[00:24] <adam_g> jamespage, http://people.canonical.com/~agandelman/ca/folsom/2012.2.3/  CA rebuilds of the SRUs that were released today
[01:31] <LargePrime> Is a Swap File just as fast as a swap partition?  Ubuntu 12.04 server.
[01:33] <sarnold> LargePrime: a swap partition allows the kernel to direct the hard drive to seek directly to a location to read or write; a swap file means the kernel must look up off disk the location of blocks for the file, and then seek to those blocks. there may be three or four block lookups before the kernel can retrieve the specific address to seek to.
[01:33] <sarnold> LargePrime: so, in theory, a swap partition should be a touch faster than a swap file. however, by the time you're going to disk to get data, you're already going significantly slower than if the data were in memory in the first place.
[01:34] <sarnold> LargePrime: so if you have a chance to make a swap partition ahead of time, that makes sense; but if you need a swap file instead, there's nothing horribly wrong with that..
[01:35] <LargePrime> so the advantage is negligible when compared to slow disks.
[01:35] <LargePrime> The issue i am facing is the tool the host gave me does not allow raid 0 on swaps
[01:35] <sarnold> yes. I expect the kernel can cache those indirection blocks in memory, so if it is -always- using the file, it might not even be a real penalty..
[01:35] <LargePrime> so a swap file could be on a raid 0 partition
[01:36] <sarnold> could you just create a swap partition on each of the drives and manually add them one at a time?
[01:36] <sarnold> granted you wouldn't get the striping speedup that way...
[01:37] <LargePrime> I guess.  Can i change that?
[01:37] <sarnold> LargePrime: the swapfile would let you do that.
[01:37] <LargePrime> can i chage my swap partitions from RAID 1 to RAID 0
[01:37] <LargePrime> manually?
[01:38] <sarnold> sorry, that's out of my experience
[01:38] <sarnold> my guess is you could remove them, create new raid 0, and then mkswap on that, but .. I've never tried.
[01:39] <sarnold> LargePrime: hah, I was wrong: https://raid.wiki.kernel.org/index.php/Why_RAID%3F#Swapping_on_RAID
[01:39] <sarnold> ""The kernel itself can stripe swapping on several devices, if you just give them the same priority in the /etc/fstab file. "
[01:46] <LargePrime> thanks you sarnold
[03:17] <adam_g> zul, can you comment on https://launchpad.net/ubuntu/+source/cinder/+bug/1150720 please? can't verify the SRU with the test case you proovided
[04:14] <bitblt> anyone here running mysql on server>12.04 have a moment to check something for me?
[04:17] <bitblt> any idea why if, on an already running service, you issue "service X start" they return 0, except for mysql, which returns 1?
[05:39] <blackjack> how to fix Invalid Partition Table
[06:17] <Freze> hi all
[06:51] <ndee> hi there, I try a publickey login on a remote server, with ssh -vvv remote, I get following message: debug3: Not a RSA1 key file /var/www/public_www/.ssh/id_dsa. I know that it's not a RSA file but a DSA file, shouldn't that also work as publickey auth?
[06:52] <ndee> ah, a little later, I see: debug1: identity file /var/www/public_www/.ssh/id_dsa type 2 so that shouldn't be a problem
[06:53] <ndee> the problem is, I can't login with "ssh -2 -vvv remotehost" although the public_key is added to the authorized_keys file, that file has 600 permission and the .ssh directory also has 600.
[06:54] <ndee> I am an idiot
[07:21] <sw> Hi, can someone explain what the --delete-after option does in rsync as I don't quite understand the man. Does it mean that files/folders that are on the destination but not source any longer are removed?
[07:22] <Shogoot> Hi people!. The issue: I got a server that i installed vsftpd on, made a user and connected to the server. When the server connects it connects to root... and i have been trying to make /var/www/html as default ftp root and that the user has write/read to that folder as this person is going to upload/download its own html pages and resources...  Another strange  thing is that when i change directory i can change to any directory that is not /var/www
[07:22] <Shogoot> /html (???) - Anyone that can help me find the solution to this?
