / bios) as well as from within mdadm/dmraid. [21:45] it's just an on-disk format and API. [21:45] it's been awhile since I looked at any of that, and it was a bit confusing with hardware raid/software raid/ fake raid [21:46] I think those names are a bit silly & misguided. With all the TRIM optimisations landing in the kernel and mdadm I am not sure which of the 'three' are better.... [21:57] Tellmarch: will I have any trouble installing to USB key? I've never messed with any version of Linux before. [21:58] you mean you want to run ubuntu from a USB key? in production? on a server? [21:58]