[00:00] im going to give it to a small business.. lol :S Ill just tell them that this is a great moneysaver.. screen and kebaord not even neccesary ehh possible :D But servers should only be connected with etehrnet and power anyway.. :) [00:04] can someone help me out with an ssh key problem [00:04] im trying to ssh-copy-id [user@host] from my friends machine to my server, trying to allow him access..but i keep get pubkey denied error messages [00:04] can i just copy/paste his rsa key into his ~/.ssh/authorized_keys on my server? [00:16] OK hooked my server up with a keyboard and a screen.. The only thing that shows up on screen when I boot it is: fsck from util-linux-ng 2.17.2 /dev/sda1: clean 765765/87658765 files, 765765/7657658765 blocks [00:18] when I press escape Ubuntu 10.14 . . . . (startupscreen in ascii shows up) [00:20] im now in tty7 (always used ssh or tty1 before this problem appeared.. tried to start from older kernel.. didnt work either.. :\ [00:23] searched all the web for answers too :\ [01:11] anyone know where I can get support for the openssh-server package/sshd? [01:12] specifically with a security question [01:32] ok who knows IT systems management really well? :) === EvilTrek is now known as Mithos === Mithos is now known as EvilTrek [01:41] EvilTrek: fire up an openbsd system and it will point you to their bug tracking... "approach [01:41] or, you know: http://openssh.org/ [01:42] hi guys framebuffer is nessesary on a server edition?? [01:43] whi wirte lsmod vga16fb is listed... [01:43] when write * === NightDragon is now known as Underbyte [01:44] I can blacklist vga16fb right?? having some problems with boot and trying to find the error. [01:47] bfcon is also running... hmm strange.. I havent activated any graphics that I know about :\ [01:54] trondpeter, while I think you could disable it without issues, it is enabled by default [01:55] I'm trying to get libpam-ldap authentication working on 10.04. 'id username' works for most users, but fails for user without the givenName attribute with 'no such user'. any idas? [02:32] Hey. [02:33] Erm, I don't know if there is a SFTP channel or something to post this in, but... [02:33] how can I use SFTP if it's not auto-sudo? [02:33] Shapeshiftr: erm.. .what? [02:33] SFTP is a module of SSH. [02:34] Shapeshiftr: SFTP doesnt need sud o.o [02:34] sudo* [02:34] The SSH channel is #openssh. [02:34] I get a "write permissions denied" error whenever I try to copy any files from this compuier over. [02:34] depends where you're copying to [02:34] oh... [02:34] 11:34 tell us EXACTLY 1) what you typed; 2) what you saw; and 3) what you expected to see. [02:34] Just a home/user/ [02:34] Linux to Windows? [02:34] Windows to linux. [02:34] or Linux to Linux? [02:34] Windows to Linux? [02:34] Yeah. [02:35] Shapeshiftr: make sure you have ownership of /home/user [02:35] Error: /home/administrator/levels/two.lvl: open for write: permission denied [02:35] Error: File transfer failed [02:35] oh... [02:35] heh [02:35] my server doesnt want to boot locally (ssh works) it "hangs" saying fsck from util-linux-ng 2.17.2 /dev/sda1: clean 50667/366480 files, 320366/1454676 blocks Anyone can help me with a solution? looked everywhere for help.. [02:35] yeah use sudo there Shapeshiftr, because sftp doesnt mean sudoftp :P [02:36] trondpeter: are you running lucid? [02:36] Shapeshiftr: since I assume your user you're using doesnt ahve ownership of the destination folder /home/administrator [02:36] I've logged into the SFTP as administrator [02:36] No, I do. [02:36] I AM administrator [02:36] hrm interesting... [02:36] Shapeshiftr: does an SFTP GET work? [02:36] check ls -l /home/ [02:36] that too [02:37] hrm, twb ? [02:37] make sure that the administrator folder though is owned by your user [02:37] how would I go about doing that? [02:37] yep server edition with samba kerberos authentication in a winblows AD domain [02:37] Shapeshiftr: its a function: within SFTP... GET [02:37] ? [02:37] i think [02:37] um, EvilTrek , I've logged in to the STFP client as administrator [02:37] So of course I have permissions :3 [02:37] ls -al ? :) [02:38] trondpeter: not fer ya, :P [02:38] chown user:group file [02:38] Ok, so I'm using FileZilla [02:39] I drag and drop a file from my computer to the remote server computer. [02:39] And, write