[07:23] <spidernik84> sw, from my understanding rsync cleans the files on the destination that are missing in the source (real sync)
[07:24] <spidernik84> the --delete-after tells rsync to remove them at the end of the entire synch operation, not before
[07:24] <spidernik84> again, from my understanding
[07:24] <sw> spidernik84, So ^ then? So confusing.
[07:25] <spidernik84> sw have you checked this? http://superuser.com/questions/156664/rsync-delete-options
[07:25] <spidernik84> might shed some light
[07:26] <spidernik84> still, I'm a basic rsync user. It has so many options :)
[07:26] <sw> spidernik84, Ah great thanks, I'd only looked at the man.
[07:29] <spidernik84> np :)
[07:34] <Shogoot> Hi people!. The issue: I got a server that i installed vsftpd on, made a user and connected to the server. When the server connects it connects to root... and i have been trying to make /var/www/html as default ftp root and that the user has write/read to that folder as this person is going to upload/download its own html pages and resources...  Another strange  thing is that when i change directory i can change to any directory that is not /var/www
[07:34] <Shogoot> /html (???) - Anyone that can help me find the solution to this?
[07:54] <stoogle> does anyone know anything about introducing new additional hdd's into a external raidarray on ubuntu?
[07:54] <AtuM> stoogle, that's something for the external raidarray admin to worry about.
[07:55] <AtuM> stoogle, however if you use DAS/SAN you can hot-add volumes/disks on the fly.
[07:56] <stoogle> welllll i have one at my house, i have 4 x 1tb drives mounted. they are currently running as file storage. i have just slotted in another 8 or so 1tb drives
[07:56] <AtuM> so are you experiencing problems?
[07:57] <stoogle> It was mounted in console along time ago with the 4x hdd's. the problem is now it still only shows 4tb even with the new drives. this would be due to the original mount size im guessing? i installed dell openmanage which showed me this status on the HDD's: http://www.freweb.com.au/whatitscurrentlyshowing.txt
[07:58] <AtuM> so where's the problem?
[07:58] <stoogle> the problem is in ubuntu its still showing 4tb not 14 x 4tb or what ever it is.
[07:58] <stoogle> 14 x 1tb rather
[07:59] <AtuM> that's not the problem with ubuntu/linux
[07:59] <AtuM> you should learn a bit more abour raid arrays and management of PERC adapters
[07:59] <stoogle> :(   im trying. im a bit confused.
[07:59] <AtuM> you need to expand the raid array it you need more space within the current volume
[08:00] <stoogle> so how would i include the Foreign disks into the mount and make them online?
[08:00] <AtuM> that's what you can do within the Perc bios util - there might be utilities to do it online.. i'm guessing this is dell.. so ask dell for software
[08:00] <AtuM> forget about mount!
[08:00] <stoogle> lol sorry.
[08:01] <AtuM> linux won't make them online.. it shouldn't.. you must configure those disks within the raid adapter itself
[08:01] <jamespage> adam_g, they all lgtm
[08:01] <AtuM> linux has no role there.. it just gets what that adapter offers
[08:01] <stoogle> ah ok, with you now. linux only reads what the adapter says. so its all hardware side really?
[08:02] <AtuM> that's correct
[08:03] <AtuM> raid array is a layer in-between the hardware and the OS.
[08:04] <stoogle> so frustrating. to my knowledge i should be able to go into dell openmanager web based. which i can. if i click on virtual disks it shows the original 4 x drives. i seleceted reconfigure, now this option is meant to allow me to add additional new hdd's and it doesnt show them. If i get out of that and click on the raidarray info it shows all 14drives with the info that was in that text
[08:04] <stoogle> file. very frustrating.
[08:05] <AtuM> most adapters can be managed online via a special tool from the vendor.. so you could expand your current volume using that from within your ubuntu.. but linux itself has no say in it.. there are many things you can do wrong trying to extend your array if you've never done it before
[08:06] <stoogle> as above, i did get the software from vendor. i must have done something incorrectly.....  as its different from what everyone else is doing which i listed above.
[08:06] <AtuM> ok.. so that's the question for openmanager developers.. most tools do allow hot-adding disks and expanding volumes online..  it is after doing that that you should check with us on how to get that "mount" bigger :)
[08:07] <stoogle> lol ok. thank you for your time and patients AtuM
[08:08] <AtuM> as you will still have a partition of the same size aswell as the filesystem.. so there's work to do after expanding the volume
[08:09] <AtuM> stoogle, sometimes reboot helps.. might even have more luck bringing up the adapter's bios util to expand the volume..
