[02:27] I'm rescuing a legacy 8.04 server, building a new array for it to live on. Anybody remember if 2.6.24 supports mdadm --metadata=1.2 ? [02:28] git indicates mdadm metadata was versioned as at 2.6.12 (as far back as git goes), but I can't see where support for 1.x was added === DWSR2 is now known as Guest36661 === Gnubie is now known as Guest31084 === wedgwood is now known as Guest73785 [04:28] PLEASE HELP: I need your help please. I have created automated ubuntu installed via dvd. For that I have created preseed file and kickstart file. Now I want to do the automated ubuntu install via usb stick. All the instructions so far I have found out seems usb installer which I think is different than automating installations. PLEASE HELP [04:35] rostam: using d-i or ubiquity? [04:38] twb: I have done that, I have already created automated installation, the media however is dvd, I copied the same image to usb and it did not work? [04:38] Which are you using? Are you using d-i? Or are you using ubiquity? [04:39] I am using d-i [04:39] OK. Are you using preseed, or kickstart? You mentioned both. [04:41] twb: it is combination of both, I do have kickstart file and also preseed file. I can use paste bin if you want to look at the files? [04:41] OK, I do not know about kickstart. [04:41] What I mainly know is from the Debian Installation Guide (apt-get install installation-guide-amd64). [04:42] All you have to do is add preseed= to the boot options, where is somewhere the installer can reach -- e.g. if it is a netinst installer, it can't be on a USB disk [04:42] You can also pull apart the initrd and put it in there. [04:43] Now, you said you're writing an existing DVD onto a USB key. Because ISO 9660 is not a writable filesystem, you'll have to either remaster it or fiddle around if you want the preseed= option to be present by default, without you having to retype it each time [04:43] twb: thank, but let me ask you this, The iso image burned on DVD which works for me, if I copied to usb as it is that should work or I need to do some extra activity to work on usb? [04:44] I don't know if current Ubuntu server install .iso is made using isohybrid. [04:44] If it is, you can write it directly to the USB key (e.g. dd if=foo.iso of=/dev/sdz) and it'll boot. [04:44] If not, it is more fiddly. [04:44] ISOs built using isolinux >= 4 can use isohybrid [04:45] twb: oh okay thank you, okay so where could I go check for isolinux version please? [04:46] Not sure how to detect it [04:47] Easiest would be to just try writing the ISO to a USB key and see if it's bootable [04:48] twb: you have almost answered all my question, specially the isohybrid which I was not aware of that. thank you so much. [04:50] rostam: but if it's isohybrid, it'll boot of USB but you won't have a preseed file on the same USB key, so you still need to deal with that [04:52] # dd bs=512 count=1 < /srv/iso/cdimage.debian.org/cdimage/release/current-live/amd64/iso-hybrid/debian-live-7.0.0-amd64-rescue.iso | file - [04:52] /dev/stdin: x86 boot sector; partition 1: ID=0x17, active, starthead 2, startsector 64, 1388480 sectors, code offset 0xed [04:52] ...that's how you tell if it's already run through isohybrid [04:52] twb: thank you thank you thank you. [04:52] I was lost completety [04:53] What I mostly do is grab the netboot d-i kernel and ramdisk, and put them on a TFTP server, and do the whole install off PXE boot [04:53] But for that you really want a local apt mirror to make it nice and fast [04:53] I think you can even apt-get install that stuff nowadays... [04:54] Hm, in debian there is stuff like this: debian-installer-7.0-netboot-amd64 -- I don't have a recent ubuntu server to compare [04:55] http://cyber.com.au/~twb/snarf/netboot is the crappy script I'm running on my 10.04 netboot server -- it has the URLs for the boot stuff. [04:56] great that is great reference for me to start with. [04:57] Strongly recommend you refer to installation-guide-amd64 docs also, that's where I learned most of this [04:58] (appendices of that doc) [05:01] ok will do that [05:02] If you're still stuck tomorrow, pm me directly -- I don't normally hang out here. [05:03] twb: sure will do that thank you again. I have been stuck on this for last a few days. thank you [05:42] Hello. I've got an ubuntu machine acting as a server for a web application (apache + tomcat, and postgresql backend) [05:43] Yesterday the machine was updated, and the following packages were upgraded: dpkg dpkg-dev libdpkg-perl libdrm-intel1 libdrm-nouveau1a libdrm-radeon1 libdrm2 libglib2.0-0 libicu48 liblockfile-bin liblockfile1 linux-firmware procps python-software-properties [05:43] Now the web application is acting sluggishly [05:45] CPU usage is minimal. Memory usage is ok I believe, there's a java process taking up 50% of the memory. [05:45] Does it normally take up 50%? [05:45] If you don't have performance monitoring already set up, it'll be harder to know what is unusual [05:46] Also desirable to find a quantitative metric, so you know when it's really fixed. Confirm that the symptoms first appeared after upgrade -- might not be correlated. [05:47] twb: I don't think java's memory grab is of any concern, that's how it ususally behaves, isn't it? [05:47] Check iostat, check for procs in D state, check logs for anything unusual/interesting. [05:47] foo357_: I think so, yes, but I boycott java wherever possible [05:49] Could also take some tcpdumps between user and app, and between internal components (e.g. pg and app) to see if there are obvious latency problems and if so, where they are. (tcpdump -s0 -wfoo.pcap, then wireshark -rfoo.pcap on a GUI system, Stats menu, also change the time column to be seconds-since-last-packet, I forget how.) [05:50] Worth pinging app's devs if they have app-specific diagnosis hints, esp. if there is long turnaround time to get a response from them [05:56] twb: hm, it seemed like the issue originated from the underlying network and some performance issue there. Ah, well, not my department :) === gaughen is now known as gaughen_afk [09:01] The dns-nameservers variable in /etc/network/interfaces appears to be working inconsistently. I've checked the documentation at https://help.ubuntu.com/12.04/serverguide/network-configuration.html -- Is there anything else I should keep in mind when running multiple bridge interfaces? [09:08] <_ruben> define "inconsistently" [09:08] <_ruben> vk01_: ^^ [09:10] @_ruben --- had to reboot the server last Saturday. Interfaces came up as planned [09:10] Forced to do a reboot today --- cat /etc/resolv.conf defined no DNS servers [09:11] It should be enough to define dns-nameservers under just br0, don't have to define it under br1 as well right? [09:12] *"Interfaces came up as planned" --- meant that DNS servers showed up when I did cat /etc/resolv.conf [09:12] <_ruben> Once should be sufficient indeed. Not working with bridges myself though, so not sure if there's any caveats involving those [09:13] Do you use static IP addresses with 12.04 server or newer? [09:13] <_ruben> Could try moving the dns-nameservers line to the definition of lo [09:13] <_ruben> Yes [09:14] I'll give that a shot, thanks. Wondering if it's worth removing resolvconf and doing it the old way [09:15] <_ruben> I've pondered that as well in the past. Then again, the dns-nameservers always worked for me. Just needed to get used to putting it into /e/n/interfaces :) [09:15] <_ruben> But like I said, no bridges here, do use a lot of bonding/vlan tagging/bare interfaces tho === shadeslayer_ is now known as shadeslayer === psivaa is now known as psivaa-lunch === zz_Gurkenmaster is now known as Gurkenmaster === psivaa-lunch is now known as psivaa === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha === freeflying_away is now known as freeflying [13:43] http://pastebin.com/d2GsbqWN - seems i can't remove a package. since this package is broke i can't seem to install, update, or upgrade other packages. === Guest73785 is now known as wedgwood === freeflying is now known as freeflying_away [14:36] hey anybody got a similir problem to me, (saucy, phpmyadmin) the current server dropdown is not changing the servers unless i first click on a link in the main pane === wam is now known as masses === masses is now known as wam [15:59] http://pastebin.com/d2GsbqWN - seems i can't remove a package. since this package is broke i can't seem to install, update, or upgrade other packages. [16:04] Hi, what should I do if I have a command looking in the wrong place for an executable? (/usr/local/bin instead of /usr/bin) === chuck_ is now known as zul_ === mdeslaur is now known as chuck_ === chuck_ is now known as mdeslaur [16:58] so i've installed both dhelp and dwww on my webserver but get the following *SIMILAR* errors in both (as well as my own webpages): The requested URL /cgi-bin/dwww/usr/share/doc/ was not found on this server. === gaughen_afk is now known as gaughen [17:26] What's