permission error. [02:39] :\ [02:39] What does ls do, btw? [02:40] Shapeshiftr, just because you _should_ have permission doesn't mean you _do_ have permission [02:40] ls -al will tell you who owns and has permission to the files [02:40] Well, how would I not have write permissions for my own home folder? ... [02:41] Ok, running, one sec. [02:41] you could have write to /home/administrator but not /home/administrator/levels [02:41] or, /home/administrator/levels/two.lvl specifically [02:41] trondpeter: did you upgrade in-place from a previous release? [02:42] Ah, the folder i'm trying to copy to is set to root. [02:42] Yeah, /levels [02:42] How would I change that? [02:43] Shapeshiftr, sudo chown -R administrator:administrator /home/administrator/levels [02:43] 3 seperate lines? [02:43] *separate [02:43] one line [02:44] Erm, I mean, spaces in between each line? [02:44] of course, that was a stupid question. [02:44] exactly as I have written it, in a terminal [02:44] that sort of thing happens when you use sudo to create directories or files [02:45] Ah... [02:45] I'll remember that for next time. [02:45] not a big deal, but you need to fix it if you want user-level access [02:45] It's a personal server, I'm really the only user besides root [02:46] New bug: #604845 in dhcp3 (main) "cannot connect to wireless networks using iwl3945 and network-manager" [Undecided,New] https://launchpad.net/bugs/604845 [02:46] Ok, thanks everyone. [02:46] Later. [02:48] no one gave shapeshifter the website to the documentation :) [02:51] bogeyd6: what documentation? [02:51] his problem was a typical permissions issue [02:51] I wasn't aware there was any documentation to recommend there [02:51] RUTE, perhaps [02:53] anyone wants to see my dmesg? :p http://pastebin.com/EanDAvmp Sorry fell out a bit.. might have missed some messages [02:55] safe to blacklist lp module? [02:57] trondpeter: so the machine booted fully? [02:58] lp is the parallel port; I wouldn't bother blacklisting it. [02:58] !serverguide | twb [02:58] twb: The Ubuntu server guide may be found at http://help.ubuntu.com/10.04/serverguide/C/ [02:59] bogeyd6: I don't think the server guide covers POSIX DACs. [02:59] yeah,, i acess it from ssh. but not possible to access it locally [02:59] twb, i was talking about asking about "ls"and etc.... [03:00] trondpeter, did you try pressing escape? [03:00] for some reason, lucid server likes to add the stupid graphical loading screen and hiding all messages [03:00] that might be hanging you up [03:02] qman__: it's a "feature" that we can't remove :-/ [03:02] Even if you comment out splash in the boot options, it still runs plymouth -- all that you disable is the chvt 7 [03:02] yuck [03:03] first thing I always did on my servers was remove 'quiet splash' from the boot line [03:03] invaluable when things break [03:03] AND, in plymouth, hitting ESC makes it print the full history, without clearing the screen first -- so if you only have five lines, and you hit ESC a few times, it prints it over and over in a confusing way [03:04] You also have to opt-out of fbcon now, and you *can't* opt out in d-i images. [03:04] I understand that vanity is an issue for desktops, but for server, come on [03:04] qman__: allegedly "it's not just splash" === dendro-afk is now known as dendrobates [03:10] I would be happy if they simply added an option that with a specific boot line it shows all messages the old way [03:10] I don't really care what's under the hood as long as I get the information [03:11] the way it is now is a total pain [03:51] Greetings! I have a question regarding network aliases for eth0, do they all have to be in the same subnet? If they do, how do I create eth1 [03:52] ie, I just got new range of ips from my ISP, but I'm having trouble adding them to my interfaces file === NightDragon is now known as Underbyte [03:55] can anyone please help with ip addresses? [04:24] hey guys [04:24] what happenin === lifeless_ is now known as lifeless === Underbyte is now known as ND-smoke [05:23] hello, anyone here? [05:27] i have a question/issue, is anyone available to help? [05:38] i install ubuntu server 10.04, and set up ssh to log in remotely, everything installs and runs without a hitch.... when i restart my computer, after post, i get what looks like a blinking cursor " _ " then my screen