[00:52] <GTAXL> How can I change the default umask for a directory?
[01:01] <markthomas> I don't believe you can set umask on a per-directory basis.
[01:02] <GTAXL> markthomas, I have a web dir that SPGrab uploads snapshots to, by default it doesn't allow others to read, resulting in forbidden when someone tries to access my screenshots, so I have to go in and change the permissions on the newly uploaded pics
[01:04] <markthomas> Not familiar with spbrab.  But umask is per-process, not per dir.  You probably need to chmod in the process or some such.
[01:05] <GTAXL> maybe setfacl?
[01:05] <markthomas> I've never worked much with ACLs.
[01:05] <markthomas> Yeah.  That I couldn't tell you.
[01:05] <GTAXL> SPGrab works fine on my CentOS machine, really wanting it on my Ubuntu one
[01:05] <GTAXL> I tried recursive permissions, don't work
[01:06] <markthomas> What file ownership are the files getting when they're uploaded?
[01:07] <GTAXL> -rw-rw----
[01:07] <markthomas> And the ownership?
[01:07] <markthomas> What are the owner and group?
[01:07] <GTAXL> the user I upload via FTP as
[01:08] <GTAXL> SPGrab is used to take snapshots of your desktop, upload via ftp to a web dir, and share a link
[01:08] <markthomas> Hmm...
[01:09] <GTAXL> I just want others to have read access on new created files in /var/www/screenshots/
[01:09] <GTAXL> :P
[01:09] <markthomas> So, you need to set the umask not for a specific directory, but as part of the FTP process.  Is this process the only thing using the ftpd on that box?
[01:09] <GTAXL> no
[01:09] <markthomas> Hmm...
[01:10] <markthomas> Do they need access immediately after upload?
[01:10] <markthomas> i.e. can there be a couple minute delay?
[01:11] <GTAXL> Sorry about that, my ssh tunnel died. :S
[01:12] <markthomas> :)
[01:12] <markthomas> I'm at the end of my workday, and I'll need to run.  If a couple minute delay after upload is acceptable, try throwing a chmod into your cron.
[01:13] <markthomas> s/cron/crontab/
[01:13] <markthomas> It's a workaround, but it may do while you look into using ACLs.
[01:13] <GTAXL> hmm
[01:14] <markthomas> No, it's not attractive.
[01:14] <GTAXL> I've heard of sticky permissions, not sure if that's it
[01:15] <markthomas> It's not.
[05:44] <digitalcha0s> trying to remove the default login on ubuntu server 12.10 -> any hints for me?
[05:45] <digitalcha0s> i know issue and issue.net and the other one
[05:46] <digitalcha0s> motd
[05:47] <digitalcha0s> seems like i change it and then resets?
[05:57] <digitalcha0s> gay
[06:08] <digitalcha0s> i figured it out
[06:43] <glitchy351> Hello! I am new to This IRC channel. I was wondering if someone knows much about load balancing webservers in ubuntu 12.10 ?
[10:12] <aldwinaldwin> Question:  i've installed exim4 with a relay,  and have set the MAILTO= in crontab.  Via 'dpkg-reconfigure exim4-config' i was able to masquerade the domain. The only thing I can't figure out, is to change  the  username 'root' in the From-Address to another username.   So, how to change  root(at)domainname.com  to   user(at)domainname.com ?
[11:13] <RoyK> hm... scripting in bash, I want to run this nagios check seven times and then return the largest exit code of all those tests combined (0-3). any idea of how to do that most easily?
[11:20] <Tzunamii> RoyK: Start Nagios from a script and put each exit status into an array. That's the start
[11:22] <Tzunamii> RoyK: After that you can sort the array and display the X number of top results
[11:38] <andol> RoyK: Alternative approachy, which may or may not apply, take a look at the check_multi plugin?
[12:39] <billy_ran_away> Hi, I've got a 5 1TB disk RAID 5 array and I'd like to migrate to 3 3TB disk array… what's the best way to do this?
[12:39] <billy_ran_away> I've got enough SATA slots to where I can run the old array in degraded state and run the new array in a degraded state as well.
