Friday, 2010-12-10

*** sirp1 has quit IRC00:13
*** johnpur has quit IRC00:14
*** hggdh has joined #openstack00:23
*** hggdh has quit IRC00:25
*** hggdh has joined #openstack00:27
*** irahgel1 has joined #openstack00:28
*** jimbaker has quit IRC00:31
*** joearnold has quit IRC00:37
*** aliguori has quit IRC00:45
*** GasbaKid has quit IRC00:54
*** GasbaKid has joined #openstack00:54
*** stewart has quit IRC01:02
*** Edwin-afk2 is now known as EdwinGrubbs01:02
*** ar1 has joined #openstack01:05
*** rlucio has quit IRC01:06
*** jc_smith has quit IRC01:07
*** kashyapc has joined #openstack01:12
*** stewart has joined #openstack01:13
*** GasbaKid has quit IRC01:17
*** zhengli has quit IRC01:18
*** dirakx has joined #openstack01:26
*** abecc has quit IRC01:28
*** abecc has joined #openstack01:31
*** jimbaker has joined #openstack01:50
*** rchavik has joined #openstack01:53
*** Cybodog has joined #openstack02:13
*** gaveen has joined #openstack02:17
*** kashyapc has quit IRC02:24
KyleM1Anyone still around?02:24
*** matiu_ has joined #openstack02:25
*** matiu_ has quit IRC02:27
*** rds__ has quit IRC02:33
Ryan_LaneKyleM1: I'm here shortly02:35
Ryan_Lanenot sure I'll be able to help ;)02:35
KyleM1Oh cool.02:35
KyleM1Getting this:02:36
KyleM1# euca-describe-instances02:36
KyleM1Warning: failed to parse error message from AWS: <unknown>:1:0: syntax error02:36
KyleM1EC2ResponseError: 403 Forbidden02:36
KyleM1403 Forbidden02:36
KyleM1Access was denied to this resource.02:36
KyleM1I'm using vishy's script.02:36
Ryan_Lanethat's on the screen with novarc sourced, right?02:37
KyleM1I don't know which one has novarc sourced.02:37
KyleM1It's on screen #702:37
KyleM1Labelled test.02:37
*** Podilarius has quit IRC02:38
KyleM1Yeah, and on screen #0 (labelled nova) it doesn't have EC2_ACCESS_KEY exported.02:38
KyleM1Someone told me a while back I needed to be running the commands in the "test" screen.02:38
*** ctennis has quit IRC02:39
KyleM1A quick google search suggested that eucalyptus throws a fit when your system clocks are wrong, but mine appears to be correct.02:40
KyleM1(and yes, I just checked nova.sh; I'm running that command in the screen with novarc sourced)>02:41
Ryan_Lanehmm02:42
Ryan_Lanenot sure then. does the api window show anything useful?02:42
KyleM1Ahh...quite possibly:02:43
KyleM1socket.error: [Errno 98] Address already in use02:43
KyleM1(and it's crashed back out to a prompt)02:43
KyleM1objectstore and volume have also crashed back to prompts, but the volume one, at least, was expected (I think...I don't think I have volumes set up yet)02:44
KyleM1The objectstore one crashed with a very similar error:02:45
KyleM1 twisted.internet.error.CannotListenError: Couldn't listen on any:3333: [Errno 98] Address already in use.02:45
*** Li has joined #openstack02:46
*** Ryan_Lane is now known as Ryan_Lane|away02:50
Ryan_Lane|awayyou likely already have instances running02:51
*** Ryan_Lane|away has quit IRC02:51
KyleM1Interesting.02:52
KyleM1How do I kill them?02:52
KyleM1Oh, you're gone.  Oops.02:53
KyleM1Well, thanks for the tip!02:53