goes blank and nothing else happens [05:57] !ubotu [05:57] Hi! I'm #ubuntu-server's favorite infobot, you can search my brain yourself at http://ubottu.com/factoids.cgi - Usage info: http://wiki.ubuntu.com/IRC/Bots [06:15] i cant seem to get xen installed on ubuntu server because xen-tools is missing [06:22] hey guys, how do i check for the number of file descriptors compiled in kernel? [06:24] Underbyte: What problem are you actually having / trying to solve? [06:25] i'm filling out an IRCD application [06:25] and that is one of the questions [06:25] i've never heard a question like that [06:32] CppIsWeird, grab it from a karmic repo [06:33] CppIsWeird, its confirmed missing from lucid === bastidrazor is now known as bastid_raZor [07:41] Anyone know of any good templates for documenting patch panels/switches cabling? [09:32] hi all [09:32] please help me to install autoconf version2.6?' [09:33] i try with command :apt-get install autoconf? [10:36] New bug: #604941 in mysql-dfsg-5.0 (universe) "package mysql-server-5.0 5.1.30really5.0.75-0ubuntu10.3 failed to install/upgrade: il sottoprocesso post-installation script ha restituito un codice di errore 1" [Undecided,New] https://launchpad.net/bugs/604941 [10:40] hi, what is the ubuntu-server-qa/regression-test ppa about? [10:45] Hi there, Will ruby1.9 every become the default version in 10.04LTS server edition? [10:48] 10.04 is released, therefore it is stable. That's "stable" as in "does not change". So if it isn't already the default, it's unlikely to become so. [10:50] Cool thanks. [10:57] twb: that definition of "stable" is nonesense - feature-freeze != stable [10:57] but then, 10.04 _is_ fairly stable [10:57] Freezes happen before the release. [10:57] yeah, but "stable" != "feature-frozen" [10:57] As at the release, it becomes stable. There are exceptions (notably, security patches), but these are exceptions. [10:58] you said `That's "stable" as in "does not change"` [10:58] Precisely. [10:58] that's abusing the word `stable` [10:59] "does not change" is "feature-frozen", and has nothing to do with the word "stable" [10:59] Plonk. [10:59] the only other place I've seen mixup of that was from the AsteriskPBX guys, but then, that project is a mess and is pretty far from any type of stability [10:59] So am I correct in assuming that 'apt-get install ruby' will never install ruby1.9? [11:00] crankyadmin: there might be other repositories if you want it from a package [11:00] crankyadmin: on 10.04, that is a reasonably safe assumption. === schmidtm_ is now known as schmidtm [11:01] Cool [11:02] crankyadmin: for a security update to change ruby from 1.8 to 1.9, it'd have to be something both critical and impossible to backport. [11:02] Of course, it you have stuff like lucid-backports or PPAs enabled, you might accidentally get 1.9 [11:02] crankyadmin: http://www.ruby-lang.org/en/downloads/ [11:03] crankyadmin: sudo apt-get install ruby1.9.1-full [11:03] it's all in lucid already [12:47] So I changed my ssh and sshd to RSAuthentication yes and password authentication to no. However I still invalid user messages in my auth.log... did I miss something? [13:38] If you change ssh_config and sshd_config to RSAuthentication yes and Password Authentication to no why would auth.log still be showing invalid user attempts? (My key is several thousand bits long and one week old I highly doubt its been brute forced yet) [13:57] is there a CLI only picture comparison program that will search through a directory of a few hundrad thousand files and look for duplicates? [13:58] md5sum ? [13:59] Jeeves_Moss: fdupes can do that, based on checksums, but if the picture is a duplicate with a different resolution, that will not work === dendrobates is now known as dendro-afk [14:16] what do u guys recommend for setting up automatic security updates on a server? [14:22] DrPoO: unattended-upgrades maybe... [14:22] joschi, is that a package? [14:23] DrPoO: yep [14:25] is there a quick way to get the user name who is running a process? [14:25] top? [14:25] ps uax ? [14:26] I now the name of process: "named" [14:26] So how to learn who is running "named" [14:26] ps uax | grep named [14:27] The first column is the user running it [14:27] bind, usually for named [14:27] Jeeves_: thanks, its what I need [14:28] * patdk-wk is lazy, stat -c %U /proc/`pidof named` [14:29] though, I normally use the uid numbers [14:29] patdk-wk: That's not lazy, that inefficient :) [14:29] well, depends on what your using it for [14:29] I normally use it in scripts [14:29] ps aux | grep named would need something to filter columns [14:29] I guess I could modify the ps output also [14:29] blain: If you change ssh_config and sshd_config to RSAuthentication yes and Password Authentication to no why would auth.log still be showing invalid user attempts? [14:30] zul: apachelogger pointed me your way about a problem we've uncovered [14:30] the problem is that libmysqld is distributed in a separate pic-enabled package [14:31] but, mysql_config --cflags doesn't have -fPIC [14:31] because the main package isn't built with -fPIC [14:31] so if you use mysql_config to get your build requirements your build will fail [14:32] jefferai: can you open a bug in launchpad [14:33] against what package? [14:33] mysql or the libmysqld-pic one? [14:35] mysql [14:36] ok [14:40] zul -- sorry, against mysql-server or mysql-dfsg? === dendro-afk is now known as dendrobates [14:40] hm [14:41] mysql-server doesn't exist.. [14:42] mysql-5.1 please [14:44] hi I get this error in my slave dns server: http://pastebin.com/XU1QRBTN [14:45] I am using 10:04 server with KVM and virsh - trying to "hot add" an LV "disk" to a guest, but having problems. [14:45] If I use virsh - and "attach-disk", the disk is "attached" but the gurst hangs and I have to destroy it. [14:46] attach-disk DOM /dev/vgserver/disk vdb [14:46] This creates a disk XML block : [14:46] [14:46] [14:46] rahman, fix your master dns server [14:46] But the guest is dead now. [14:47] Does anyone use "attach-disk" with KVM/virsh? [14:48] hmm, actually I think it's the tmp file, guess named doesn't have write permission [14:48] If I try and restart the guest (start), I get an error : [14:48] patdk-wk: master has allow section for slave and it doesn't report any error in syslog for slave [14:48] error: internal error unsupported driver name 'phy' for disk '/dev/vgserver/rum-disk-server' [14:48] Jul 13 16:18:09 artvin named[647]: client 79.123.252.52#46624: transfer of 'artvin.edu.tr/IN': AXFR started Jul 13 16:18:09 artvin named[647]: client 79.123.252.52#46624: transfer of 'artvin.edu.tr/IN': AXFR ended [14:48] ya, I just noticed the tmp file failed [14:48] There is very little documentation for the "--driver" and "--subdriver" options to "attach-disk" ... [14:48] fix your filesystem permissions [14:49] do I need to give -wrx permissions to "bind" user for /tmp dir? [14:50] heh? [14:50] zul: ok, done: https://bugs.launchpad.net/ubuntu/+source/mysql-5.1/+bug/605021 [14:50] Launchpad bug 605021 in mysql-5.1 "mismatch in mysql_config output and libmysqld cflags" [Undecided,New] [14:50] what exactly did you do to your /tmp? [14:50] and I wouldn't know, I don't use named, I hate named [14:51] last I saw, named used chroot, dunno if that is the case in ubuntu [14:51] jefferai: thanks [14:51] patdk-wk: :) well its a fresh installed bind9, I don't know where "named" is trying to write tmp-mAacM75gcD [14:52] zul: sure...now I just have to figure how to work around this breakage for people for now [14:52] or trying to read it from [14:52] I wouldn't know either :) [14:52] patdk-wk: thanks anyways :) [14:54] jdstrand, ping [14:56] New bug: #605021 in mysql-5.1 (main) "mismatch in mysql_config output and libmysqld cflags" [Undecided,New] https://launchpad.net/bugs/605021 [15:00] smoser: yes? [15:00] you're archive admin? i think. [15:01] https://bugs.launchpad.net/ubuntu/+source/cloud-init/+bug/605013 is in need of help if you had some spare time. [15:01] Launchpad bug 605013 in cloud-init "[MIR] cloud-init (legacy-grub-ec2)" [Undecided,New] [15:01] smoser: I am, but my day isn't until Friday (see ArchiveAdministration). what do you need? [15:02] jdstrand, sorry, i didn't know of days. i'm ok to bug jonathan or wait. [15:03] smoser: I should mention that I (or jonathon for that matter) are not on the MIR team [15:03] oh. sorry. yeah, i was confused. [15:03] so never mind then. thank you. [15:05] jdstrand, can you verify for me that legacy-grub-ec2 should at least be in universe at the moment ? it does'nt seem to be in the archive right now, but i'm under the impression that no other buttons need pushing to get it