[01:31] <dassouki> is there a way i can host a dropbox type solution on my server
[01:31] <dassouki> with some form of security and allowing cliens to log in
[04:02] <TheEvilPhoenix> how can i find what's binding to what ports?
[04:09] <SpamapS> TheEvilPhoenix: sudo netstat -tnlp
[04:09] <TheEvilPhoenix> SpamapS:  without using netstat, i.e. lsof
[04:10] <SpamapS> TheEvilPhoenix: sudo lsof -n | grep LISTEN
[04:10] <SpamapS> but.. they both basically read the same place
[04:10] <SpamapS> not sure why netstat -sn't desirable. ;)
[05:38] <mrroth> how would I grep / awak this when the out shows me ""Wan1AliasMaskIp1"value='255'" "Wan1AliasMaskIp1"value='somevaule'  and all I want is that somevalue  this is as far as I gotten grep -i '"Wan1AliasMaskIp"*'|awk '{print $4}' | cut -c 8-11
[05:50] <twb> mrroth: ow
[05:51] <twb> mrroth: wtf is that from
[05:51] <mrroth> this is the output http://pastebin.com/bWyiZZ4r from printf %s $(curl -s -u sometuser:somepass http://hostname/some.html |grep -i '"Wan1AliasMaskIp"*')
[05:51] <twb> You realize that if you pass a password as a CLI argument, every user on the host can see the password as clear text?
[05:51] <twb> If you're parsing XML, use xmlstarlet, not awk
[05:52] <twb> If you're parsing SGML HTML, use tidy -asxml --numeric-entities yes to turn it into XML
[05:52] <twb> Not sure why you're doing printf %s, either
[05:53] <mrroth> oh tidy hmm
[05:54] <mrroth> so what would a easier syntax
[05:54] <mrroth> and it html
[05:55] <twb> Or use a dedicated scraping library
[05:55] <twb> Or best of all: use the REST or SOAP or whatever API that upstream provides specifically for use by non-humans
[05:56] <twb> curl -s URL | egrep name=[\'\"]?Wan1AliasMaskIp1 | grep -o value=[\'\"]?[0-9.]+ | head -1 | egrep -o [0-9.]+
[05:56] <mrroth> oh thanks
[05:56] <twb> What are you talking to, CPE?
[05:57] <mrroth> a rv042 firewall / router
[05:57] <twb> Can you reflash it with OpenWRT?
[05:57] <mrroth> I want for work, a quick cli script
[05:57] <mrroth> NOPE
[05:57] <twb> Bummer
[05:57] <twb> I'm amazed it works at all
[05:57] <mrroth> it bassically a small bussiness linksys
[05:57] <twb> Most of the firmwares I've seen are js out the wazoo
[05:57] <mrroth> with vpn
[05:58] <mrroth> yea It may be some js. but I was able to logon and grab subnet mask
[05:58] <twb> Reflashing with a normal linux distro would be my preferred choice in these cases
[05:58] <mrroth> but parse is what I want to do
[05:58] <mrroth> I am no programmer
[05:58] <twb> Either openwrt or just ubuntu on a 1RU peecee
[05:58] <mrroth> if it were me I would also
[05:59] <mrroth> do something else
[06:00] <twb> fai enough
[06:00] <twb> *fair
[06:03] <mrroth> so your sample syntax should do the trick
[06:06] <twb> That's what I use for screen-scraping image pages
[06:06] <twb> curl URL | egrep -oi [\'\"]+.jpe?g | wget -i- --base URL
[06:07] <mrroth> oh how I wish i was a programmer
[06:07] <mrroth> make my life simpler
[06:07] <twb> That's not programming man
[06:07] <twb> That's being a sysadmin
[06:07] <mrroth> or scirpting
[06:07] <mrroth> yea
[06:44] <red2kic> oh how I wish i was a millionare
[06:44] <red2kic> make my life simpler (too). :P
[06:46] <twb> red2kic: move to africa then
[06:46] <twb> Some of those places have wicked hyperinflation
[06:48]  * red2kic commands twb to make him some money!
[06:48] <twb> I think you missed the point
[06:48] <red2kic> Probably
[06:49] <twb> http://en.wikipedia.org/wiki/Hyperinflation
[06:50] <twb> Hell, even Japan still has a low net value for the Yen
[06:50] <red2kic> Inflation goes up. Real values going down.
[06:50] <red2kic> I treat every 5 bucks like it's 15 bucks here.