*** KyleM1 has left #openstack02:56
*** kashyapc has joined #openstack02:56
*** daleolds has quit IRC02:57
*** gaveen has quit IRC03:00
*** gaveen has joined #openstack03:01
*** rchavik has quit IRC03:18
*** rchavik has joined #openstack03:33
*** Ryan_Lane has joined #openstack04:08
*** abecc has quit IRC04:17
*** dubsquared has joined #openstack04:22
*** dragondm has quit IRC04:56
*** stewart has quit IRC05:03
*** ramkrsna has joined #openstack05:07
*** stewart has joined #openstack05:17
*** zul has quit IRC05:24
*** ctennis has joined #openstack05:34
*** dailyio has joined #openstack05:44
*** f4m8_ is now known as f4m805:49
*** dailyio has left #openstack05:49
*** dailyio has joined #openstack05:53
*** ramkrsna has quit IRC05:53
*** krish has joined #openstack05:54
*** krish has left #openstack05:55
*** iammartian has quit IRC05:55
*** mjmac has quit IRC05:55
*** burris has quit IRC05:55
*** karmabot has quit IRC05:55
*** _0x44 has quit IRC05:55
*** cory__ has quit IRC05:55
*** holoway has quit IRC05:55
*** smoser has quit IRC05:55
*** ctennis has quit IRC05:57
*** iammartian has joined #openstack05:59
*** mjmac has joined #openstack05:59
*** burris has joined #openstack05:59
*** karmabot has joined #openstack05:59
*** _0x44 has joined #openstack05:59
*** cory__ has joined #openstack05:59
*** holoway has joined #openstack05:59
*** smoser has joined #openstack05:59
*** verne.freenode.net sets mode: +v _0x4405:59
*** ramkrsna has joined #openstack06:06
*** smoser has quit IRC06:16
*** aimon has quit IRC06:29
*** aimon has joined #openstack06:29
*** dubsquared has quit IRC06:30
*** dailyio has left #openstack06:32
*** smoser has joined #openstack06:40
*** smoser has quit IRC06:51
*** 45PABJNDX has joined #openstack06:51
*** smoser has joined #openstack06:56
*** GasbaKid has joined #openstack07:04
*** MarkAtwood has quit IRC07:07
*** aimon_ has joined #openstack07:07
*** aimon_ has joined #openstack07:09
*** aimon has quit IRC07:10
*** aimon_ is now known as aimon07:10
*** GasbaKid has quit IRC07:13
*** GasbaKid has joined #openstack07:13
*** Ryan_Lane has quit IRC07:16
*** rchavik has quit IRC07:37
*** almaisan-away is now known as al-maisan08:01
*** brd_from_italy has joined #openstack08:02
*** rchavik has joined #openstack08:06
*** rchavik has quit IRC08:07
*** rchavik has joined #openstack08:08
*** stewart has quit IRC08:09
*** larstobi has joined #openstack08:09
*** Adiantum has quit IRC08:10
*** rchavik_ has joined #openstack08:16
*** rchavik has quit IRC08:17
*** stewart has joined #openstack08:21
*** allsystemsarego has joined #openstack08:23
*** allsystemsarego has joined #openstack08:23
*** calavera has joined #openstack08:24
*** GasbaKid has quit IRC08:25
*** GasbaKid has joined #openstack08:28
*** GasbaKid has quit IRC08:32
*** GasbaKid has joined #openstack08:34
*** ibarrera has joined #openstack08:41
ttxalekibango: ping08:42
*** Li has quit IRC08:43
ttxalekibango: about https://blueprints.launchpad.net/nova/+spec/resource-partitioning08:43
*** rchavik_ has quit IRC08:44
alekibangottx: i know, but can we please meet later, for example 6-8 hours later?08:44
*** rchavik has joined #openstack08:44
alekibangoor start pvtmsging me, i will try to answer08:44
ttxalekibango: sure. I thought EU TZ was better for you ;)08:44
*** kashyapc has quit IRC08:57
*** ramkrsna has quit IRC09:02
larstobihmm…all I changed in the Swift config was to add SSL certificates for auth-server.conf and proxy-server.conf, and now "swift-auth-recreate-accounts -K passwd"  fails with "Recreating accounts failed. (503)" from the client, and "auth-server ERROR Unhandled exception in ReST request: Connection refused" in the syslog09:17
larstobiSwift seems to be working, though, with that exception.09:18
*** rchavik has quit IRC09:20
larstobiboth port 11000 and 8080 are open and listening09:22
larstobiare there any other ports I should be aware of?09:22
*** vivekbabs has quit IRC09:25
*** deshantm_cosi has quit IRC09:28
larstobiI "solved" it by setting the default_cluster_url in auth-server.conf, but now it returns "ERROR attempting to create account https://10.2.2.102:8080/v1/AUTH_asdf: 500 Internal Server Error". It's progress, anyway. :)09:28