[08:10] <AtuM> stoogle, hope I helped make things more clear.. :)
[08:14] <sw> So if someone is logged in via password they appear in users but when using SSH keys they don't?
[08:15] <sw> i.e. if I login on two sessions with the password users shows root twice, but not when one of those is via SSH keys.
[08:22] <ruben23> h guys i  have mysqlk server but does not run, i wa able to restart but when login it say socket /var/run/mysqld. scok 2
[08:29] <kai> hi folks.
[08:30] <kai> before I go and download the 13.04 server installer, does that have a sane interface to set up LUKS-based encrypted LVM of parts of the disk while still allowing me to "partition" the encrypted LVM?
[08:30] <kai> the graphical installer fails miserably at that
[08:32] <ruben23> guys this is my error -----> http://pastebin.com/BaTyqZWR
[08:40] <sw> So if someone is logged in via password they appear in users but when using SSH keys they don't?
[08:40] <sw> i.e. if I login on two sessions with the password users shows root twice, but not when one of those is via SSH keys.
[09:02] <xnox> kai: yes it does. it has a one click option for that, which will then ask for a password as well.
[09:02] <xnox> kai: mind you desktop installer has that as well, but server cd is more flexible as it allows manual partitioning for lvm & crypt as well.
[09:03] <kai> xnox: the latter is what I'm looking for. the 1-click option is fine, but I don't want to take the performance hit of encryption for my data partition where I keep open source code and publicly available data
[09:05] <kai> but I clearly want / and swap encrypted. :)
[09:11] <kai> ok, now I just need to figure out how to connect to the wireless network via the command line and set up the radius magic
[10:55] <LargePrime> When I run smartctl as root i get permission denied.  Can't find any info.  Thoughts?
[12:38] <smoser> zul, could you loook at comments in https://bugs.launchpad.net/bugs/1098688
[12:38] <smoser> (last ones rguarding cloud archive)
[12:39] <zul> smoser:  should be fixed in grizzly
[12:40] <zul> smoser: havent looked at folsom yet
[14:12] <stoogle> asking for some suggestions, i have a md1000 raid array running raid 5 with 13 x 1 tb hdds.. i want to take a external backup of everything to take home once a week for safe keeping incase of fire/flood or raid failure. what is the best way to do this? i can get another md1000 with 13 x 1tb hdds in it also, put them into a padded brief case and take them home once a week. insert another
[14:12] <stoogle> 13 x 1 tb hdd in there and then copy everything over again until i swap again? i just need it to be reliable!
[14:13] <sw> Is it possible with rsync to only copy files that are new or have been modified, or is that behaviour standard?
[14:14] <xnox> sw: look at rsnapshot.
[14:14] <sw> At the moment using: rsync -avz --delete-after.
[14:14] <sw> xnox, Is there not an option in rsync?
[14:15] <xnox> sw: rsnapshot is what you want.
[14:15] <xnox> it's a script around rsync more or less.
[14:15] <sw> xnox, So are you saying that rsync is not capable of this? Or just that we should use rsnapshot? :b
[14:16] <stoogle> asking for some suggestions, i have a md1000 raid array running raid 5 with 13 x 1 tb hdds.. i want to take a external backup of everything to take home once a week for safe keeping incase of fire/flood or raid failure. what is the best way to do this? i can get another md1000 with 13 x 1tb hdds in it also, put them into a padded brief case and take them home once a week. insert another
[14:16] <stoogle> 13 x 1 tb hdd in there and then copy everything over again until i swap again? i just need it to be reliable!
[14:17] <sw> Oh, appears it works as we want by default anyway.
[14:18] <xnox> stoogle: you dropped a connection. look at incremental back ups to tape. as by taking a copy home you simply add a redundancy against physical location, but not e.g. silent corruption of data in _both_ arrays (the one at primary location and the one at home)
[14:19] <xnox> and / or cloud storage & backups which are becomming very cheap for very large data sets.
