[01:22] <smoser> hallyn, no, user data scripts do not get any information about their launch-index. that was a feature request made recently by hazmat, and i plan to have something for that soon.
[01:22] <smoser> feel free to open a bug, though (i'd appreciate if you do, specifically mentinging desire for "launch-index" available somehow -- likely it will be environment)
[04:50] <blendedbychris> anyone know tools to follow a request from haproxy->nginx->php-fpm->glustterfs and figure out latency issues?
[04:51] <blendedbychris> i'm having sparatic response time issues with my web cluster nodes
[04:51] <blendedbychris> it'll jump from 300ms to 3000s
[04:51] <blendedbychris> er
[04:51] <blendedbychris> 3s
[09:25] <mcloy> can anyone see my text?  is there a way not to install kde (remove it) but run dolphin or xchat or a video (just these applications) and do all work at command line?
[09:31] <twb> ikonia: is that the same guy as was doing it about two weeks ago?
[09:32] <ikonia> twb: yes
[11:31] <master> hello, i have a problem with my NICs, both of them are gigabit however, for some reason I can't get them to work on gigabit. They're stuck ot 100 mbs. Tried to set the manually with ethtool however no result
[11:52] <haxxpop> hi, I'm try to use Phusion Passenger, but Apache2 doesn't know RailsBaseURI command. What should I do? I only did the installation before I asked you.
[11:55] <sofco3> how can i change an NIC speed from  Ethernet 100Mb/s to gigabit. the nic supports its
[11:56] <sofco3> i dunno why it defaults to 100
[11:57] <mardraum> what is it plugged into?
[11:57] <e_t_> How have you determined the speed?
[12:03] <sofco3> mardraum: into tplink 1043 gigabit router with 2 meter cable
[12:03] <sofco3> e_t_:  i'm setting a homeserver and noticed that file transfer don't peak above 12 mb/s
[12:04] <sofco3> e_t_: used webmin is showing speed at 100mbs and also used ethtool
[12:04] <e_t_> Is it a cat5e or cat6 cable that will support gigabit speed?
[12:08] <sofco3> e_t_: its a cat5e
[12:10] <e_t_> Can you pastebin the output of 'lspci' ?
[12:11] <sofco3> http://paste.ubuntu.com/1009592/
[12:13] <sofco3> I'm using the broadcom nic
[12:13] <e_t_> Do you get any better speed if you use the Realtek ?
[12:15] <sofco3> nope, still goes to 100 mbs
[12:16] <sofco3> http://paste.ubuntu.com/1009599/
[12:17] <e_t_> Same problem on two disparate cards means it's not hardware-related.
[12:20] <sofco3> e_t_: yep, i think that is some configuration info
[12:21] <e_t_> What's the output of 'cat /etc/udev/rules.d/70-persistent-net.rules' ?
[12:25] <sofco3> http://paste.ubuntu.com/1009615/
[12:26] <sofco3> just did a quick recabling so now the server and the desktop go not directly to the router but via gigabit swich, still 100 mbs
[12:30] <e_t_> It looks like this is not a unique problem [ https://www.google.com/search?q=ubuntu+gigabit+100mb+only ] but I didn't see a sure solution, sorry.
[12:37] <sofco31> hmm
[12:48] <sofco31> hmmm
[12:48] <sofco31> maybe i found the problem
[12:48] <RoyK> e_t_: you can run gigabit over good old cat5 these days too
[12:50] <twb> RoyK: IIRC cat5e is not *rated* for gigE, which means that while it will usually work (especially for short runs), it is not guaranteed to by quality specifications.
[12:50] <RoyK> twb: cat5e is, cat5 isn't ;)
[12:51] <twb> RoyK: I believe you are wrong
[12:51] <sofco31> gotta go, but from first look it's some stupid bug >:( after 2 days of brainstorming it appears that for some fucking reason you can't connect to gigabit on auto config.