*** deshantm has quit IRC09:29
*** zhengli has joined #openstack09:29
*** deshantm has joined #openstack09:41
*** deshantm_cosi has joined #openstack09:41
*** jkakar_ has joined #openstack09:44
*** larstobi has quit IRC09:45
*** jkakar has quit IRC09:45
*** jkakar_ is now known as jkakar09:46
*** dkocher has joined #openstack09:47
*** larstobi has joined #openstack09:51
*** kashyapc has joined #openstack09:54
*** termie has quit IRC10:23
*** termie has joined #openstack10:23
*** dkocher has quit IRC10:26
*** gaveen has quit IRC10:31
*** al-maisan is now known as almaisan-away10:35
*** 45PABJNDX has quit IRC10:50
*** ramkrsna has joined #openstack10:55
*** ramkrsna_ has joined #openstack10:56
*** kashyapc has quit IRC11:01
*** jkakar has quit IRC11:01
*** jkakar has joined #openstack11:04
uvirtbotNew bug: #688508 in nova "xenapi raise Exception rather than NotFound in get_info" [Undecided,New] https://launchpad.net/bugs/68850811:06
*** ramkrsna has quit IRC11:13
zykes-what was the reason for Gundlach leaving ?11:21
chmouelpersonal reason  it seems11:23
*** GasbaKid has quit IRC11:26
*** Suh has joined #openstack11:27
*** GasbaKid has joined #openstack11:28
*** zul has joined #openstack11:45
*** ar1 has quit IRC12:02
*** novahelp has joined #openstack12:03
*** brd_from_italy has quit IRC12:34
*** aimon has quit IRC12:40
*** brd_from_italy has joined #openstack12:50
*** novahelp has quit IRC13:02
uvirtbotNew bug: #688545 in nova "nova-compute (and possibly others) crashes if cannot connect to datastore" [Undecided,New] https://launchpad.net/bugs/68854513:21
*** jaypipes has quit IRC13:28
*** SaadTalaat has joined #openstack13:38
*** SaadTalaat has left #openstack13:38
*** almaisan-away is now known as al-maisan13:49
*** jbaker_ has joined #openstack13:50
*** jimbaker has quit IRC13:50
*** Im0 has joined #openstack14:00
*** rds__ has joined #openstack14:04
alekibangoanyone here knows irc nicks of Shigetoshi Yokoyama, Nobukazu Yoshioka  ?? are they coming here?14:05
*** rds__ has quit IRC14:06
*** fitzdsl has left #openstack14:12
*** calavera has quit IRC14:16
*** krish has joined #openstack14:19
*** aliguori has joined #openstack14:20
Im0many people using openstack in production at the moment?14:24
sorenIt's free software. There's no way to tell.14:25
alekibangolm0 i would think they are mostly betatesting it...  i think it will be more adopted for production after next release14:26
alekibangobut thats only opinon14:26
Im0alekibango: Q1 next year if I read correctly?14:26
alekibangoyes14:26
Im0cool14:27
Im0Just looking to set something up.  Looks really good.14:27
larstobiI'm currently considering using it in production14:27
Im0larstobi: has it been fairly stable for you?14:27
jeroi'm currently considering using it in destruction14:27
*** rds__ has joined #openstack14:28
alekibangoIm0: as i said few days ago - its moving so fast that we need to hold our hair. hats are lost already...14:28
alekibangofast moving duck on steroids14:28
Im0alekibango: :)  exciting14:28
larstobiwell, the things I can get to work are stable, but there are some tings I can't get to work with certain configurations14:28
larstobiI'm only using Swift, btw14:29
alekibangoswift is pretty stable compared to nova14:29
alekibango:))14:29
alekibangoi would think stable release is now too old, and new is still ahead...14:29
*** sirp1 has joined #openstack14:29
Im0I'm still reading the docs... there appear to be a lot of different names to understand. swift/nova/compute/object storage ?14:30