there. [15:05] smoser: sure, np :) [15:05] https://launchpad.net/ubuntu/maverick/+queue?queue_state=3&queue_text=cloud-init shows component universe for that version [15:06] smoser: is legacy-grub-ec2 a binary provided by the cloud-init source? [15:06] yes [15:06] smoser: was legacy-grub-ec2 added in a recent cloud-init upload? [15:06] yes [15:07] before this morning it was in the NEW queue [15:07] smoser: https://launchpad.net/ubuntu/+source/cloud-init shows grub-legacy-ec2 (as opposed to legacy-grub-ec2) for 0.5.12-0ubuntu5 [15:08] grub-legacy-ec2 is correct. [15:08] updating bug [15:09] smoser: it is in LP. when did it get deNEWd? [15:09] i'm a moron [15:09] sorry [15:09] * smoser apologizes for stupidity [15:09] thank you for catching my bad package name. it is in universe. [15:10] smoser: heh, no worries. happens to the best of us :) [15:50] In sshd what is pubkey authentication? [15:51] rsa/dha key exchanges [15:51] OK so pubkey must be allowed to use RSA or DSA [15:52] yep [15:52] If you change ssh_config and sshd_config to RSAuthentication yes and Password Authentication to no why would auth.log still be showing invalid user attempts? [15:52] AuthorizedKeysFile %h/.ssh/authorized_keys [15:53] do you have any authorized keys? [15:53] Yes I have a working key, I am just curious as to why random (hackers or bots probably) are showing up in auth.log with invalid user attempts [15:54] crackers I should say [15:55] oh? [15:55] cause it's invalid :) [15:55] lol yeah I guess that makes sense. [15:56] I was expecting it to say invalid key rather than invalid user I guess [15:56] is the user valid? [15:56] negative [15:56] I don't believe sshd exposes the fact that keyboard-interactive is disabled, just like sshd won't expose that the username someone is attempting doesn't exist. [15:57] kozmund: Yes for security that would make sense [15:58] hmm, I don't get that [15:58] with mine, configed with passwordAuthenication = no [15:59] it doesn't attempt passwordauth at all, on the client side [15:59] no error messsages logged [15:59] mine too. Some attempts say invalid key, which would be expected, but some say invalid user which I guess I don't how they got that far with password attempts disabled [16:00] ya, I get invalid user [16:00] but the user check happens long before the password check [16:01] as it needs to know hte user, to find the directory for the certs [16:01] before the key check even? [16:01] yes [16:01] you can't find the key, without looking in the users home dir for them [16:01] Ah, that makes me feel better [16:02] I was worried I had missed disabling something. I have had so many brute-force attempts that I felt I needed key based authentication [16:03] heh, simple firewall based rules normally do good with that [16:03] if not, fail2ban [16:04] Yeah I would like to set a number more like 3 === dendrobates is now known as dendro-afk [16:08] Maybe set a banner too.. with idle threats of prosecution. (since most of these attacks are coming from China and Korea) [16:12] what is the dfeault working directory for bind in 10.04? I get a named[23441]: the working directory is not writable when I restart bind and could'nt find where the option is set. [16:15] alvin, sorry, I was AFK, I'm looking for something that'll take into account the diff sizes, res, etc. I have ~300,000 pics that I don't want to manually go through [16:20] ttx, you happen to know much about the uec provisioning cgi script? [16:21] ttx, i'm running into a problem with it executing powerwake, tried to contact kirkland but he appears to be afk for a while [16:25] I'm running 10.04 server, 64-bit, with software RAID5, and i'm finding that occasionally some operations take like 10 seconds to complete [16:25] for example closing vim or logging out of SSH [16:25] but only occasionally, and just that one operation [16:25] Jeeves_Moss: I found a utility 'findimagedupes' in universe. I didn't try it, but it looks like what you're searching for [16:26] and then it's back to being funke [16:26] has anybody else hasd this happen before/ [16:31] hi guys. I have a program that's giving me a weird error in bash. I've seen it before but I can't remember what caused it [16:34] http://pastebin.com/DxXtpA6K [16:34] I don't understand why it doesnt' execute pclient. It's chmodded 755 [16:36] alvin, thanks, I'll