[06:51] <twb> So if you move there post-inflation, your foreign currency gets converted 1:1bn or whatever
[06:51] <twb> Thus, you are instantly a millionaire or billionaire
[06:51] <red2kic> Until it get stablized then everything goes higher again.
[06:51] <red2kic> Ya?
[06:52] <twb> That would be filed under "it all depends"
[06:52] <twb> In any case, your wish would be granted
[06:52] <red2kic> "Germany 1923, banknotes had lost so much values that they were used as wallpaper."
[06:52] <red2kic> Heh.
[07:14] <hypetech> Anybody know why my ubuntu server would be able to ping any IP except for the google public DNS IPs?  I've had an outgoing connection issue for days, and it turns out I just wasn't able to communicate with 8.8.8.8 and 8.8.4.4 but other IPs work fine :(
[07:25] <twb> hypetech: because google DNS servers don't respond to echo requests?
[07:25] <twb> Though FWIW "ping 8.8.8.8" and "ping 8.8.4.4" work for me
[07:26] <twb> Maybe your ISP is blocking access to them, to force you to use their broken DNS?
[07:26] <twb> Have you asked your ISP about it?
[07:27] <hypetech> twb: the server is in a datacenter, and we have other servers using google DNS from the same switch/firewall just fine
[07:27] <twb> Hm
[07:27] <twb> In that case maybe you fucked up your routing table or such
[07:28] <twb> Pastebin the output of "ip l;ip a;ip r"
[07:28] <hypetech> It's the same on a fresh install of 10.10 and 11.04
[07:29] <twb> Or maybe google have blacklisted your IP because they've decided you're a dickhead
[07:29] <twb> I can't think of a *likely* explanation, so I'm just brainstorming
[07:29] <hypetech> I considered that, but I don't think they block on DNS, plus the server didn't really do anything outgoing at all, much less anything that would cause them to block me :(
[07:30] <hypetech> plus I can ping google.com's main IP
[07:30] <hypetech> and others
[07:30] <hypetech> just not the DNS ones
[07:30] <twb> Maybe the last to have your IP did
[07:30] <hypetech> we've had the IP it uses for years :p
[09:22] <smoser> jamespage, around ?
[09:22] <jamespage> smoser: sure am
[09:22] <smoser> sweet.
[09:22] <smoser> let me check in #ubuntu-release really quick
[09:22] <jamespage> smoser: you're up early :-)
[09:22] <smoser> but i think we should start testing ISOs
[09:23] <jamespage> smoser: ack - I will just spin a small test to check archive mirrors OK?
[09:23] <smoser> s/isos/uec images/
[09:23] <smoser> jamespage, good thinking sir.
[09:23] <smoser> jamespage, have you made the changes to collect pids and initctl list ?
[09:24] <smoser> (hoping to get more info on that hang we see)
[09:24] <jamespage> smoser: ah - so initctl list is now part of the collected data and so is a full ps
[09:24] <jamespage> smoser: however I've not make the change to collect stuff on timeout if the instance is ssh-able.
[09:25] <smoser> i thought you were alrady doing that
[09:27] <jamespage> so it collects a console log; however it does not ssh in and collect data
[09:28] <smoser> oh.. not on hang.
[09:28] <smoser> but it does that at some point
[09:31] <smoser> jamespage, i'm thinking/hoping that it will give us enough data if we get those two things. at least give us some hints.
[09:32] <jamespage> smoser: it does a ssh in to collect data as soon as the instance is running - but that was a little early?
[09:33] <smoser> maybe. i dont know.
[09:34] <jamespage> yeah it was - look I'll do a fix now and then we can run the tests; I will trigger a collection via ssh on timeout if its ssh accessible
[09:35] <jamespage> smoser: archives all look OK - two tests in each region did not return any errors
[09:36] <smoser> how possible would it be to collect the same bit later?
[09:36] <smoser> maybe just before terminate?
[09:38] <maedox> Anyone here running and IRC server?  I want to set up a private on at work and I'm wondering what's the ircd of choice? There seems to be a bunch to choose from in the repos.
[09:45] <smoser> jamespage, ^
[09:45] <jamespage> smoser: sorry - lemme take a look - think we might already do that
[09:47] <jamespage> smoser: so we currently collect metadata 1) as soon as we can ssh 2) after the in-instance testing is complete and 3) as soon as its ssh-able after a reboot or stop/start
[09:47] <jamespage> 2) is just before reboot
[09:48] <smoser> so when is 2, if the "i'm done" file never appears
[09:53] <jamespage> smoser: thats the issue - if we hit the timeout we don't collect the metadata
[09:53] <jamespage> smoser: just running a quick test now on the change
[09:53] <smoser> oh. great.