alekibangonova = compute14:30
alekibangoswift = storage14:30
alekibangonova/swift  are names for developers14:30
alekibangoand compute/storage for big customers :)14:30
Im0ahh right, thanks.14:30
*** westmaas has joined #openstack14:31
alekibangonova prolly means: no other viable alternative :)14:31
larstobiI'm think that Swift configuration files aren't very thouroughly documnted, so I often find myself reading the code to figure things out14:31
alekibangolarstobi: i am just now reading nova's code with the same motivation lol14:32
alekibangobut its readable!14:32
alekibangoand clear14:32
larstobiyeah :)14:32
alekibangoIm0: python is your friend14:32
Im0so nova/swift are what the devs are currently expanding on?  compute/object storage are general releases?14:32
alekibangoIm0: nova/swift are internal names14:32
sorenOpenstack compute == Nova14:33
alekibangoyou know, for customers they are hidden behind the cloud14:33
sorenOpenstack storage == Swift14:33
Im0same things then14:33
Im0ok, thanks.14:33
*** jaypipes has joined #openstack14:33
alekibangoluckily customers are not hoodwinked here...14:33
*** mdomsch has joined #openstack14:33
alekibangoits only somewhat cloudy sometimes :)14:33
Im0I heard about this some months ago on floss weekly I think.. looks like a lot has been happening quickly.14:34
larstobiI heard about it at OSCON in July, and started testing it this Monday14:36
larstobia lot is working very well14:36
Im0Nice.14:36
alekibangoIm0: that flosweekly is pretty old now :)14:36
*** Podilarius has joined #openstack14:36
Im0alekibango: yeah, I didn't checkout the site until recently.  Looking to replace our current setup with something a bit more "open"14:37
*** kashyapc has joined #openstack14:37
sorenIm0: What are you running now?14:37
alekibangoIm0: how many nodes?14:37
Im0soren: Virtuozzo and Parallels management system..14:37
Im0alekibango: probably about 10-15 nodes?14:38
sorenIm0: Ok.14:38
alekibangoIm0: interesting. i am on 4 now, plan few 100s soon14:38
alekibangobtw my fai installation is giving me some fruits now :)14:38
alekibangoif there are people interested in automated installation of nova14:39
larstobifai?14:39
Im0alekibango: sounds nice.  Have you got some sort of PXE installer?14:39
alekibangogpxe  iso  - by ipmi hehe14:39
alekibangoiam planning to hack it some more and possibly release it, maybe combine with some puppet or chef, maybe add some mcollective into it also14:40
larstobiI'm guessing http://fai-project.org/14:40
alekibangoyes sorry i got 2 questions at once :D14:40
Im0alekibango: a few 100 nodes.. so, this looks to be a bit different to what we have atm.. of thsoe 100 nodes will some be storage systems and others host virtual servers etc?14:40
larstobihmm, I didn't know about that one. thanks!14:41
alekibangoIm0: mst my hosts will be serving nova14:41
alekibangostorage hosts only to serve for nova14:41
alekibangovia glance14:41
alekibango.. i need cactus release to move forward it seems14:41
alekibangowith live migration etc14:42
creihtlarstobi: if you change the default cluster url you have del the auth.db then recreate your users14:42
Im0alekibango: sounds like I have a bit of a learning curve to experience here... ie glance/cactus all new words to me :)14:42
creihtormanually edit the db to channge the urls for eachuser14:42
Im0alekibango: no live migrations as yet?14:43
larstobicreiht: good morning! okay, I'll try it now14:43
alekibangoIm0: cactus is name of release some 5 months ahead14:43
creihtmorning :)14:43