research it. === dendro-afk is now known as dendrobates === dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates [16:43] is there a way of creating a temporary partition in swap? [16:53] drpoo, tmpfs :) [16:53] patdk-wk, tmpfs? [16:54] ya [16:54] can one create a tmpfs on a ssd? [16:54] tmpfs is a swap backed ramdisk [16:54] nice [16:54] you said in swap [16:54] ya my swap is on an ssd [16:54] well, it's a ramdisk that can get swapped [16:56] New bug: #605062 in autofs5 (main) "(lucid) autofs had a strange behaviour with paths" [Undecided,New] https://launchpad.net/bugs/605062 [16:59] Question about KVM and LV - In Xen, an LV is seen/inserted into a guest as a partition (e.g. xvda1). In KVM, an LV is is seen as an entire disk in the guest i.e. /dev/vdb. [16:59] actually, in xen it's optional [16:59] I do it both ways [16:59] I personally feel inserting it as a partition is evil [16:59] Is there a problem (downside) to not putting a partition table on the LV inside the guest (vdb1) and just a filesystem on the entire device vdb? This seems "cleaner". [17:00] sherr, generally you will be fine [17:00] but some things will freak out, and *install* a partition table for you [17:00] Thanks - it seems to work OK, but just planning. [17:00] I have run without a partition table on systems for awhile, and it works [17:00] I *am* concerned about compatibility and data safety. [17:00] but when I tried openfiler, it wipes the drive if it doesn't have a partition table [17:02] Thanks. I will note that concern about things like OpenFiler - don;'t think it will affect me. [17:02] ya, openfiler I know will screw it up :) [17:02] haven't had anyhting else yet, but I normally don't use gui stuff [17:03] I think I am going to have to search LP for bugs and probably log one though - trying to use "virsh" to "attach-disk" (hotplug) will hang the guest. [17:03] Not a show-stopper for me, but a shame. [17:03] ya, I haven't used kvm at all, but lots of xen [17:04] switching them all over to vmware though [17:04] I use Xen too ... but have had to switch to KVM because of new hardware. [17:05] Debian Stable/Xen - unsupported h/w. Debian Testing/Xen - unstable ... so, on Ubuntu Lucid 10:04 now :-) [17:05] I'm still in 8.04 [17:05] cause of xen [17:05] once the move to vmware is good, it will all be 10.04 [17:06] some have already moved [17:08] I don't know VMware well enough. And I need to provision a new server this week - with something I can support ... [17:09] I exported a database before running a script which caused problems, now i want to restore it to what it was before I ran the script. I exported databse in phpmyadmin, so how can I go about restoring it? [17:16] oru_work: don't know phymyadmin, but isn't there a "restore db" section/button as well? === dendrobates is now known as dendro-afk [18:54] How do i disable the initial password to my server- if i remote reboot, I lose connection, and have to manually take my keyboard and plug it into my server box and enter a password... [18:56] zeeeeee: like, a BIOS password? [18:57] hmmm it seems like it's some sort of hd access password- once I enter it, it gives me the ubuntu-server main info page [18:58] but there's no way (that I've figured out) to enter the password remotely [18:58] upon rebooting [18:58] surely it was some setting I enabled during install [19:07] you can't [19:07] if that is a harddrive password [19:07] that has to be entered long before ubuntu starts [19:08] to disable it is annoying, if possible [19:13] it's harddrive encryption that i installed w/ubuntu server. oi vei.. is there some encryptfs file I can delete? [19:14] whole disk encryption? [19:14] normal encryption setup by ubuntu is on the user level, not the whole system [19:16] seems really pointless to have an encrypted system, where the system can boot without a password [19:16] I can only guess your used cryptsetup to do it [19:16] and in that case you need to modify your /etc/crypttab file [19:18] http://linux.die.net/man/5/crypttab [19:18] 3rd field, sets the password [19:19] you will have to remake your initrd files [19:26] anyone aware of a network stats program similar to top, and iotop? [19:27] iftop? [19:27] iftop [19:27] there we go :) [19:27] trafshow? [19:27] heh [19:27] there are like 10+ of them [19:27] patdk-wk, thanks [19:27] or