the best and easy way to mange LDAP users? I tried to install phpLDAPadmin as mentioned in https://help.ubuntu.com/community/InstallingphpLDAPadmin but I failed to configure it === psivaa is now known as psivaa-afk [17:56] so i've installed both dhelp and dwww on my webserver but get the following *SIMILAR* errors in both (as well as my own webpages): The requested URL /cgi-bin/dwww/usr/share/doc/ was not found on this server. [18:00] rtdos: anything in the logs? [18:00] sarnold: how would i view the logs? === chuck_ is now known as zul [18:01] rtdos: check /var/log/, most services log to files in there. [18:02] ok gimme a sec. [18:05] ok, sarnold, here is the paste of /var/log/apache2/access.log: http://pastebin.ubuntu.com/6278330/ [18:06] andol, sarnold, here is the paste of /var/log/apache2/error.log: http://pastebin.ubuntu.com/6278339/ [18:06] ersi, sorry andol: that was a typo. [18:06] damn spelling checker. === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha [18:54] In varnish 3, how do I tell the difference between a backend actually giving me a 503 response, and the backend being too slow to answer, and hitting varnish' timeout for the backend? [18:54] (in vcl specifically, not varnishlog) [19:54] so i've installed both dhelp and dwww on my webserver but get the following *SIMILAR* errors in both (as well as my own webpages): The requested URL /cgi-bin/dwww/usr/share/doc/ was not found on this server. [20:06] I have a really simple mysql backup script that works when I invoke it manually, but cron isn't executing it. I don't see anything in the syslog about it. the script is here http://pastebin.com/ce4GJJXC The permissions on it are -rwxr-xr-x. Can anyone help me figure out why it's not working? [20:15] Lownin: is that the cron file? [20:16] raub: no it's a script I put in cron.daily [20:18] Lownin: how about starting a bit more humble? [20:18] Create a cron.hourly file that does notthing but echo something to a /tmp/file [20:19] Lownin: I myself usually create something in cron.d [20:19] I can do that. [20:19] So I can set the time I want it to run [20:19] raub: this is a pretty venella install of ubuntu server 12.04 [20:20] I have not had much luck with cron.daily :) [20:20] so I just use cron.d [20:21] i.e. if i were in your shoes I would move your file to /usr/local/bin/take-a-dump [20:21] and then create /etc/cron.d/take-a-dump, which looks like this [20:22] 15 3 * * * root /usr/local/bin/take-a-dump [20:22] once that works, add the opton to send stder to syslog/whever [20:24] Lownin: that is IMHO. /etc/cron.d/take-a-dump does not even need to be executable [20:24] so i've installed both dhelp and dwww on my webserver but get the following *SIMILAR* errors in both (as well as my own webpages): The requested URL /cgi-bin/dwww/usr/share/doc/ was not found on this server. [20:25] rtdos: you know for sure you want to have that dir, right [20:25] raub: i do. [20:26] at least i think i do. [20:26] i do have it in /usr/share/doc [20:26] Kinda odd it is in cgi-bin [20:26] raub: i used the default settings for dhelp/dwww as well as for apache2 [20:27] it's almost as if apache2 is messed up. [20:27] rtdos: I've never used dhelp/dwww, but do know apache has config files in /etc/apache2/sites-enabled telling where things are supposed to be [20:28] raub: ok. hold on. i'll double check. [20:28] rtdos: the message seems to indicate it is trying to find a dir in the wrong location [20:29] raub: the only file there is 000-default.conf [20:29] is it pointing to the right locations? [20:30] raub: to /var/www correct? [20:30] which means i would have to link /usr/share/doc to /var/www then? [20:30] It depends on what rwww/dehlp wants. Docs better help [20:31] Maybe [20:31] ok. seeing if either have man pages. [20:31] rtdos: you now know what the problem is. Might be able to find in the installation docs for thos eprograms what it wants [20:33] raub: looking at dwww.conf now === Gurkenmaster is now known as zz_Gurkenmaster [20:47] Oct 8, 2013 15:03 from Capa [20:47] It's mostly the gas-wasting light trucks and SUVs that use the oil ... 