Im0Hello14:43
alekibangoglance is system of managing image repository for nova...14:43
Im0Ah, cool.14:43
Im0Thanks.14:43
alekibangoIm0: it will not be that hard, documentation now exists. at least some14:44
alekibango:)14:44
creihtthe config settigs are also documented, but i will have to give you the link a little later as i amo14:44
creihterm am on my tablet14:44
larstobicreiht: I have already manually altered the url per user in auth.db to reflect the new cluster ip, but it still gives me 50014:44
creihtyou will also add an ssl=true setting to the auth middleware in proxy-server.conf14:45
larstobiok14:45
creihtif you look at the multi server install dcs, everything is set up with ssl14:46
creihtso that may ao be a good place to look14:46
Im0alekibango: you mentioned live migrations. Is this currently not available with nova/swift?14:47
alekibangounfortunatelly, its not14:47
alekibangobut someone is working on it14:47
alekibangowe all would like to have it14:47
Im0what vm software do you use for the virtual servers? xen/kvm?14:48
alekibangoIm0: yes14:48
alekibangoand more14:48
Im0what is your preference?14:48
alekibangokvm/xen14:48
alekibangokvm first14:48
Im0Yeah, it looks quite nice.14:49
odyiAnyway to make an instance in nova persistant? The root disk isn't destroyed when you destroy the instance and comes back up with the same mac address? Is this something that comes with the object store as a service work?14:50
Im0If I have manually provisioned KVM systems, would they be easy enough to move into nova later?14:51
larstobicreiht: I have followed the multi server document quite closely. I have most things working with SSL now, but the user management isn't working for meg with SSL yet. I get 200 OK from swift-auth-recreate-users, but then "auth-server ERROR attempting to create account https://10.2.2.102:8080/v1/AUTH_asdf: 500 Internal Server Error" and "auth-server recreate_accounts(_, _) = 1 accounts, failures ['AUTH_asdf'] [0.11]" shows up in the syslog, and apparently14:57
alekibangoIm0: yes they will be easy to move in and out14:58
Im0alekibango: Thanks for your help.  Hopefully I'll be a little more clueful shortly. :)15:00
*** skrusty has quit IRC15:00
*** abecc has joined #openstack15:03
*** abecc has joined #openstack15:03
*** ppetraki has joined #openstack15:04
larstobiI'm not quite sure what to make of the Internal Server Error.15:04
*** guigui1 has joined #openstack15:04
*** f4m8 is now known as f4m8_15:04
alekibangolarstobi: you need to make real question if you want us to help15:05
larstobiI'm sorry, I'm trying to ask a real question :)15:07
*** piken_train is now known as piken15:07
larstobiAdding a new user to the existing 'test' account also works fine. Adding a new user to a new account, doesn't work, and "swift-auth-add-user -K mmmsaltyhash  -a test1 tester1 testing" returns "Update failed: 503 Service Unavailable" to the client, and this is logged in syslog: "auth-server ERROR attempting to create account https://10.2.2.102:8080/v1/AUTH_asdf: 500 Internal Server Error", "auth-server FAILED create_user('test1', 'tester1', _, True, False15:07
larstobiso the question is: why does this happen?15:08
*** krish has quit IRC15:08
alekibangoswift, right?15:08
larstobiyeah15:08
alekibangoat least you given us some clues :)15:08
larstobi:)15:08
alekibangobut i am sorry, i dont know swift much15:08
alekibangoi would try to rgrep 'FAILED create_user' in src15:09
*** westmaas has left #openstack15:09
alekibangocheck all logs if there are some more15:09
larstobiI can only find it logs to /var/log/messages15:10