[12:40] <billy_ran_away> Is this the recommended procedure? Run both in degraded, copy all the data over, then, remove the old disks and reboot with all the three new disks in?
[12:46] <davegarath> billy_ran_away: if you place a 3T disk into a raid with 1T disk and rebiuld it you will lost 2T because raid will cuts it on 1T
[12:46] <mardraum> billy_ran_away: I probably would not do it like that, I would use two machines in your situation, or another controller added in
[12:47] <billy_ran_away> Can't I run both in degraded state?
[12:47] <mardraum> davegarath: I don't think he wants to do that, he's just running out of sata ports?
[12:47] <mardraum> billy_ran_away: you probably can. like I said, I wouldn't do it like that. what happens if you have disk failure in that state?
[12:48] <mardraum> depends how important your data is, I assume your backups are all current :D
[12:48] <billy_ran_away> Then I insert my 5th drive back in and order another 1TB drive
[12:48] <mardraum> I don't think you are listening to me. good luck.
[12:49] <Ralf22> Hey. When I install 13.04 beta now, will I be up to date when the final version is released?
[12:49] <Nafallo> yes
[12:50] <Ralf22> thanks Nafallo
[12:50] <Nafallo> yw
[12:55] <billy_ran_away> Why is mdadm -C /dev/md0 -l 5 -n 3 /dev/sd[hg]1 missing not making a 6TB array when I'm using 3TB disks?
[13:05] <billy_ran_away> Why is neither cfdisk nor fdisk making the right size partitions on this 3Tb drive?
[13:06] <mardraum> maybe your hardware can't handle the sector size change?
[14:07] <billy_ran_away> mdadm -C /dev/md0 -l 5 -N media -R -n 3 /dev/sdd1 /dev/sdc1 missing
[14:07] <billy_ran_away> mdadm: cannot open /dev/sdd1: Device or resource busy
[14:07] <billy_ran_away> WTF? I just rebooted...
[14:10] <zastaph> if I upload my files by ftp how do I make sure they get www-data:www-data 775 like my other files in /home/zastaph/web ?
[14:10] <zastaph> setgid and umask? and what user should I use for ftp
[16:41] <roaksoax> kirkland: howdy!! did you upload testdrive 3.18 to the ubuntu archive or only ppa?
[16:51] <hallyn> ppetraki: hey - what would you say would be the most appropriate way to determine (from C) whether a blockdev is an lvm2 LV?  The functions in /usr/include/lvm2app.h ?
[16:51] <ppetraki> hallyn, /me looking
[16:56] <hallyn> thx!
[16:57] <ppetraki> hallyn, just posting publicly now :)
 the chessy way out to test for the negative, INQUIRY
[16:57] <ppetraki>  block devices don't respond to SCSI cmds
[16:57] <ppetraki>  all this code assumes you're trying to manage LVM
[16:57] <ppetraki>  so you pick a VG, and use that handle to get a list of LVs
[16:57] <ppetraki>  which is a royal pain in the ass
[16:57] <ppetraki>  shortcut, just interrogate device mapper
[16:57] <hallyn> ppetraki: what i'm doing is a generic blockdev layer in lxc,
[16:57] <hallyn> which should take the 'source path', the blockdev in lvm case, and decide 'yes, i can handle this'
[16:58] <hallyn> (so then it knows how to do snapshots)
[16:58] <ppetraki> yeah, anything with a linear driver in DM table is either an LV or MP path
[16:58] <ppetraki> well, an LV
[16:59] <hallyn> ppetraki: in the current script-based clone, i just do lvdisplay $dev
[17:00] <ppetraki> hallyn, you could also use the sg api and just pound all the devices with inquiry, not difficult, http://pastebin.ubuntu.com/5713668/
[17:00] <hallyn> suppose i could go see how that command works
[17:00] <ppetraki> hallyn, it tried to fail back, SCSI and ATA passthrough, both failed.
[17:01] <ppetraki> hallyn, as if you concat a bunch of devices, which ones ID do you return?