[14:20] <stoogle> mmm. cloud probably would not work due to the huge ammount of data. i understand doing daily backups of only new or modified data onto a cloud. but it would take a week to download on adsl
[14:21] <stoogle> people have suggested tape drives but it really does put me off as i heard they are really unreliable
[15:26] <GrueMaster> Hmm, he left.  I was going to say that I have successfully restored data from some of my old QIC-80 tapes (250M) from my consulting days in the early 90's.  Hardest part was resurecting an old 486 with floppy controller, dos, and Central Point Backup (had to dig deep into my archives for that).
[16:37] <markthomas> Happy $LOCALTIME, everyone.  Does anyone know how I can simulate a udev device insertion event for a USB device I don't have?  In older versions, I think there was a udevtest utility, but I can't find one in Precise.
[17:23] <zul> hallyn:  http://www.gossamer-threads.com/lists/openstack/dev/26505?do=post_view_threaded#26505
[17:30] <adam_g> zul:  http://people.canonical.com/~agandelman/ca/folsom/2012.2.3/
[17:31] <zul> adam_g:  +1 (im not here btw)
[17:34] <hallyn> zul: ?
[17:42] <GeorgeJ> Hello folks!
[17:42] <GeorgeJ> Is there any guide one could follow to install ubuntu 13.04 server with a btrfs root?
[17:44] <xnox> GeorgeJ: use manual partitioning, select btrfs, see Ubuntu Server Guide on how to enter manual partitioning.
[17:44] <sarnold> GeorgeJ: (a) I'm not sure btrfs is mature enough for that use yet (b) I think it is just selecting a different fstype in the installer, right? (c) I understand you can convert ext3 or ext4 to btrfs inplace afterwards .. check wikipedia for info there
[17:44] <xnox> GeorgeJ: note, I would not recommend you to use btrfs =)
[17:44] <xnox> best to install straight away onto btrfs, rather than conversion
[17:45] <sarnold> xnox: ack, thanks :)
[17:46] <GeorgeJ> It's a personal webserver, theres no important data. And, I've read some recent reviews of BTRFS, there's usually no loss of data involved, even in tested, production load environments.
[17:46] <GeorgeJ> xnox: Do the same limitations described here apply? https://help.ubuntu.com/community/btrfs#Fresh_Install_on_11.04_Natty
[17:47] <GeorgeJ> Disregard the section, I ment to link to "Ubuntu-specific subvolume layout in 11.04 and later"
[17:48] <xnox> yes.
[17:49] <xnox> GeorgeJ: i'm not sure why are you installing btrfs then.....
[17:49] <GeorgeJ> xnox: Snapshots.
[17:49] <xnox> GeorgeJ: install with lvm2 and use snapshots. They are far more stable & easier than btrfs snapshots.
[17:50] <xnox> lvm2 is the default option when doing server install, with one click.
[17:50] <GeorgeJ> I'm allready using LVM, it's really not the same thing.
[17:52] <GeorgeJ> LVM doesn't know about the FS, it just keeps COWs blocks afaiks. Also, snapshot require pre-set sizes. While on BTRFs I can create as many snapshots as I want, without having to worry about much.
[17:52] <GeorgeJ> There's even a plugin to apt to snapshot before each apt-get operation.
[17:53] <xnox> with lvm2 when you run out of space a snapshot is dropped; with btrfs you run out of disk space on your '/' and it all becomes very sad =)
[17:53] <GeorgeJ> I'm mostly just giving it a whirl, to test it out. I'll probably test it on my personal laptop aswell if I'm pleased.
[17:53] <GeorgeJ> Yeah, that seems to be an issue.
[17:54] <sarnold> GeorgeJ: thanks for giving it a test :)
[17:54] <GeorgeJ> sarnold: Hehe, I'd test ZFS too, but, there seem to be some performance issues.
[17:56] <GeorgeJ> I like ZFS's volume managing better than BTRFS tbh. But atm, the only thing that ZFS implements that BTRFS doesn't(yet) is deduping, which I don't care much for, tbh.
[18:01] <GeorgeJ> Gotta love that most usb creators fail to create a good usb instalation drive, and dd works without a single issue.
[18:01] <GeorgeJ> After creating an usb drive, with the usb installer creators, the instalation fail to find a cd drive.