*** skrusty has joined #openstack15:13
*** gondoi has joined #openstack15:14
Im0alekibango: to migrate a vserver I presume you can still do a cold migration?15:14
*** Suh has quit IRC15:28
*** Suh has joined #openstack15:30
*** guigui1 has quit IRC15:31
larstobiI've tracked down what the auth-server sends to the proxy-server that crashes. In 'auth/server.py':L204 the auth-server connects to the proxy-server: 'conn = http_connect(parsed.hostname, port, 'PUT', parsed.path, {'X-Auth-Token': token}, ssl=(parsed.scheme == 'https'))' and the proxy-server responds with 500.15:42
larstobiThis translates to this curl command: http://paste.openstack.org/show/268/15:42
larstobiI haven't been able to figure out what happens on the proxy that makes it fail.15:44
larstobiThe proxy doesn't log anything either15:45
*** hggdh has quit IRC15:49
larstobigotta go now, going away for the week end. contact me on larstobi15:54
larstobi@conduct.no if you think you know the answer :)15:55
larissalarstobi: Error: "conduct.no" is not a valid command.15:55
larstobilarstobi at conduct.no if you think you know the answer :)15:55
larstobibye15:56
*** larstobi has left #openstack15:56
*** JuanPerez has joined #openstack16:03
*** jdarcy has joined #openstack16:03
*** GasbaKid has quit IRC16:03
*** sophiap has joined #openstack16:03
*** GasbaKid has joined #openstack16:04
*** mdomsch has quit IRC16:05
*** dirakx has left #openstack16:07
*** daleolds has joined #openstack16:07
*** dfg_ has joined #openstack16:10
*** MarkAtwood has joined #openstack16:12
*** johnpur has joined #openstack16:12
*** ChanServ sets mode: +v johnpur16:12
*** hggdh has joined #openstack16:20
creihtdoh... just missed larstobi :/16:21
*** dirakx has joined #openstack16:29
*** kashyapc has quit IRC16:34
*** iammartian has quit IRC16:38
*** creiht has quit IRC16:46
*** creiht has joined #openstack16:47
*** ChanServ sets mode: +v creiht16:47
*** jimbaker has joined #openstack16:49
*** jbaker_ has quit IRC16:49
*** deshantm_ has joined #openstack16:54
*** deshantm_ has quit IRC16:55
*** deshantm has quit IRC16:55
*** deshantm has joined #openstack16:55
*** hadrian has joined #openstack17:00
*** Im0 is now known as Im0Away17:00
hadrianhi, any idea where br100 gets configured in nova?17:01
hadriani.e. where the 10.0.0.1 ip comes from?17:01
chmouelhadrian: nova-network?17:01
hadrianchmouel: thanks, but that i don't see where the config is stored (db maybe?)17:02
hadrianor any way to change that17:02
chmouelhumm not sure sorry and i don't have test machine near by to look at17:04
* ttx endofweeks17:04
hadrianchmouel: looked everywhere, /etc, /lib17:04
chmouelsqlite /var/lib/nova/nova.sqlite .dump ?17:05
*** KyleM1 has joined #openstack17:06
*** zhengli has quit IRC17:07
hadrianchmouel: not familiar with sqlite, trying to figure it out17:08
hadriani get: Unable to open database "/var/lib/nova/nova.sqlite": file is encrypted or is not a database17:09
hadrianfor now17:09
KyleM1From what I know about sqlite it stores the DB in that flat file; if you're just trying to get set up and don't have anything important in it yet you could probably just delete it and try again.  I think nova creates it fresh when it needs to.17:10
chmouelapt-get install sqlite317:10
chmouelsqlite3 instead of sqlite17:10
KyleM1Or, of course, you could get a real answer :-)17:10
KyleM1Thanks chmouel17:10
*** ibarrera has quit IRC17:11
*** dendrobates is now known as dendro-afk17:12
hadrianchmouel: that worked, one sec to pipe the out to a file17:14
hadrianchmouel: found it, thx17:15
chmouelhadrian: sweet17:16
