[00:32] <rsully> Any reason `hdparm` shows my drives as only 2TB instead of 3 and 4?
[00:32] <rsully> er. sorry not hdparm, lshw
=== Lcawte is now known as Lcawte|Away
=== markthomas is now known as markthomas|away
[08:36] <apeskalle> Hello. I have a problem with my NFS server, that I hope you can help me with. I've added a new raidstripe to the nfs server, tar'ed all the data over to it. Deleted the old symlinks in /export, and linked them to the new stipe. But now when I try to mount the exports on the nfsclients, it times out. When I check tcpdump on the nfsserver, it complains about stale filehandles. I've tried rebooting, deleting rmtab, xtab, sm, s
[08:37] <apeskalle> uninstalling nfs server and client with purge and autoremove. Then ran out of ideas
=== Lcawte|Away is now known as Lcawte
[09:56] <lordievader> Good morning.
[09:57] <hadifarnoud> on nginx, what should I set for client_body_buffer_size? the guides I found are confusing. Should it be a large number and then set body size also on a large number?
[11:29] <blanoz> What would you consider to be more appropiate for PHP shared hosting between 3 friends (each with its own /home and www folder under /home): apache & mod_php + nginx for static content OR apache mpm event (static files only) + php-fpm, each user with its own fpm pool?
=== eaud is now known as ubuthetechguru
[16:11] <kevinde> Are there any good paid ebooks around to learn more about linux server security?
[16:13] <ivoks> i learned a lot from 'hacking linux'
[16:13] <ivoks> hacking exposed linux
[16:13] <ivoks> http://www.amazon.com/Hacking-Exposed-Linux-3rd-Edition/dp/0072262575
[16:14] <ivoks> it's a bit outdated thou
[16:14] <ivoks> still, a good read
[16:14] <ivoks> i had this one - http://www.amazon.com/Linux-Hacking-Exposed-Brian-Hatch/dp/0072127732/ref=pd_sim_b_5
[16:26] <kevinde> thanks
[16:26] <kevinde> will check that out
[17:55] <apeskalle> Only took most of the day, but I've figured out the NFS problem (posted earlier today)
[17:56] <bekks> And what was the problem? :)
[18:10] <apeskalle> nfs has 32bit inodes, and i used xfs on the new stipe, which has 64bit inodes (old stipe had ext3). So I added fsid to /etc/exports , and that solved the problem
[18:13] <bekks> So actually you are using NFSv4 now, and I guess you actually had this issue, which is not a "nfs inode size" issue: http://www.tcm.phy.cam.ac.uk/sw/inodes64.html - correct?
[18:15] <apeskalle> Yes. nfsv4
[18:29] <apeskalle> Never got so far to try any programs or ls like in the link. I couldnt even mount the filesystem on the clients. nfsv3 and 4 just timed out. Tcpdump and -v showed stale nfs handle errors
[18:50] <bekks> apeskalle: Which sounds like nfs misconfiguration, not like an inode issue.
[21:14] <ilken> do i need network manager on ubuntu server? its using 2% of my ram
[21:14] <ilken> can i remove it?
[21:14] <bekks> If you configure networking without it, yes.
[21:18] <pmatulis> ilken: how did it get installed on your server?
[21:25] <ilken> its an old box, it had KDE on it before, but i needed a stick of RAM so i stole one out
[21:25] <ilken> i removed all the WM's
[21:26] <soahccc> Why does a default crtmpserver installation (via apt-get) does not require any authentication at all? I read a lot and the only thing I could find is a bug (which broke authentication) but it's from 2011 and >300 revisions in the past. Would such a bug be in an official package still without anyone noticing? I know that crtmpserver is very complicated and the documentation / howto's / etc. are rare and often outdated
[21:26] <ilken> its been messed with for over a year i may need to re-install a minimal iso
[21:27] <pmatulis> ilken: that's what i would do, reinstall 14.04 LTS
[21:27] <pmatulis> (server edition)
=== bilde2910|away is now known as bilde2910