[03:33] <halcyonCorsair> i have an HP quad port server adapter that i'm using the intel e1000 driver for, but i want to try the HP driver, which only comes for RHEL3, 4, 5, or SLES9, or 10, would I be able to use any of those?
[03:34] <kgoetz> doubtfull
[03:34] <kgoetz> can you get source and build it yourself?
[03:34] <halcyonCorsair> i don't know....i don't think so, but i'm investigating that
[04:59] <fujin_> anyone here using Puppet?
[05:50] <nealmcb> fujin_: Not me.  But a more specific question might elicit a response.  Seriously good guide to asking questions on IRC: http://www.sabi.co.uk/Notes/linuxHelpAsk.html
[01:57] <ivoks> grrr
[01:57] <ivoks>  /etc/network/interfaces is bad concept :/
[01:58] <kgoetz> :(
[01:58] <ivoks> if you have 3-4 interfaces set up there, and if you missconfigure one, none of them will start
[01:58] <soren> ivoks: Yeah.. Who needs network configuratino?
[01:58] <kgoetz> ivoks: automatically
[01:59] <soren> ivoks: You'd have to *really* misconfigure it to make that happen, surely?
[01:59] <kgoetz> they still start if yyou ifup them dont they?
[01:59] <ivoks> soren something like iface static lo:100 (instead of iface inet static lo:100)
[01:59] <ivoks> that will crash whole network
[02:00] <ivoks> and servers are 2000-3000km away :/
[02:00] <soren> Yeah. Maybe we could provide a wrapper around it like with sudoers or passwd to avoid that sort of thing.
[02:00] <kgoetz> soren: what sort of wrapper?
[02:01] <ivoks> we should disect it to files
[02:01] <kgoetz> that knows the syntax of every config files in the archive/?
[02:01] <soren> kgoetz: You familiar with visudo ?
[02:01] <ivoks> include files from conf.d
[02:01] <kgoetz> soren: yes
[02:01] <soren> kgoetz: Nono, just interfaces.
[02:02] <kgoetz> hm.
[02:02] <soren> kgoetz: Just a wrapper script you call that opens an editor and checks that you haven't b0rken your /e/n/interfaces file.
[02:03] <kgoetz> mmm.
[02:08] <ivoks> we shall create blueprint for that
[02:08] <ivoks> lol, shall... should
[02:09] <soren> Ah...
[02:09] <soren> I'm not sure it's possible, actually.
[02:10] <ivoks> i think that separate files for interfaces would be easier
[02:11] <soren> Oh, please no.
[02:11] <ivoks> :)
[02:11] <kgoetz> how redhat
[02:11] <ivoks> i'm so frustrated right now, since i can't get to those servers...
[02:11] <ivoks> kgoetz: all distributions have single file for single interface
[02:12] <ivoks> iirc
[02:12] <soren> http://bugs.debian.org/159884
[02:13] <kgoetz> ivoks: not sure i follow
[02:13] <ivoks> you have one file with IP, interface etc...
[02:13] <ivoks> then you have another file with another IP, interface, etc..
[02:14] <soren> ivoks: I don't think the syntax check would be feasible. The way the parser figures out that a new interfaces is being described is by recognising the iface lines. If it doesn't match a set list of patterns, it's considered an option to the previous interface.
[02:14] <ivoks> then, if there is an error in second file, that error doesn't effect first file
[02:14] <ivoks> why not?
[02:15] <ivoks> for every file include file
[02:15] <ivoks> ifconfig $IFACE up $ADDRR netmask $NETMASK
[02:15] <ivoks> etc..
[02:17] <kgoetz> wouldnt it be better to make ifupdown support not bailing on errors?
[02:17] <kgoetz> or reverting to last good config?
[02:17] <soren> That's not the issue.
[02:17] <soren> The issue is that it's impossible to tell if it's been misconfigured or it's something else that's not working properly.
[02:18] <soren> The the syntax for interfaces is very forgiving, so it's practically impossible to detect errors in it (since it might just be someone doing something "interesting").
[02:19] <kgoetz> agree
[02:22] <ivoks> i would say that much bigger problem is that mistake on one interface destroys all other
[02:23] <soren> ivoks: That shouldn't be the case.