[09:58] <jamespage> smoser: OK worked so going to kickoff the main test now.
[09:58] <jamespage> smoser: so will be testing 20110426
[09:59] <smoser> yes
[09:59] <jamespage> smoser: spinning now - http://jenkins.qa.ubuntu-uk.org/view/natty-ec2/job/natty_server_ec2/10/
[10:00]  * jamespage goes for coffee
[10:37] <jamespage> smoser: two failures so far - http://jenkins.qa.ubuntu-uk.org/view/natty-ec2/job/natty_server_ec2/10/ARCH=i386,REGION=eu-west-1,STORAGE=instance-store,TEST=multi-part-ud,label=ubuntu-server-ec2-testing/
[10:38] <jamespage> bug 712026
[10:38] <jamespage> and one capacity issue in us-west-1b
[11:46] <jamespage> smoser: around?
[11:46] <smoser> here
[11:47] <jamespage> smoser: OK - so I'm seeing a new 'condition' in the ec2 testing - InstanceInitiatedShutdown state
[11:47] <smoser> state?
[11:47] <smoser> you're getting that back in the api ?
[11:47] <jamespage> yes
[11:47] <jamespage> not supported by boto yet but I can see it in the logs
[11:48] <jamespage> I've had 3-4 instances of this across all of the testing.
[11:48] <jamespage> the instance starts and then switches to this state before it gets to running.,
[11:48] <jamespage> no console data (just checked)
[11:48] <smoser> that is strange.
[11:48] <smoser> but it does get to running ?
[11:48] <jamespage> no
[11:48] <smoser> oh.
[11:48] <jamespage> sorry - that was not clear
[11:49] <smoser> so this would be failed boot then
[11:49] <smoser> right?
[11:49] <smoser> where do you see it?
[11:49] <smoser> region
[11:49] <jamespage> http://tinyurl.com/6caqmvv
[11:50] <jamespage> us-east-1 by the looks of it
[11:50] <smoser> do you have ec2 api tools there?
[11:51] <smoser> i'm interested in what that says right now
[11:51] <smoser> ec2-describe-instances <instance-id>
[11:52] <jamespage> looking now
[11:54] <jamespage> smoser: http://paste.ubuntu.com/599225/
[11:55] <jamespage> smoser: they are all in us-east-1 - might be related to outage stuff I guess
[11:55] <smoser> bug 557483
[11:55] <smoser> its not new.
[11:55] <smoser> i suspect hypervisor/platform gremlins as the cause
[11:56] <smoser> and there is good reason to believe that there are gremlins hanging around us-east-1 datacenters
[12:05] <jamespage> smoser: got one cloud-init not completing - http://tinyurl.com/3hg8h2b
[12:05] <jamespage> smoser: some stack traces as a full in instance capture on timeout
[12:11] <smoser> jamespage, we weren't seeing those errors before though
[12:11] <smoser> i did suspect the apt... but no real good evidence
[12:12] <jamespage> smoser: so that may be a bit of a red herring - all of the cloud-config tests throw that exception
[12:16] <trapmax> how to enable ssl for apache?
[12:26] <pmatulis> trapmax: get a valid certificate and then tell apache about it.  guides abound
[12:28] <jamespage> smoser: all of the us-east-1 instances that failed to start where EBS as well
[12:32] <Dramaturg> Hey everybody! I have an issue with Spamassassin =( It seems to me it is loaing too few plugins: http://paste.debian.net/115085/ and is rating an obvious spam message with 1.4. In e-mail headers I see tests[none]. Where Do I start investigating the issue?
[12:35] <jamespage> smoser: any reason we don't have ebs tests in the test tracker for ap regions?
[12:35] <smoser> no
[12:35] <smoser> no reason. there should no
[13:00] <ScottK> Dramaturg: I would ask on a spamassassin specific forum.  Other than needing to know package names for additional things to install, there is very little that's Debian/Ubuntu unique about setting up spamassassin.
[13:01] <Dramaturg> ScottK: You know one good forum&
[13:01] <ScottK> No.  Sorry.
[13:20] <jamespage> smoser: I've populated the test cases in ISO tracker with all of the ec2 results
[13:20] <jamespage> and linked to the associated bug reports
[13:20] <jamespage> only one I have not done is the HVM stuff.