[17:01] <hallyn> you're talking mp?
[17:01] <ppetraki> hallyn, that's LVM
[17:02] <hallyn> concat devices?
[17:02] <ppetraki> hallyn, likely true of MP but I don't have system access to confirm
[17:03] <ppetraki> hallyn, sysfs is also useful, as block devices don't have make and model attrs, nor are they attached to the sd driver in sysfs device model
[17:03] <hallyn> ok, so your comment was about sg...  i demand the actual lvm device, not one of the backing physical devices, so hopefu
[17:03] <hallyn> oh yeah, lemme check sys
[17:05] <ppetraki> hallyn, http://pastebin.ubuntu.com/5713690/
[17:06] <hallyn> so i see:
[17:06] <hallyn> cat /sys/dev/block/252:9/dm/uuid
[17:06] <hallyn> LVM-B78EDPMWoBOMaP0szouAOQ6QDLsHCszyqKBLlxvtB41UudUh3CJyIFO0cDM35Df6
[17:06] <hallyn> I wonder if LVM- is predictable?
[17:06] <ppetraki> hallyn, you can do set exclusion at that point
[17:07] <hallyn> dm is not necessarily lvm right
[17:07] <ppetraki> hallyn, actually, yes :)
[17:07] <ppetraki> hallyn, http://pastebin.ubuntu.com/5713696/
[17:07] <ppetraki> hallyn, fixed?
[17:08] <hallyn> ppetraki: sweet, thanks :)
[17:08] <ppetraki> hallyn, UUID_PREFIX is hardcoded, so I think you're good, unless some uses udev to rename it
[17:08] <ppetraki> hallyn, at a minimum you've covered the common case
[17:08] <hallyn> if they want to mess with udev like that, they can get what's coming :)
[17:08] <ppetraki> hallyn, agreed :)
[17:09] <hallyn> ppetraki: thanks!   ttyl
[17:09] <ppetraki> hallyn, anytime!
[17:50] <ddsss> is keyring and seahorse installed by default in ubuntu?
[18:48] <Jaith> I just did apt-get update/upgrade on my server running Ubuntu 12.04 LTS and noticed that PHP is ancient (5.3.10).  The latest version of php is 5.4.14.  The latest release of php 5.3 is 5.3.24.  Why on earth has canonical (or someone) not provided more up-to-date packages for PHP?  This is kind of ridiculous.
[18:55] <sarnold> Jaith: see https://wiki.ubuntu.com/SecurityTeam/FAQ#Versions and http://www.debian.org/security/faq#oldversion
[18:56] <sarnold> Jaith: see also http://people.canonical.com/~ubuntu-security/cve/pkg/php5.html
[18:56] <RoyK> sarnold++
[18:57] <sarnold> hey RoyK :)
[18:58] <RoyK> sarnold: ;)
[19:04] <Jaith> sarnold: thx
[19:04] <Jaith> oops..
[19:38] <disturbed> hello
[19:39] <disturbed> i'm having issues with automatically start tightvncserver at boot, any volunteers to help out?
[19:39] <disturbed> not... all... at... once... plz :)
[19:40] <bean> disturbed: chillax.
[19:40] <bean> disturbed: how did you install it
[19:40] <disturbed> sudo apt-get install tightvncserver
[19:40] <bean> okay, and how do you currently start the server
[19:41] <disturbed> tightvncserver -geometry 1980x1020 :0
[19:41] <disturbed> that works...
[19:41] <disturbed> then i made a /etc/init.d/tightvncserver, chmod +x
[19:41] <bean> okay
[19:41] <bean> and that script works to start it?
[19:42] <bean> this post seems to explain it pretty well
[19:42] <bean> http://askubuntu.com/questions/120973/how-do-i-start-vnc-server-on-boot
[19:42] <disturbed> if i execute: sudo /etc/init.d/tightvncserver it works
[19:42] <disturbed> but it doesn't do that at boot
[19:42] <bean> right
[19:43] <bean> you have to update it to run at boot
[19:43] <bean> update-rc.d tightvncserver defaults
[19:43] <bean> should do it
[19:43] <disturbed> can i do: update-rc.d tightvncserver -geometry 1980x1020 :0 ?