[12:51] <RoyK> I beleive I'm right ;)
[12:51] <sofco31> after everything is booted you have to manually do a ifdown and ifup
[12:51] <RoyK> http://www.connectworld.net/syscon/support.htm
[12:52] <sofco31> and they'll connect to gigabit
[12:52] <sofco31> wtf....
[12:52] <RoyK> twb: some info there http://www.connectworld.net/syscon/support.htm
[12:53] <twb> https://en.wikipedia.org/wiki/Cat5e indicates it is often used for gige but makes no mention of it being rated for same
[12:53] <RoyK> sofco31: some autonegotiate bits vary between makers, on 100Mbps it was rather bad, on gigabit it's been far better. the trick back in the 100Mbps days, was to use fixed speed and duplex rate
[12:53] <RoyK> twb: from the link above "CAT-5e components were designed with high-speed gigabit Ethernet in mind."
[12:54] <twb> gigE standard requires autonegotiation, btw.
[12:54] <twb> RoyK: well, OK, thought that's not quite the same :-)
[12:54] <RoyK> oh, it does?
[12:55] <twb> RoyK: AFAIK yes
[12:55] <twb> RoyK: I can't back that up with facts atm tho
[12:55] <twb> (bedtime)
[12:55] <RoyK> from http://en.wikipedia.org/wiki/Gigabit_Ethernet - IEEE 802.3ab, ratified in 1999, defines gigabit Ethernet transmission over unshielded twisted pair (UTP) category 5, 5e, or 6 cabling and became known as 1000BASE-T. With the ratification of 802.3ab, gigabit Ethernet became a desktop technology as organizations could use their existing copper cabling infrastructure.
[12:56] <RoyK> sofco31: seems twb is right - gigE *requires* autoneg
[13:42] <reisi> is anyone else spending their weekend at work because of the new kernel image?
[13:42] <reisi> that is on 10.04 lts server; linux-image-2.6.32-41-server
[13:43] <reisi> we've got every process and task hanging like http://pastebin.com/1bGnpAHp
[13:43] <reisi> loads are 30+, normally < 1
[13:47] <reisi> ok.. now that kvm FINALLY booted (only one guest) everything started to get unstuck
[13:47] <reisi> this is 22 minutes after booting
[14:11] <koolhead17> hi all
[14:11] <reisi> well, i guess my problem died away by killing collectd
[15:57] <tarvid> system hang this morning - http://paste.ubuntu.com/1009834/
[15:57] <tarvid> last message - May 27 06:51:14 fairfax2 kernel: [1294212.040024] ata3: SRST failed (errno=-16)
[15:59] <patdk-lap> kind of hard for a system to keep working
[15:59] <patdk-lap> when it's harddrive is failing
[16:07] <Soekris> hello
[16:07] <Soekris> I got som troblues met my self and get MAAS notworking
[16:08] <Soekris> I got 2 nodes with status REady is that oke ?
[16:08] <Soekris> so i can run juju bootstrap ?
[16:13] <tarvid> there are no hard drive failure messages
[16:13] <tarvid> and this is RAID0 and no recovery action was taken
[16:14] <patdk-lap> no failure messages? what do you call all those messages?
[16:14] <patdk-lap> it comes down to a few issues, harddrive, controller, or cable
[16:15] <qman__> yeah, one of those three failed
[16:15] <tarvid> supposed to be RAID0 , I don't see any attempt at recovery
[16:16] <qman__> there wouldn't be one
[16:16] <qman__> the drive quit while operating, raid 0 doesn't offer redundancy, nothing to recover
[16:16] <tarvid> two hard drives, two cables
[16:16] <qman__> and?