bmon :) === XyliosPC1 is now known as Xylios [20:01] soren: did you ever publish gfsharefs ? [20:01] does anyone know the specific directive for a ks file so that server only installs base by itself? not development or other parts, just base [20:03] would it be %packages then @base without anything else? [20:03] in 10.04 it seems to install a bunch of development libs if I do that [20:06] lowridah: you could just use the mini iso. ;) [20:07] that's a workaround, I want to know how to do it right =) [20:07] i'll just bang my head on it some more [20:08] (right is probably by passing a preseed file I know) === MTeck is now known as MTecknology [20:19] kees: I doubt it. [20:19] kees: It seems not :( === ericboehs is now known as Guest76109 [20:31] if you moved your /var/www to /srv/www (www is it's own partition) would you consider symlinking /var/www ? [20:32] or maybe mount --bind [20:39] how can I check php log ? [20:40] webPragmatist: I'd change the configs [20:40] SpamapS: well i have to do that anyway because i have like a jillion sites :) [20:40] webPragmatist: but given the amount of stuff that expects to be served from /var/www ... a symlink is probably in order. :) [20:41] SpamapS: doesn't httpd stuff like a default site in there [20:41] oh no it's just this index.html [20:41] soren: if you send me a copy, I'd be happy to publish it! :) [20:42] webPragmatist: no [20:42] php logs anyone ? [20:42] um. [20:43] turn on error reporting because I don't think there are php logs [20:43] you should see the errrors in the httpd logs [20:43] or edit your php.ini and add it [20:44] here's a really good link about it [20:44] http://lmgtfy.com/?q=php+log+file+location+ubuntu [20:45] !google | lowridah [20:45] lowridah: While Google is useful for helpers, many newer users don't have the google-fu yet. Please don't tell people to "google it" when they ask a question. [20:45] lol [20:45] you seriosuly tell people not to google simple strings? [20:45] that's ridiculous [20:46] lowridah: Thats the way we do things around here. [20:46] kees: Appreciated :) [20:46] wow here kinda sucks then [20:46] kees: I wonder where it is :-/ === Underbyte is now known as Underbyte|away [20:49] soren: nooo [20:49] soren: no losing it! [20:51] kees: I'm thousands of miles from home right now. I'm sure it's on one of my laptops at home. [20:51] kees: Remind me in a couple of weeks. [20:52] soren: hah, okay. you in the states right now? [20:53] kees: Oui. Austin. [20:53] kees: I can see Dustin from here. [20:53] kees: :) [20:54] heh === Underbyte|away is now known as Underbyte [20:57] !away > Underbyte [20:57] Underbyte, please see my private message [21:05] i restarted my server, and apache did not appear to start as normal. i was able to start it manually, but i don't know why it didn't start on it's own, it worked ok before [21:06] i also noticed this yesterday on my laptop running ubuntu, that boincd did not start when I rebooted. is this a recent bug in ubuntu [21:06] not sure if that issue is related but it seems like it === Underbyte is now known as Underbyte|away [21:10] could be a race condition problem, but you'd have to do some testing to make sure === Guest76109 is now known as ericboehs [21:12] qman__, how would i test such a thing? [21:13] I'm not exactly sure of the best way to go about that, far as logging the information [21:14] but a race condition is when something (ie apache) tries to start before something it depends on (ie the network) has finished starting, and fails [21:14] if it attempted to start during boot and failed, would it be in any particular log file? i don't know very well what everything in /var/log relates to [21:14] but then would start fine once the system is booted and you try it manually [21:14] it would [21:14] probably messages or syslog [21:14] there would also probably be something in apache/error.log [21:21] qman__, what files configure what is run on startup? [21:22] I don't know where they're at now, since the change to upstart [21:22] argh [21:23] looks like I still have /etc/init.d/apache2 [21:24] how does it know to call /etc/init.d/apache2. is there a file that lists what to call in init.d, or what order? or does it just automatically call everything in etc/init.d [21:25] well, there used to be sysvinit configuration, but now it's all upstart [21:25] looks like apache is still started the sysv way, just through upstart [21:28] actually i had boinc on this server as well, and that doesn't start either [21:28] ah, it still has the rc?.d file structure [21:28] update-rc.d should work too [21:29] *sigh* i don't understand this well enough. i just don't get what could have broken it when it all worked fine before [21:30] dammit i'm a software developer not a sysadmin [21:31] well, if you still have /etc/rc[2,3,4,5].d/S91apache2 [21:31] it's telling apache to start [21:32] qman__, ://paste.ubuntu.com/463152/ [21:33] i don't have any rc*.d/*apache [21:34] well, that explains it [21:34] not sure how it happened [21:35] sudo update-rc.d apache2 defaults [21:35] ought to fix it [21:35] err [21:35] sudo update-rc.d apache2 defaults 91 [21:36] what's 91 for? [21:36] that's the ordering [21:36] without it, it defaults to 20, which could cause it to start before it can run [21:44] ahhhhhhh, cups is down too. nothing freaking works === Underbyte is now known as Underbyte|away [21:52] qman__, i was mistaken, that find command did not work like i though it did. I already have S91apache2 file === Underbyte|away is now known as Underbyte === dendro-afk is now known as dendrobates === Underbyte is now known as Underbyte|away [22:15] what is "committed" memory? [22:22] if i had to make an uneducated guess it would be memory that has been used by a process [22:22] or "committed" to a process [22:34] what is the pae in package linux-generic-pae ? i've never seen this before [22:36] New bug: #605172 in libvirt (main) "virt-manager fails if target system uses tcsh as root shell" [Undecided,Confirmed] https://launchpad.net/bugs/605172 [22:37] it has the same package description as linux-generic [22:41] is git unsupported in ubuntu server 10.04? [22:42] CppIsWeird: I use git all the time with no problem [22:43] what package do i install to get it? [22:43] sudo apt-get install git-core === Underbyte|away is now known as Underbyte [22:46] peeps[work], it has PAE enabled, which allows non-x64 kernels to use more than 4GB of RAM [22:46] oh [22:47] aside from performance tweaks, the only place it can cause problems is with badly written drivers [22:49] thesheff, thx [22:49] np === Underbyte is now known as Underbyte|away === Underbyte|away is now known as Underbyte [22:49] if you can't find a package you can always do apt-cache search packageName [22:50] still don't know how to fix my upstart issues [22:50] what service is not working throug upstart? [22:51] apache2, cups, boinc-client, maybe others that i haven't noticed yet [22:51] they worked fine days before. i restarted today and they did not come up automatically [22:51] well I really new to 10.04 but I have still been using /etc/init.d/ scripts fine for apache2 and boinc [22:52] and i tried another reboot and they didn't come up again [22:52] apache is still a sysv script, but upstart is doing all the sysv work now [22:52] eventually will the init scripts go away? [22:52] yes [22:53] i have some cron jobs i added recently, which might take a long time to run, could this possibly cause problems? [22:53] that's the only thing i can think that's really changed [22:53] apache should be instant [22:53] only if they run on startup [22:54] i just put them in cron.daily, cron.weekly, cron.monthly. i don't know if those get called on startup or not [22:54] what is the method for adding a service to startup...I'm assuming update-rc.d is going to go away as well [22:54] you shouldn't be starting services through cron [22:54] he isn't [22:54] well for startup at least [22:55] i'm not starting service through cron, i'm just saying the only thing that's changed is i added some cron jobs, they back up my files [22:55] oh ok [22:55] anything in cron.daily, etc would only run on startup if you start up at the time it picked arbitrarily to run daily jobs [22:55] which would only be about a one minute window in the day [22:56] do you still use update-rc.d for starting scripts on reboot? [22:56] services I mean [22:57] for sysv-style ones [22:57] not sure about upstart jobs [22:57] ok === Underbyte is now known as Underbyte|away [23:10] i tried asking for help in #upstart, but it's dead in there === dendrobates is now known as dendro-afk === Underbyte|away is now known as Underbyte [23:53] rshali?