[19:43] <bean> no
[19:44] <bean> literally
[19:44] <bean> update-rc.d tightvncserver defaults
[19:44] <bean> :|
[19:45] <disturbed> ok brb
[19:45] <disturbed> right, now i also see it in chkconfig, so looks good, tnx, => rebooting
[19:49] <disturbed> still not rebooted...
[19:49] <bean> heh
[19:49] <disturbed> btw i saw that link before and already tried rc.local and @reboot crontab thingy
[19:50] <bean> did you try the accepted answer
[19:50] <bean> lol
[19:51] <disturbed> @reboot seemed the easiest one :)
[19:51] <disturbed> right, rebooting :)
[19:51] <bean> @reboot likely won't work like you think it does, but alright :)
[19:53] <disturbed_> no luck: ps -ef|grep vnc => nothing
[19:54] <bean> ok
[19:54] <bean> a vnc server is hard to get to work at boot, since it requires the display and stuff to already be working
[19:55] <bean> not really a server if it has a GUI imo
[19:55] <bean> ...
[19:55] <disturbed_> :)
[19:55] <disturbed_> well, i would have been happy with mint linux
[19:55] <disturbed_> except that it could not boot without a monitor connected to it
[19:55] <disturbed_> not a nas when it needs a monitor constantly
[19:56] <RoyK> perhaps #ltsp
[19:58] <disturbed_> i'll try webmin and see how far i get :)
[19:58] <RoyK> !webmin
[19:58] <disturbed_> *$!# :)
[19:59] <RoyK> disturbed_: better use the commandline - it doesn't take to long to learn the basics
[19:59] <RoyK> and it helps a good while in the long term
[20:00] <disturbed_> where's my xanax
[20:00] <RoyK> (by actually understanding what you're doing instead of just clicking some fancy GUI)
[20:00] <RoyK> benzodiazepines aren't good for you, disturbed_ - beleive me
[20:01] <disturbed_> should be a nas, so samba, daatp, owncloud, plexserver, ...
[20:01] <RoyK> well, buy one
[20:01] <RoyK> get a nas box
[20:01] <RoyK> doesn't cost that much
[20:01] <disturbed_> see, i'm not doing basics :)
[20:02] <RoyK> disturbed_: it's rather easy to learn ;)
[20:02] <disturbed_> i'd like to have a small server that can do samba, bittorrent, itunes server, ...
[20:02] <RoyK> yeah, it's easy
[20:02] <RoyK> easy peasy
[20:02] <RoyK> just learn the basics
[20:02] <disturbed_> but, since it's in the fitness room, i also sometimes want to connect a monitor and play music on that
[20:02] <RoyK> don't be so bloody lazy
[20:03] <disturbed_> without having to use my playstation 3 for the dlna
[20:03] <disturbed_> don't see how i'm gonna play my music on my nas, in text mode :)
[20:03] <RoyK> over the net - sure
[20:04] <RoyK> but then - use a laptop or something if you don't want to learn too much
[20:04] <RoyK> setup a homeserver and samba and you're all set
[20:04] <disturbed_> think you're missing the point
[20:05] <RoyK> probably more hassle with a webbased admin interface than with something on the commandline
[20:05] <disturbed_> i want it to server a nas and media server
[20:06] <RoyK> ps3mediaserver?
[20:06] <RoyK> xbmc?
[20:06] <RoyK> xbmc works well
[20:07] <disturbed_> openmediavault, nas4free, ...
[20:07] <RoyK> well, learn it, please
[20:07] <RoyK> don't just ask
[20:07] <RoyK> or join the channels where those products are supported
[20:08] <RoyK> ubuntu server can give you storage on raid and support for a bunch of applications
[20:08] <RoyK> but this channel isn't for end users for media server setup
[20:09] <disturbed_> nvm
[20:09] <RoyK> heh
[20:09] <RoyK> (idiots)
[20:22] <disturbed> single, doesn't surprise me