[13:48] <elb0w> root@dbrm:/home/gtsafas# service mysql start
[13:48] <elb0w> start: Job failed to start
[13:48] <elb0w> I just did a apt-get remove --purge mysql-server
[13:48] <elb0w> then reinstalled it
[14:03] <RoAkSoAx> morning all
[14:17] <moonpup> can anyone tell me where to find the 'fully commented' config files like the postfix main.cf instead of the abbreviated version in /etc/postfix
[14:18] <jMCg> moonpup: postfix.org? man 5 postfix?
[14:19] <jMCg> moonpup: they have... documentation!
[14:19] <jMCg> http://www.postfix.org/documentation.html
[14:19] <moonpup> does ubuntu not provide fully commented config files like red hat and other distros?
[14:19] <moonpup> i come from a red hat world that has well commented config files
[14:20] <jMCg> moonpup: I come from a world where documentation exists.
[14:21] <moonpup> jMCg i understand about the documentation, but i'm used to a distro providing the default fully commented config files
[14:22] <jMCg> That's a terrible habit. I think distros should provide empty configs with links to the documentation.
[14:22] <shauno> I'm on 10.04 LTS, the first line in my main.cf reads:
[14:22] <shauno> # See /usr/share/postfix/main.cf.dist for a commented, more complete version
[14:23] <jMCg> I think the single best advise I took from Ivan Ristic's book "Apache Security" is: When you start a new configuration, start with an empty file.
[14:23] <moonpup> shauno thank you, i'm new to ubuntu server and not used to their abbreviated config files
[14:24] <moonpup> jMCg i like to know what variables are available in a config file and the comments describing it without having to hunt around for documentation
[14:25] <jMCg> moonpup: if you have to hunt stuff down in documentation, then it's lousy documentation, and you should complain at the project's issue tracker.
[14:25] <moonpup> jMCg again, i'm a red hat guy trying to understand the ubuntu/debian way of doing things
[14:26] <robos> moonpup, I'm in the same boat
[14:26] <moonpup> jMCg it's no different if your comfortable in an ubuntu environemnt and then come over to red hat
[14:26] <robos> trying to learn the ubuntu way of things from redhat
[14:27] <kpettit> moonpup, the apache directory setup is setup a bit different by default
[14:27] <smoser> jamespage, you rock. thank you.
[14:27] <moonpup> both distros name packages differently, config files are in different places and package mgmt is completely different
[14:27] <robos> not trying to fork the conversation too much, but in redhat it's a huge nono to compile anything. Everything should be a rpm. Is the same school of thought with dpkg?
[14:27] <jMCg> 15:35:18 <fajita> Pop along to http://wiki.apache.org/httpd/DistrosDefaultLayout for an idea on where the main operating systems place Apache files.
[14:28] <jamespage> smoser: np - must get round to sorted out the HVM test as well
[14:28] <kpettit> moonpup, THis might help: https://help.ubuntu.com/10.04/serverguide/C/httpd.html
[14:28] <jMCg> moonpup: I work on Ubuntu/Debian, Fedora/RHEL, Solaris and FreeBSD on a daily basis.
[14:28] <smoser> jamespage, well, its not htat big of a deal. i have just run them by hand.
[14:28] <moonpup> kpettit thx, i've read that... good stuff
[14:28] <jamespage> smoser: well the framework supports it - I just can't pass it as an option at the moment!
[14:30] <moonpup> jMCg that's great, but i'm red hat certified and have spent my career on red hat so the ubuntu/debian way of doing things is new to me
[14:31] <moonpup> jMCg i'm just trying to figure it all out ;)
[14:32] <moonpup> thanks for the help... gotta run
[14:35] <robos> is there a way to tell which repository a package was installed from?
[14:36] <raphink> robos, you can see which repository provides a package
[14:36] <raphink> but not which repository provided a package
[14:36] <robos> bummer
[14:36] <robos> that's pretty bad :-/
[14:37] <robos> what happens when i update a package. How does the system know which repository to apply the updates from?
[14:37] <raphink> what's your need?