[18:10] <Praxi> does this message say that I'm using USB 2 on the port I plugged a drive into?  usb 2-2: new high-speed USB device number 12 using ehci_hcd
[18:19] <genii-around> Praxi: ehci is USB2, yes
[19:15] <GeorgeJ> Automatic partitioning has setup an UEFI partition, however, ubuntu doesn't seem to boot from the HDD. However, UEFI boot works with the usb pendrive, is there any reason for this?
[19:41] <HSak> Hello, I did some apt-get update on ubuntu server 12.04 2 days ago. After the restart, I'm getting "The disk drive for / is not ready yet or not present Continue to wait; or Press S to skip mounting or M for manual recovery."
[19:41] <HSak> Are there anything I can do to fix it?
[19:44] <RoyK> HSak: start with m
[19:44] <RoyK> HSak: apt-get update won't upgrade anything, though, it just updates the apt index
[19:44] <RoyK> HSak: on a raid?
[19:44] <HSak> hmm I wonder what's wrong then. Because that's the only thing I did
[19:44] <HSak> Yes
[19:44] <HSak> I have raid 5
[19:44] <RoyK> probably a dead drive
[19:45] <HSak> hmm I don't think so.
[19:45] <RoyK> pastebin /proc/mdstat
[19:45] <HSak> The hdd is around 1 month old
[19:45] <RoyK> doesn't matter
[19:45] <HSak> and that can happend after restart too?
[19:45] <HSak> i readed somewhere it could be fstab is the problem
[19:45] <RoyK> drives usually die either the first few months or after a year or five
[19:46] <HSak> hmm ic
[19:46] <RoyK> 21:45 < RoyK> pastebin /proc/mdstat
[19:46] <HSak> I'm not with the server atm.
[19:46] <RoyK> no ssh access?
[19:46] <HSak> nope, not even internett want to start
[19:46] <HSak> i did skip skip and got in, connected to internett. Couldn't connect to it with ssh from laptop
[19:48] <HSak> I'm heading home in 1-2hours. Then I will try and do /proc/mdstat
[19:48] <RoyK> well, ask again when you have console access. it can be anything
[19:48] <HSak> okey
[19:48] <HSak> thanks
[19:49] <RoyK> "cat /proc/mdstat"
[19:49] <RoyK> mdstat isn't a command :P
[19:51] <HSak> okey : )
[19:51] <HSak> I will probably ask again when I'm home
[19:59] <glitchd> hello all
[19:59] <glitchd> anyone have any experience with openbox?
[20:00] <RoyK> !ask | glitchd
[20:01] <glitchd> RoyK,  my server with open box installed will not start x, the startx command just reads back ".Xauthority" errors.
[20:01] <RoyK> glitchd: sorry - just barking - I don't know openbox
[20:01] <glitchd> shit.
[20:02] <glitchd> well for anyone that might think they know, this is the exact output of "startx"
[20:02] <glitchd> :~$ sudo startx
[20:02] <glitchd> xauth:  error in locking authority file /home/glitchd/.Xauthority
[20:02] <glitchd> xauth:  error in locking authority file /home/glitchd/.Xauthority
[20:02] <glitchd> Fatal server error:
[20:02] <glitchd> Server is already active for display 0
[20:02] <glitchd> 	If this server is no longer running, remove /tmp/.X0-lock
[20:02] <glitchd> 	and start again.
[20:02] <glitchd> Please consult the The X.Org Foundation support
[20:02] <glitchd> 	 at http://wiki.x.org
[20:02] <glitchd>  for help.
[20:02] <RoyK> !pastebin | glitchd
[20:02] <glitchd>  ddxSigGiveUp: Closing log
[20:03] <RoyK> also, X questions belong in #ubuntu, not here, this is a server-centric channel
[20:03] <glitchd> welp, i just figured it out..lol
[20:04] <glitchd> i guess i had a zombie X session still running somewhere some how
[20:04] <glitchd> "sudo pkill X" fixed the problem for me
[20:05] <glitchd> i tried "killall X" but killall is depreciated so i had to use pkill
[20:05] <glitchd> *pkill X
[20:05] <RoyK> glitchd: X isn't related to servers
[20:06] <glitchd> RoyK, well im not always the best at command line, so i run a gui on my server
[20:06] <glitchd> a gui requires X
[20:06] <glitchd> so my server has X
[20:07] <RoyK> glitchd: really!