[16:16] <qman__> those messages indicate hardware failing, it could just be a loose cable but it's probably the drive
[16:17] <qman__> raid 0 does not offer any redundancy, there's nothing to recover from, if either drive goes you're done
[16:17] <qman__> it's strictly for increased performance
[16:17] <tarvid> hmm, I am not sure how to check the raid array
[16:17] <qman__> there's nothing to check
[16:18] <tarvid> it is running at the moment
[16:18] <qman__> check the hard drives' smart info
[16:18] <patdk-lap> if it's raid0 there is only one answer, it's screwed
[16:18] <qman__> smartctl
[16:18] <patdk-lap> unless you can get that drive working again
[16:18] <tarvid> maybe it is raid 1 mirroring
[16:18] <patdk-lap> this is hopeless
[16:19] <patdk-lap> we can only hlep based on what you can tell us
[16:19] <qman__> those error messages have _nothing whatsoever_ to do with raid
[16:19] <qman__> they're hardware errors
[16:19] <qman__> one of your drives, cables, or controllers failed
[16:19] <tarvid> installing smartmontools
[16:21] <tarvid> noW SMARTCTL WAS HELPFUL
[16:24] <patdk-lap> well, it's only as useful to read the data from the drive
[16:24] <patdk-lap> so the better the drive is at knowing what is wrong, the better smartctl will be
[16:27] <qman__> right
[16:28] <qman__> and if the drive is showing a bunch of hardware errors, the drive is going bad
[16:28] <qman__> if not, it might be going bad but might also be a cable or controller problem
[16:33] <tarvid> http://paste.ubuntu.com/1009931/ both sda and sdb  do not look promising
[16:37] <tarvid> http://paste.ubuntu.com/1009938/ mdadm -D /dev/md0
[16:37] <tarvid> looks like it did what it was supposed to do - sda removed and I am up with sdb
[16:40] <tarvid> thanks for the help, maybe I learned something
[17:59] <Ketsueki_K> anyone have experience with ARToolKit?
[18:16] <harushimo> I'm trying to run the ubuntu cloud cd
[18:16] <harushimo> it isn't working
[18:17] <harushimo> I keep getting this error message: This kernel requires x86-64 CPU. but only detected an i686 cpu. Unable to boot
[18:17] <harushimo> what is that error message mean?
[18:19] <harushimo> any ideas
[18:19] <harushimo> ?
[20:57] <tarvid> what is a good alterntive to vsftpd?
[20:58] <tarvid> what are ny chances of forcing everyone to use sftp?
[21:02] <RoyK> tarvid: scp or rsync over ssh?
[21:02] <RoyK> rsync uses ssh by default these days, since, what, 2005 or so?
[21:03] <tarvid> most of my clients use filezilla
[21:04] <tarvid> rsync makes sense
[21:04] <tarvid> i use it for inter server transfers
[21:11] <RoyK> doesn't filezilla support ftp over ssh?
[21:16] <smw> RoyK, it probably supports sftp
[21:17] <smw> RoyK, which is NOT FTP over ssh but is a file transfer protocol that works over ssh
[21:17] <RoyK> I've seen sftp used as both ftp over ssh and ftp over ssl, a bit confusing
[21:18] <smw> RoyK, no, sftp is a separate protocol
[21:18] <smw> ftps is ftp over ssl
[21:18] <smw> RoyK, but yes... I see how this might cause confusion
[21:19] <smw> RoyK, http://en.wikipedia.org/wiki/SSH_File_Transfer_Protocol
[21:19] <RoyK> mhm
[21:19] <tarvid> going to take some play
[21:20] <zastaph> how far are aptitude usually behind release schedule?
[21:20] <tarvid> first try suggests manual transfer over stfp will work from filezilla
[21:20] <zastaph> for instance git is at 1.7.5.4 but current is 1.7.10.2
[21:21] <smw> RoyK, http://en.wikipedia.org/wiki/File_Transfer_Protocol#Secure_FTP
[21:22] <smw> RoyK, that explains the difference in a nut shell
[21:22] <RoyK> I know... not very nerdy tonight, after 44km on the bike
[21:22] <smw> lol
[21:24] <RoyK> smw: I'm quite aware of the differences, I just find it confusing that SFTP is used for both. I also find it somewhat worrying that people still use FTP, a protocol that should have been abandoned years ago
[21:27] <streulma> RoyK: still asking the same question as a few days ago ?