[14:37] <raphink> robos, it uses the package version first
[14:37] <Pici> apt-cache policy packagename might   provide the data you need
[14:37] <raphink> and applies weights from /etc/apt/preferences
[14:37] <raphink> Pici, that gives which repository currently provides the package
[14:37] <raphink> (if any)
[14:38] <jMCg> hrm... the problem with this:         su -s /bin/sh -c 'exec "$0" "$@"' komunalbedarf.at -- /opt/bw/bin/httpd-worker -f /etc/bw/apache/vhosts/komunalbedarf.at/httpd.conf -k start
[14:38] <Pici> raphink: *might* ;)
[14:38] <raphink> robos, a package installed from one repository might very well be upgraded from another one
[14:38] <raphink> (which is good)
[14:38] <robos> raphink, I inherited a bunch of ubuntu systems. My main concern right is if the version of Apache (and other things) are based off of SRU releases
[14:38] <raphink> it's a good thing that packages you installed from the main repositories are upgraded using the security repository for example
[14:38] <jMCg> Is that you cannot expect fork or expect daemon in upstart... Expect fork expects one, expect daemon two, but these should be three or four forks...
[14:39] <jMCg> And this is the wrong channel for this to discuss.
[14:39] <raphink> what version do you have robos ?
[14:39] <robos> raphink, that sounds good... but what if you want to control how your package is updated? For example, I want SRU versions and only want security fixes applied. I don't want any new features
[14:39] <raphink> (and which version of the OS)
[14:39] <raphink> robos, that's what apt/preferences is made for
[14:39] <robos> raphink, 2.2.12-1ubuntu2.2 and I'm running version 9.10
[14:40] <raphink> you can set weights by repository/releasename/package/version, etc. which will override the version weights
[14:40] <raphink> let me see
[14:40] <robos> cool; i need to learn how to do that asap
[14:40] <Pici> robos: 9.10 loses support at the end of this month, you really should upgrade.
[14:40] <raphink> 9.10... that's karmic
[14:40] <Pici> robos: support includes security updates.
[14:40] <Pici> !9.10
[14:41] <raphink> ubuntu|karmic|main|source: apache2 2.2.12-1ubuntu2
[14:41] <raphink> ubuntu|karmic-updates|main|source: apache2 2.2.12-1ubuntu2.4
[14:41] <raphink> I don't have karmic-security here, but it looks like that could be it
[14:41] <Pici> apache2 | 2.2.12-1ubuntu2.4 | karmic-security
[14:41] <raphink>    apache2 | 2.2.12-1ubuntu2.4 | karmic-security | source, all
[14:41] <raphink> right
[14:41] <raphink> :-)
[14:41] <raphink> rmadison said :-)
[14:41] <Pici> ♥ rmadison
[14:42] <raphink> robos, so your version is an old security update
[14:42] <raphink> 2 versions behind
[14:42] <robos> i have soo much to learn
[14:42] <raphink> is that a bad thing? :-)
[14:42] <robos> haha, nope
[14:44] <robos> so to be sure I understand, there is no way of telling if i'm running an SRU version of apache right now?
[14:45] <robos> SRU as in https://wiki.ubuntu.com/StableReleaseUpdates
[14:45] <raphink> sure there is
[14:45] <SpamapS> robos: sure there is, look at the top of the changelog ... if it is    distro-proposed then it is an SRU
[14:45] <raphink> SpamapS, you beat me to the explanation :-)
[14:45] <robos> is there some dpkg magic to find the changelog?
[14:46] <raphink> look in /usr/share/doc/$pkg/changelog.Debian.gz
[14:46] <raphink> it's more $EDITOR magic than dpkg ;-)
[14:47] <sbeattie> SpamapS: did you ever test php5 sbuild with the latest natty upstart? Cuz it hangs here.
[14:48] <robos> It doesn't appear to be distro proposed
[14:48] <SpamapS> sbeattie: not with the latest latest one no
[14:48] <robos> apache2 (2.2.12-1ubuntu2.2) karmic-security; urgency=low
[14:48] <SpamapS> sbeattie: I did test w/ 0.9.6 and it worked
[14:48] <SpamapS> sbeattie: also it varies based on what you have on your local machine.. do you have mysql server installed locally?
[14:48] <SpamapS> or rather, your... hosting machine..
[14:49] <sbeattie> SpamapS: I did at first, uninstalling it in the host didn't make a difference
[14:49]  * SpamapS isn't sure what the term is for the non chrooted OS. :)
[14:49] <SpamapS> sbeattie: damn.
[14:49] <SpamapS> and I see that I'm still running 0.9.6 :-P
[14:49] <SpamapS> speak of the devil :)
[14:50] <sbeattie> jhunt: did you ever test php5 sbuild with upstart 0.9.7? Cuz it hangs installing mysql-server here.
[14:51] <sbeattie> (either with or without mysql-server installed in the sbuild host)
[14:53] <jhunt> sbeattie: not aware of such an issue. could you raise a bug with more details?
[14:54] <sbeattie> jhunt: sure. what information do you need?