[20:07] <glitchd> RoyK, yupyup
[20:07] <RoyK> glitchd: it doesn't take you long to learn the commandline
[20:07] <RoyK> glitchd: this channel is for servers, meaning commandline
[20:07] <RoyK> period
[20:08] <glitchd> thats kind of a shitty way not offer help, thx anyways, i figured my problem out.
[20:08] <glitchd> *thats kind of a shitty way to not offer help, thx anyways, i figured my problem out.
[20:26] <RoyK> some people don't get the idea of what a server is - glitchd certainly had a glith there :P
[20:30] <GeorgeJ> RoyK: Well, the X server IS a server. So.. heh
[20:31] <RoyK> heh
[20:31] <RoyK> it's not a server thing, though
[20:32] <GeorgeJ> Isn't it? What if he needed remote X sessions, and the server is still a headless box?
[20:32] <RoyK> he was talking about X running on a server
[20:33] <genii-around> GeorgeJ: xvfb ?
[20:39] <Arrick> can someone point me to the ubuntu server specific man page for getting NTLM SSO working with apache2?
[20:40] <Arrick> specifically I am trying to follow this, but nothing matches for commands.
[20:40] <Arrick> http://code.google.com/p/mod-auth-external/wiki/Installation
[20:53] <GeorgeJ> What tool could one use to test a network connection? I'm interested in speed, dropped packets, etc.
[20:53] <GeorgeJ> I've got two ubuntu machines running on the same network.
[20:54] <sarnold> GeorgeJ: ping -f is a good first-attepmt
[20:54] <sarnold> (just don't aim against an os x box, they rate-limit icmp packets)
[20:55] <GeorgeJ> I don't think that's enough. I'm trying to test this NIC I modiified: http://imgur.com/a/uwKtH
[20:56] <GeorgeJ> I want to check wether extending the port like that affects network performance
[20:58] <sarnold> GeorgeJ: hahaha, that's awesome. and a bit terrifying. :)
[20:58] <GeorgeJ> Yeah, I'm brave(crazy) like that.
[20:59] <sarnold> GeorgeJ: you can specify some 'payload' data to send in your ping packets, you can grow the packets quite large, so you're not just stuck testing little 56 byte packets all day long...
[20:59] <GeorgeJ> Is the integrity of the data checked?
[20:59] <sarnold> GeorgeJ: if you fiddle with the payloads, you can check if it'll swap 0 bits to 1 bits or the other way around...
[20:59] <sarnold> GeorgeJ: every packet is crc32'd at wire level; you can see error counts with ifconfig
[21:00] <GeorgeJ> Ah, fair enough. And I guess it should count as dropped packets.
[21:03] <sarnold> granted crc32 won't catch them all, but it'll catch a lot. I think ping will repotr the others, but .. I haven't injected those sorts of faults to find out :)
[21:12] <HSaka> RoyK you there?
[21:12] <HSaka> I'm here now
[21:12] <RoyK> I am
[21:13] <HSaka> just booted now, I've got this. " Ext4-fs(sdb1):re-mounted. Opts: errors=remount-rp.
[21:13] <RoyK> rp? or ro?
[21:14] <HSaka> hda-intel: no codecs found! the fisk drive for /dev/mapper/cryptswap1 is not ready yet or not present and it says the same for /mnt/big
[21:14] <HSaka> ro
[21:14] <HSaka> sorry typed wrong
[21:14] <RoyK> fakeraid?
[21:15] <HSaka> what do you mean?
[21:15] <RoyK> HSaka: pastebin /dev/mdstat
[21:15] <RoyK> !pastebinit
[21:16] <HSaka> Royk, It's not possible for me to write everything in pastebin. cuz it's happening on the other computer
[21:16] <HSaka> physicaly
[21:16] <RoyK> is it mdraid or some sort of hardware raid_
[21:16] <RoyK> ?
[21:16] <HSaka> mdraid is software raid right?
[21:17] <RoyK> ys
[21:17] <RoyK> yes
[21:17] <HSaka> yeah it is mdraid
[21:17] <RoyK> so what /dev/mdstat has to tell?