[02:23] <ivoks> but it is
[02:24] <ivoks> make a mistake in interfaces and try
[02:24] <kgoetz> ivoks: sure it is? or have you stufffed up worse then you think
[02:26] <ivoks> i'm sure
[02:26] <ivoks> that already happend to me once
[02:26] <kgoetz> i'd test it, but i dont want to loose my connectivity ;)
[02:31] <ivoks> just add false lines
[02:31] <ivoks> and try ifdown lo
[02:32] <ivoks> it won't work, even if you don't do mistake with lo
[02:33] <ivoks> $ sudo ifdown lo
[02:33] <ivoks> /etc/network/interfaces:23: too few parameters for iface line
[02:34] <ivoks> it works if you add:
[02:34] <ivoks> sdgfsagr rg rg tedfrgb
[02:34] <ivoks> but it doesn't if you add:
[02:34] <ivoks> iface ath6 static
[02:35] <ivoks> funny, isn't it? :)
[02:55] <soren> ivoks: Ah... *That* we could check.
[02:56] <soren> ivoks: But in the real world, a more sensible solution would probably be to make vi hilight it in bright red.
[02:56] <ivoks_> eh, wasn't here the whole time :D
[02:57] <ivoks_> we should check that
[02:57] <ivoks_> and bring interfaces which aren't cripeld
[03:03] <soren> omg... Have you ever actually looked at the ifupdown code?
[03:03] <soren> If you haven't.. don't!
[03:03] <kgoetz> lol.
[03:03] <kgoetz> i wont
[03:04] <soren> I wish I hadn't.
[03:08] <ivoks_> :)
[03:08] <ivoks_> i will :)
[03:08] <ivoks_> i don't want this to happen again :/
[03:08] <ivoks_> so lame :(
[04:19] <m11> hello all
[04:27] <DustWolf> where in
[04:27] <DustWolf> ubuntu settings
[04:27] <DustWolf> e.g. what file
[04:27] <DustWolf> do I find the setting that makes
[04:27] <DustWolf> ubuntu not start mysql with the --skip-networking flag?
[04:28] <soren> DustWolf: Which version of Ubuntu?
[04:28] <DustWolf> 6.06 LTS I think
[04:29] <soren> You think? :)
[04:29] <soren> Could you find out?
[04:29] <soren> "lsb_release -cs" will tell you
[04:29] <DustWolf> I'm quite sure
[04:29] <DustWolf> it says edgy
[04:30] <DustWolf> so.. ? :P
[04:30] <DustWolf> :)
[04:30] <soren> ...so it's probably Edgy (i.e. 6.10).
[04:30] <mralphabet> edgy is not 6.06 LTS ;)
[04:30] <DustWolf> sorry ;P
[04:31] <soren> I'll just fire up my edgy vm. Hang on.
[04:34] <soren> DustWolf: And you claim it's started with --skip-networking?
[04:34] <soren> DustWolf: As in it has "--skip-networking" on the command line?
[04:35] <mralphabet> I think he _wants_ the --skip-networking flag, IE to find the place to put it in
[04:35] <mralphabet> I could be wrong
[04:35] <DustWolf> yeah
[04:35] <DustWolf> well
[04:35] <DustWolf> I want to remove it
[04:35] <mralphabet> ahh
[04:35] <DustWolf> it got in durring a software upgrade I guess
[04:35] <DustWolf> mysql     6329  0.0  1.7 127904 18000 pts/0    Sl   16:17   0:00 /usr/sbin/mysqld --basedir=/usr --datadir=/var/lib/mysql --user=mysql --pid-file=/var/run/mysqld/mysqld.pid --skip-networking --skip-locking --port=3306 --socket=/var/run/mysqld/mysqld.sock
[04:35] <DustWolf> this is why
[04:36] <DustWolf> I assume it's there ;P
[04:36] <DustWolf> that's output from ps aux
[04:37] <mralphabet> DustWolf: do you need mysql networking?
[04:37] <DustWolf> yes
[04:37] <DustWolf> I can configure it securely.. trust me :P
[04:37] <DustWolf> but my php aren't very happy without being able to connect
[04:38] <mralphabet> php on another machine?
[04:38] <DustWolf> (and I just want to fix this... because it's downtime until I do)
[04:38] <DustWolf> php on the same machine
[04:38] <mralphabet> umm, okay
[04:38] <DustWolf> and the mysql backuping apps on other machines :P
[04:39] <mralphabet> ahh
[04:39] <soren> DustWolf: What does 'dpkg -l mysql-server-5.0' say?