[14:54] <SpamapS> sbeattie: actually mine appeared to lock up until I killed mysqld
[14:54] <SpamapS> sbeattie: are you positive you're not getting stuck in the post-start loop ?
[14:55] <SpamapS> sbeattie: when I say killed mysqld.. I mean I had something else using port 3306 which was making mysqld fail to start
[14:55] <sbeattie> SpamapS: uninstalling the package killed it and that particular sbuild continued on (after I'd already sigint'ed out of it)
[14:55] <sbeattie> SpamapS: but a subsequent run, after the uninstall had completed, also hung in the same spot.
[14:56] <raphink> robos, when it says -security, it's a security update, so it's SRU
[14:56] <sbeattie> SpamapS: but no, I'm not sure of anything, it just sits silently at 'Setting up mysql-server-5.1 (5.1.54-1ubuntu4) ...'
[14:57] <raphink> robos, when it says -proposed, it means it went through -proposed and you probably got it in -updates (unless you're using -proposed, which is not recommended)
[14:57] <SpamapS> sbeattie: right, mine did too.. because it was spinning on mysqld failing/respawning
[14:57] <raphink> robos, when -security is definitely SRU
[14:58] <SpamapS> either way, I need to install the latest updates and try again anyway
[14:58]  * SpamapS notes that his Unity desktop gives him zero indication that there are updates to install.. and that is rather confusing
[14:58] <sbeattie> SpamapS: argh, yeah, there's still a mysqld process running.
[15:02] <SpamapS> sbeattie: probably left over in the schroot session.. which likely was not ended
[15:02] <sbeattie> yeah
[15:02]  * sbeattie reboots build-server to clear out any crap :-/
[15:02] <SpamapS> sbeattie: I think we may need to patch schroot to stop any upstart jobs added in the current session
[15:02] <SpamapS> how to do that.. I have no idea
[15:07] <robos> thanks raphink
[15:28] <robos> is there a way to look at any hardware errors in ubuntu? I'm looking for something like mcelog in redhat
[15:29] <Pici> !info mcelog
[15:31] <robos> Pici, that doesn't come with ubuntu by default, right?
[15:31] <Pici> robos: Correct.
[15:31] <robos> gotcha. So installing it won't help after a crash, correct?
[15:31] <robos> Any recommendations on how to figure out why this server crashed?
[15:32] <robos> nothing in /var/log/messages
[15:32] <robos> nor dmesg has any clues
[16:03] <robbiew> hggdh: hey...so going with your last comment...can we move bug 717166 back to Fix Released?
[16:04] <hggdh> robbiew: I am not sure -- it is true we cannot replicate, but I do not know what causes Will's issue
[16:05] <hggdh> robbiew: for a fact, he should not be getting the dhcpd errors
[16:05] <robbiew> hggdh: hmm...so I think I'll move it to Fix Released
[16:05] <robbiew> he can open a new bug
[16:05] <robbiew> and we can work it there
[16:07] <hggdh> robbiew: good enough. I personally see no real reason to keep it open, except for the purist in me
[16:10] <tyreza> hello there
[16:10] <tyreza> what 's LVM ? what it can do ?
[16:13] <patdk-wk> lvm is like a partition table
[16:13] <patdk-wk> but it can also do snapshots
[16:13] <patdk-wk> and very very limited raid1 type things
[16:14] <SpamapS> and raid0 actually
[16:14] <robbiew> hggdh: heh...if I we went by the "purist" in you or me...we'd have to fix every bug :P
[16:14] <hggdh> heh. So true...
[16:15] <tyreza> but i can't difference between lvm and disk partition
[16:15] <patdk-wk> sure you can
[16:15] <patdk-wk> spamaps, well not raid0 at all :)
[16:15] <patdk-wk> linier or raid1
[16:16] <SpamapS> striping is pretty much raid0
[16:16] <patdk-wk> you can concat disks, or you can span files onto multible disks
[16:16] <patdk-wk> but lvm doesn't stripe
[16:16] <SpamapS> [-i|--stripes  Stripes  [-I|--stripesize StripeSize]]
[16:16] <patdk-wk> or atleast I don't think it does :)
[16:16] <SpamapS> from man lvcreate
[16:16] <SpamapS> I've used it
[16:16] <patdk-wk> oh ok, in my usage, it doesn't stripe :)
[16:16] <SpamapS> to stripe across two RAID5's
[16:17] <patdk-wk> I just extend my lv to other pv's
[16:18] <sbeattie> SpamapS: I think the workaround I'll go with for php is to build-depend on mysql-server-core-5.1 and mysql-client-5.1, rather than pull in mysql-server-5.1
[16:19] <SpamapS> sbeattie: I've actually always wondered why it uses mysql-server anyway.