[21:17] <RoyK> s/has/have/
[21:18] <HSaka> i need to skip it first?
[21:19] <RoyK> HSaka: just cat /etc/mdstat
[21:19] <RoyK> if something is broken, it'll show
[21:19] <RoyK> erm
[21:20] <RoyK> cat /proc/mdstat
[21:20] <HSaka> i type " /etc/mdstat" it sayd no suck file or directory
[21:20] <RoyK> my fault
[21:20] <HSaka> ?
[21:20] <RoyK> cat /proc/mdstat
[21:21] <HSaka> Personalities: raid 6, raid 5, raid 4, linear, multipath, raid 0, raid 1 raid 2
[21:22] <RoyK> nothing about the raid?
[21:22] <HSaka> md0 : active raid5 sdc[1], sdd[3], sda[0]. 5860270080 block super 1.2 level 5, 512k chunk, algorithm 2[3/3] [UUU]
[21:22] <HSaka> unused device: none
[21:23] <RoyK> that's a healthy raid-5
[21:23] <RoyK> mount it
[21:23] <RoyK> or run fsck -f /dev/md0
[21:24] <RoyK> don't fsck a mounted filesystem
[21:25] <HSaka> if you want to check the consistency of an xfs filesystem or repair a damged file system, see xfs_check(8) and xfs_repair(8)
[21:25] <RoyK> yes
[21:26] <RoyK> make sure the filesytem isn't mounted, check it and peraps do a repair
[21:27] <HSaka> could uou tell me step by step how to do it? I'm still new to linux
[21:27] <RoyK> pastebin df -h
[21:27] <RoyK> !pastebin | HSaka
[21:30] <HSaka> k
[21:32] <HSaka> http://pastebin.com/WDx748YE
[21:35] <RoyK> and /proc/mdstat?
[21:35] <RoyK> pastebinit
[21:36] <HSaka> permission denied
[21:37] <RoyK> cat
[21:37] <RoyK> you can't run a text file
[21:37] <RoyK> cat /proc/mdstat | pastebinit
[21:38] <genii-around> If it's a straight text file you can just do: pastebinit /filepath/filename
[21:40] <HSaka> http://pastebin.com/YCrCkKpG
[21:40] <HSaka> I have to write everything over to laptop
[21:44] <Iszak> I've got a ubuntu 12.04 LTS server, It's already installed, anyone got a guide on setting up RAID?
[21:44] <Iszak> s/RAID/RAID 1/
[21:44] <ScottK> Iszak: Did you look in the server guide that's listed in the channel topic?
[21:45] <Iszak> No, will take a look.
[21:45] <HSaka> Royk, any clue?
[22:17] <halvors1> How do i setup an ubuntu-server as a router, which routes between 2 subnets? Subnet 1 is internet and subnet 2 is lan.
[22:18] <halvors1> Subnet 1 has a gateway which nats :)
[22:23] <ScottK> halvors1:  Did you look in the server guide that's listed in the channel topic?
[22:39] <halvors1> ScottK: Nothing about it there...
[22:40] <ScottK> halvors1: OK.  It's been long enough since I've done it, I don't have any great suggestions.  Please, while you're doing it, take good notes and we'll write them up and get that fixed, ok?
[22:41] <ScottK> Documentation written by people who already know stuff is usually useless to people that don't, so we need someone like you to help.
[22:48] <shauno> is there anything on 'subnet 1' that needs to reach 'subnet 2' ?
[22:49] <sarnold> halvors1: potentially useful to you -- though it won't be any help at all for what to write to which configuration files: http://www.lartc.org/
[23:03] <shauno> it should be as simple as net.ipv4.ip_forward = 1, but the gotcha will be architectual; subnet 1 doesn't know to use the server as a gateway to reach subnet 2 (and the existing gateway won't know to point subnet 2's replies at the server).  the easy out is just the classic nat/masq with subnet 2 as the 'inside'
[23:09] <GeorgeJ> Awesome, no packet loss whatsoever!
[23:09] <sarnold> GeorgeJ: nice
[23:29] <RoyK>  
[23:59] <halvors1> ScottK: I thought it was just to enable ip forwarding in /etc/sysctl.conf? Anything more i have to do? That setup is working with masquarading...
[23:59] <ScottK> not sure