[21:29] <smw> RoyK, here is my opinion on ftp: http://mywiki.wooledge.org/FtpMustDie
[21:29] <RoyK> streulma: me? asking questions? about what?
[21:30] <qman__> RoyK, SFTP isn't used for both, FTP over SSL is FTPS
[21:30] <qman__> anybody who refers to FTPS as SFTP is just plain wrong
[21:31] <RoyK> qman__: I didn't say it was a standard, I just said it's used for both by certain parties
[21:31] <smw> qman__, people do say sftp for ftp+ssh
[21:31] <smw> qman__, this is also wrong... but too comon
[21:31] <qman__> and those people are wrong
[21:31] <smw> common*
[21:32] <qman__> of course, someone's always going to be wrong on the internet
[21:32] <RoyK> http://xkcd.com/386/
[21:32] <smw> qman__, I am saying that it is too common to be ignored
[21:32] <qman__> I disagree
[21:33] <RoyK> qman__: if a million people make the same mistake, it's not merely a stupid error, but a trend, and should therefore be allowed focus
[22:42] <ringzer0> How do I get colorful tty?
[22:42] <ringzer0> e.g. blue directories in bash?
[22:44] <RoyK> should work by default
[22:46] <ringzer0> RoyK: not in the Rackspace ssh distro/whatever
[22:47] <RoyK> ls -F --colour=auto
[22:47] <ringzer0> Interesting, i type bash and the colors load up.
[22:47] <RoyK> ls -F --color=auto
[22:47] <RoyK> ah
[22:47] <RoyK> chsh
[22:47] <ringzer0> what is chsh?
[22:47] <RoyK> man chsh ;)
[22:48] <ringzer0> The chsh command changes the user login shell. This determines the name of the user's initial login command. A normal user may only change the login shell for her own account; the superuser may change the login shell for any account.
[22:48] <RoyK> yes, I know
[22:49] <RoyK> so chsh -s /bin/bash
[22:49] <RoyK> it'll change your default shell to bash
[22:49] <ringzer0> weird, yeah I did that
[22:49] <ringzer0> it seemed to change.
[22:49] <ringzer0> but then i logged out and back in via ssh
[22:49] <ringzer0> and it came back to non color
[22:49] <streulma> why is dash the default and not bash ?
[22:49] <RoyK> what does "ps $$" say?
[22:50] <RoyK> streulma: no idea - it's stupid
[22:50] <ringzer0> my etc/passwd is : root:x:0:0:root:/root:/bin/bash
[22:50] <ringzer0> root:x:0:0:root:/root:/bin/bash
[22:50] <ringzer0> oops
[22:50] <ringzer0> sorry
[22:51] <ringzer0> 7328 pts/0    Ss     0:00 -bash
[22:51] <streulma> bin sh is linked to bin dash
[22:51] <ringzer0> but when I make it work, by typing bash, it gives me: 7431 pts/0    S      0:00 bash
[22:51] <ringzer0> bash without the - in front
[22:52] <ringzer0> is -bash different than bash?  cause it works without the - after i type in bash again after logging in.
[22:52] <ringzer0> and which bash shows /bin/bash
[22:53] <ringzer0> could this have something to do with sshd?
[22:55] <streulma> ringzer0:I have also -bash
[22:55] <streulma> you have no colors ?
[22:55] <streulma> do something like sudo apt-get install htop
[22:55] <streulma> run the htop command
[22:56] <streulma> and see if you have colors
[22:56] <ringzer0> i see colors
[22:56] <ringzer0> if i type in bash
[22:56] <ringzer0> 'b-a-s-h' in the term, then it reloads, and I have colors
[22:56] <ringzer0> its like ssh'd in bash vs the bash binary are 2 different things.
[22:59] <streulma> I have a solution
[22:59] <streulma> what do you have in ~/.bashrc ?
[23:00] <streulma> a lot of text ?
[23:03] <streulma> hello ringzer0 ?