[16:19] <sbeattie> that will prevent mysqld from getting started during package setup.
[16:19] <SpamapS> sbeattie: we should think about creating a mysql-server-testharness that automates creating an ephemeral mysqld
[16:19] <sbeattie> SpamapS: it needs mysqld available during the testsuite, but it doesn't need the initscript et alia.
[16:20] <sbeattie> SpamapS: yeah, that has potential.
[16:20] <SpamapS> tho I bet the code for that exists in the mysql testsuite already ..i wonder if it could be extracted
[16:24] <tyreza> what is a logical volume ?
[16:27] <SpamapS> tyreza: an unemotional volume
[16:45] <RoAkSoAx> zul: ping
[17:00] <jamespage> ttx: around?
[17:07] <RoAkSoAx> SpamapS: jamespage kirkland smoser hallyn robbiew Daviey isn't there a meeting today?
[17:07] <Daviey> o/
[17:07] <robbiew> heh...yeah
[17:07] <robbiew> who's chair this week?
[17:08] <Daviey> isn't it kirkland this week?
[17:08] <RoAkSoAx> robbiew: zul I believe
[17:08] <zul> im at the openstack summit today
[17:10] <robbiew> well...given we release in 2 days...maybe we skip it this week?
[17:10] <robbiew> the only think I can think of is to tell folks to create blueprints
[17:10] <zul> pos wireless
[17:10] <Daviey> Blueprints.. Blueprints... Blueprints...
[17:12] <robbiew> jdstrand: https://bugs.launchpad.net/ubuntu/+source/samba/+bug/750786
[17:12] <robbiew> so I'm thinking we just lost this one
[17:12] <robbiew> grrrrr
[17:12]  * robbiew hates Incomplete status!!!!!!
[17:15] <jdstrand> robbiew: well, that is nmbd. I saw that bug, but nmdb starts fine here after several reboots. My bug was smbd. I wasn't sure they were related, so filed a new bug
[17:15] <robbiew> ah
[17:15] <robbiew> right
[17:15] <Daviey> Going to talk to jhunt to see if he thinks it is related
[17:16] <robbiew> will release note it for Natty
[17:17] <zul> jdstrand, amd64?
[17:18] <jdstrand> zul: my bug? yes. libvirt/kvm with 1768M disk, 256M swap (the rest to /) and 384M of ram
[17:18] <jdstrand> zul: everything else is iso defaults
[17:18] <zul> jdstrand: ok ill try to reproduce it as well
[17:27] <hggdh> and I am re-running the jenkins test in debug mode to see what happens there
[17:28] <Eitan> hey guys
[18:03] <loopback_br> somebody is having problem installing USB Wi-Fi with Ralink 2870 chipset on Ubuntu Server 10.04 Kernel 2.6.32-21-generic-pae?
[18:12] <RoyK> loopback_br: not me, but there might be fixes in the 2.6.35 kernel
[18:12] <RoyK> installable from standard repos
[18:13] <RoyK> maverick backport, really, but due to some unfixed issues in 3.6.32, they let that pass
[18:14] <loopback_br> apt-get is not upgrading my kernel
[18:15] <patdk-wk> you used, apt-get dist-ugprade?
[18:15] <RoyK> loopback_br: it is if you apt-get install linux-image-2.6.35-25...
[18:15] <RoyK> patdk-wk: dist-upgrade won't update to 2.6.35
[18:15] <RoyK> (last I checked)
[18:16] <patdk-wk> oh, to that kernel, nope
[18:16] <patdk-wk> you have to install it manually
[18:17] <loopback_br> okay i'll try
[18:43]  * RoAkSoAx off to lunch
[18:47] <loopback_br> One of my adapters worked but the other with the same chipset didn't
[18:49] <loopback_br> with kernel 2.6.32-31-generic
[18:55] <pangrazi> anyone here use ebtables or vlans?
[19:22] <bdmurray> kirkland: you might be interested in bug 771395
[19:23] <kirkland> bdmurray: looking
[19:24] <kirkland> bdmurray: hmm, yes, interesting
[19:25] <kirkland> bdmurray: what's the output of uname -a ?
[19:25] <bdmurray> 2.6.38-8-generic-pae
[19:25] <bdmurray> well plus some more
[19:27] <kirkland> bdmurray: okay, perhaps I should egrep for pae
[19:27] <bdmurray> kirkland: that might be a desktop system though couldn't it?