[04:39] <DustWolf> Desired=Unknown/Install/Remove/Purge/Hold
[04:39] <DustWolf> | Status=Not/Installed/Config-files/Unpacked/Failed-config/Half-installed
[04:39] <DustWolf> |/ Err?=(none)/Hold/Reinst-required/X=both-problems (Status,Err: uppercase=bad)
[04:39] <DustWolf> ||/ Name                                   Version                                Description
[04:39] <DustWolf> +++-[04:39] <DustWolf> ii  mysql-server-5.0                       5.0.24a-9ubuntu2                       mysql database server binaries
[04:41] <soren> Does 'grep -l -- --skip-networking /etc/init.d/mysql /etc/default/mysql* /usr/bin/mysqld_safe' give any result?
[04:41] <DustWolf> # grep -l -- --skip-networking /etc/init.d/mysql /etc/default/mysql* /usr/bin/mysqld_safe
[04:41] <DustWolf> grep: /etc/default/mysql*: No such file or directory
[04:41] <DustWolf> erm
[04:41] <soren> Apart from that?
[04:41] <DustWolf> /usr/bin/mysqld_safe
[04:41] <DustWolf> this is missing ;P
[04:41] <DustWolf> irc stripped it
[04:42] <soren> Missing? The file is not there?
[04:42] <DustWolf> no
[04:42] <DustWolf> it's there
[04:42] <DustWolf> just got removed when I pasted ;P
[04:42] <DustWolf> or rather.. parsed
[04:42] <DustWolf> e.g. I got this:
[04:42] <DustWolf> # grep -l -- --skip-networking /etc/init.d/mysql /etc/default/mysql* /usr/bin/mysqld_safe
[04:42] <DustWolf> grep: /etc/default/mysql*: No such file or directory
[04:42] <DustWolf> /usr/bin/mysqld_safe
[04:43] <soren> In that case, you've edited that file. It's not a conffile and the stock one does not contain '--skip-networking'
[04:44] <DustWolf> ok
[04:44] <DustWolf> I'll fix it then
[04:44] <DustWolf> and it might have been my co-admin playing smart again
[04:45] <soren> You can find the original one at http://warma.dk/mysqld_safe
[04:46] <DustWolf> nah I'll just let it be
[04:48] <soren> Alright. I'll remove it from there again, then.
[04:49] <DustWolf> :)
[04:49] <DustWolf> thanks for the help :)
[05:01] <mabhobs>  I have a problem with my cms. The following code is not parsed: http://test.mabhobs.de/dynimages/100/files/_galleries/gallery/design.jpg . My provider found this out http://pastebin.ca/690295 . The php code is supposed to create a resized image of the original image, this also why there are no thumbnails ( http://test.mabhobs.de ). Does anybody have an idea why this is the case? (I think it is a FastCGI problem.)
[09:08] <roadkill> man why dose digg.com make my system jump to max cpu usage
[09:09] <roadkill> constant,
[09:11] <mralphabet> all the ajax?
[09:19] <roadkill> i dunno it work fine this morning i havnt installed anything since
[09:38] <roadkill_> anyone here know electronics stuff, i have a part i've got a question about
[10:12] <mathiaz> lamont: about bug 135393, what is the security view of it ?
[10:12] <ubotu> Launchpad bug 135393 in bind9 "init.d-script uses control channel instead of signals" [Wishlist,Triaged]  https://launchpad.net/bugs/135393
[10:13] <mathiaz> lamont: is the control interface opened to anyone by default ?
[10:13] <lamont> mathiaz: root and/or the user bind, depending
[10:14] <mathiaz> lamont: the reporter said that most of the user disable rndc by default.
[10:14] <lamont> yeah, well.  if you choose to modify your config away from doing it the right way, then you can modify init.d too.
[10:15] <mathiaz> lamont: I was wondering if the control channel would only listen to localhost
[10:15] <lamont> and I doubt that 'most' is anywhere near accurate
[10:15] <mathiaz> lamont: agreed.
[10:17] <lamont> mathiaz: well...
[10:18] <lamont> anyone on localhost can talk to the control port.  and if you have read access to /etc/bind/rndc.key (or can otherwise get the key), then you can tell it to do stuff
[10:18] <lamont> default is to listen on lo only
[10:19] <mathiaz> lamont: ok I see.. Thanks.
[10:20] <lamont> (and yes, I had to go looking...)
[10:21] <lamont> bug 82178: I 'm inclined to just not deliver *idn*
[10:21] <ubotu> Launchpad bug 82178 in bind9 "idnconv manual page exists, but no binary" [Low,Confirmed]  https://launchpad.net/bugs/82178
[10:21] <lamont> manpages that is, we already don't deliver the binary