30% goes [20:47] into truck/SUV tanks. [20:47] They average 19mpg according to http://en.wikipedia.org/wiki/Criticism_of_sport [20:47] _utility_vehicles#Fuel_economy [20:47] Cars are the next largest consumers of oil at 28%. [20:47] Then large trucks 20%. [20:47] Aircarft 9% [20:47] Boats 4% [20:47] Trains and buses 3% [20:47] Military 3% [20:47] [Planes Trains And Automobiles> msg #105690 (78 remaining)] Read cmd -> [20:47] Sorry for that [20:47] wrong screen tab [20:48] hahahaha... [20:52] raub: in the config file for dwww it looks like /cgi-bin/dwww/doc/HTML is commented out. i un-commented that line. how do i re-start apache2? [20:54] rtdos: sudo service apache2 restart [20:54] rtdos: or: sudo service apache2 reload to just reload configs [20:55] hitsujiTMO: thanks. 'restart' also reload configs as well, correct? [20:55] yes [20:55] it just doesn't bring the daemon to a halt [20:55] it just doesn't bring the daemon to a halt if you just reload [20:56] oh, ok. [20:56] this is the message i get now: AH00558: apache2: Could not reliably determine the server's fully qualified domain name, using 127.0.1.1. Set the 'ServerName' directive globally to suppress this message [20:56] i thought i had it set to my servers name. that is set in apache2.conf correct? [21:02] rtdos, you can set it anywhere as long as it's global ... [21:02] you can also ignore the error if you wish [21:03] hitsujiTMO: ok. doesn't look like it worked. neither dhelp nor dwww are loading anything from /usr/share/doc/HTML in the browser. [21:04] rtdos are there any errors in the global log in /var/log/apache2/error.log ? [21:04] though typing my host or localhost - http://localhost - works fine. [21:05] can you post your virtualhost config? [21:06] hisujiTMO: here is the output of /var/log/apache2/error.log - http://pastebin.ubuntu.com/6279342/ - where would i find the virtualhost config? [21:06] /etc/apache2/sites-available should be the non ssl one [21:07] hmm indeed no errors [21:08] where is the directives for /usr/share/doc/HTML declared? [21:10] hitsujiTMO: i'm not sure where the directies for /usr/share/doc/HTML are declared but here is the output for /etc/apache2/sites-available - http://pastebin.ubuntu.com/6279369/ [21:11] HI all, for ubuntu 12.04 update 3 I get the following error: lib32gcc1: Depends libc6-i386 (>= 2.11) but it is not going to be installed [21:14] rtdos whats the output of: ls -l /var/www [21:14] rtdos: http://manpages.ubuntu.com/manpages/jaunty/man8/dwww.8.html [21:16] i think that might be helpful [21:17] rtdos can you post: /etc/dwww/apache.conf [21:19] hitsujiTMO: kinda silly they are running a separate apache [21:20] raud: maybe not, could be a global directive [21:20] hitsujiTMO: you have a point there [21:52] hitsujiTMO: the output of ls -l /var/www is - total 8 [21:52] drwxr-xr-x 4 root root 4096 Oct 21 15:11 dwww [21:52] -rw-r--r-- 1 root root 177 Oct 20 22:31 index.html [21:52] sorry, i should've used pastebin for that. [21:53] two lines is usually fine.. [21:53] rtdos whats happening for http://localhost/dwww ? [21:54] hitsujiTMO: there is no /etc/dwww/apache.conf but when i type http://localhost/dwww it loads fine except when i try view a document in /usr/share/doc(/HTML) folders. [22:04] RoyK: hey, you use zfs right? have you written anything about your experiences using it? I've gone through a pile of the official docs and would like the perspective of a user now.. :) === gallomimia is now known as Gallomimia [22:45] hitsujiTMO: even running dpkg-reconfigure on dhelp and dwww didn't work. [22:54] hitsujiTMO: i think i know what the problem is. in /usr/lib/cgi-bin both dwww and dhelp are symbolic links to files and not directories. how can i fix this? [22:56] they should be symlinks to files. they are executables [22:59] hitsujiTMO: but why would i get this error: "The requested URL /cgi-bin/dwww/usr/share/doc/ was not found on this server." - wouldn't that indicate that a /usr/share/doc is not symbolic linked to /usr/lib/cgi-bin ? [22:59] this is the URL that is returned from dwww when i attempt to click on the 4th menu option at the top: http://localhost/cgi-bin/dwww/usr/share/doc/?type=dir [23:01] gdos: cgi-bin/dwww is the executive usr/share/doc/?type=dir would be params ... could be a rewrite issue. [23:02] dgos: ls -l /etc/apache2 [23:09] hitsujiTMO: http://pastebin.ubuntu.com/6279905/ [23:10] ls -l /etc/apache2/conf.d [23:10] ls -l /etc/apache2/conf-available [23:18] hitsujiTMO: here is the output of /etc/apache2/conf-available/serve-cgi-bin.conf file: http://pastebin.ubuntu.com/6279937/ [23:19] hitsujiTMO: here is the output of ls -l /etc/apache2/conf.d: http://pastebin.ubuntu.com/6279943/ [23:20] hitsujiTMO: here is the output of ls -l /etc/apache2/conf-available: http://pastebin.ubuntu.com/6279947/ [23:21] gdos can you post your apache2.conf from /etc/apache2