[19:28] <kirkland> bdmurray: yeah, perhaps ... i'd need to check your package list for ubuntu-desktop
[19:41] <robos> any general thoughts on compiling vs using dpkg?
[19:44] <pmatulis> robos: ubuntu is a package-based linux distro.  apt is very good and should be used.  not doing so can lead to serious problems unless you know exactly what you're doing
[19:44] <robos> package-based linux distro. That means all it's updates are handled via dpkg?
[19:45]  * RoAkSoAx back
[19:45] <hggdh> usually via apt-get
[19:45] <robos> So dpkg should be aware of everything on your system. Thus compiling is an absolute no-no. If compiling is needed we should build our own .deb package, correct?
[19:45] <hggdh> ideally, yes
[19:46] <RoAkSoAx> SpamapS: powernap's SRU is a 0 day SRU that you can "copy" to Oneiric with the same version number
[19:46] <hggdh> or you can install off the system places, like on /usr/local
[19:46] <RoAkSoAx> SpamapS: I've seen that done many times :)
[19:47] <SpamapS> RoAkSoAx: cool, I wasn't sure thats why I accepted it.
[19:47] <pmatulis> !apt | robos
[19:49] <RoAkSoAx> SpamapS: yeah, once Oneiric is open, then you'd have to release the SRU for natty, and then "copy" it to Oneiric with same version number
[19:51] <SpamapS> RoAkSoAx: hopefully pitti will have time to teach me some of that. :)
[19:52] <RoAkSoAx> SpamapS: :)
[20:35] <Anpheus_> Does anyone know why Ubuntu 10.10 Server's minimal virtual install (kernel -virtual) does not contain the Hyper-V modules, if it is possible to restore those to that kernel, and if this is rectified in the next release?
[20:37]  * Gnea notes the echo effect when saying anything here...
[21:06] <jMCg> Anpheus_: check the changelog for linux-image-virtual maybe that gives a clue.
[21:23] <robertj> hey all, I got a box that is coming up to busy box and says that /sbin/init is not found in the target, but mounting /dev/mapper/mybox-root /root works just fine, /sbin/init is there, life appears good...?
[21:24] <robertj> I was unmounting a whole bunch of lvm snapshots, it got stock and all disk related cmds began to act unhappy, gave it a reboot, and that's how I got into this predecimate
[21:25] <robertj> don't _think_ I unmounted the live root ;P
[22:19] <quentusrex> Anyone know what part of the ubuntu-vm-builder is suppose to generate the libvirt xml file?
[22:19] <quentusrex> I seem to be running into an issue where I can provision a VM, but the xml file is no longer loaded into the libvirt directory.
[22:50] <RoAkSoAx> TREllis: ping?
[23:17] <ignarps> quentusrex, I think qemu would be the one to write out the xml file.  permissions issue ?
[23:18] <ignarps> quentusrex, sorry it would be libvirt not qemu
[23:27] <quentusrex> ignarps, I don't see any errors even with --debug and -v turned on
[23:27] <quentusrex> and I'm running the provision script as root, so I don't think that could cause the permission issue.
[23:32] <quentusrex> ignarps, I don't see the vmbuilder script ever actually loading the image into libvirt. I do see the image converted to qcow2 though.
[23:32] <ignarps> what --libvirt= option did you use ?
[23:32] <quentusrex>  --libvirt qemu:///system
[23:33] <ignarps> man page shows = not sure if it is important
[23:35] <quentusrex> trying it with the =
[23:38] <quentusrex> ignarps, no difference.
[23:46] <grout> How can i mount my usb thumb drive on my ubuntu server?  I did a sudo fdisk -l but it dosnt come up.  The drive comes up in lsusb.
[23:51] <ignarps> quentusrex, I usually do this as my user account.  but have you checked this works okay.   virsh -c qemu:///system
[23:51] <ignarps> quentusrex, to see if you can use the libvirt shell okay
[23:51] <quentusrex> I do have the shell ok.
[23:53] <ignarps> quentusrex, running out of ideas.  Could it be an apparmor issue ?  Are you using a non default location for some files.
[23:54] <quentusrex> no non default locations. I haven't touched apparmor since I installed the box.
[23:54] <quentusrex> I do have ufw enabled, but I don't see anything in the script that can run into that.
[23:54] <ignarps> grout, after I connect a device I usually look in the kernel logs to see what the device is.  ie run dmesg