hadriannow the question is how to manage the networks, there are some 6 six of them over there17:16
*** dragondm has joined #openstack17:16
hadriansome are clearly nova-manage network create shoot in the dark attempts to fix the problem :(17:17
KyleM1Hey, I was talking with Ryan_Lane last night about a problem I'm getting.  I run nova from vishy's script (so I have all my workers in different threads, etc.) but when I try something like `euca-describe-instances` I get "EC2ResponseError: 403 Forbidden".  Checking out my workers I see that the api and objectstore workers both crash out on launch with a socket error [Errno 98] Address already in use.17:17
KyleM1Ryan suggested that it looks like I already have instances running, but I've been unable to find them to terminate them if that is the case.17:17
KyleM1I've tried using `nova.sh terminate` and `nova.sh scrub` (although scrub gave me an awk error so I had to run the virsh command manually).17:19
KyleM1Neither found any instances to terminate.17:19
chmoueldid you try to match the pid and see if they are running from there?17:19
KyleM1Match the pid?17:20
*** ccustine has joined #openstack17:20
*** maplebed has joined #openstack17:21
hadriankylem1: on what os/version do you use nova.sh?17:21
KyleM1This is on Debian Squeeze.17:21
*** dubsquared has joined #openstack17:21
*** iammartian has joined #openstack17:22
*** jdurgin has joined #openstack17:25
chmouelin the nova directory can you do a ls *pid17:25
KyleM1Nothing in there.17:26
chmoueli dont have the novascript maybe it's in the nova source directory17:26
KyleM1And `virsh list` gives me an empty list.17:26
KyleM1I've seen the pid files while I had it running; they show up in the directory I run the script from.17:27
chmouelyou are running on bare hardware ?17:31
KyleM1I believe so, yes.17:31
chmouelis your CPU support kvm ?17:33
*** daleolds has quit IRC17:33
*** brd_from_italy has quit IRC17:33
chmouelsee with the command kvm-ok17:33
KyleM1I don't know.  How would...Ah...let me check.17:33
KyleM1# kvm-ok17:34
KyleM1-bash: kvm-ok: command not found17:34
jaypipesdang adding i18n markers is tedious... grr. 35 files done...dozens to go...17:34
KyleM1should I run it in the shell or somewhere else?17:34
*** hazmat has quit IRC17:35
*** dubsquared has quit IRC17:37
*** gondoi has quit IRC17:38
*** zhengli has joined #openstack17:38
jbryceif you don't have kvm-ok try doing  egrep --color "(vmx|svm)" /proc/cpuinfo17:39
KyleM1Okay.  I'm actually rebooting right now, so I'll try that when it comes back up.17:39
chmouelKyleM1: yeah you run it in the shell.. while you are at reboot you probably want to dig in your bios making sure you have the virtualization option enabled..17:41
*** daleolds has joined #openstack17:41
*** dendro-afk is now known as dendrobates17:45
*** hazmat has joined #openstack17:47
*** KyleM1 has left #openstack17:51
*** johnpur has quit IRC17:56
*** Podilarius has quit IRC17:56
*** joearnold has joined #openstack18:03
*** miclorb has joined #openstack18:04
*** sophiap has quit IRC18:05
*** ppetraki has quit IRC18:07
*** Ryan_Lane has joined #openstack18:23
*** BK_man has joined #openstack18:24
BK_manhi all18:25
*** atobey_ has quit IRC18:29
*** al-maisan is now known as almaisan-away18:35
*** iRTermite has quit IRC18:47
*** iRTermite has joined #openstack18:48
*** irahgel1 has left #openstack18:53
vishyKyleM1: sounds like you have another copy of the workers running deamonized or in another screen18:55
vishytry ps ax | grep nova-18:56
*** brd_from_italy has joined #openstack19:06
*** dirakx has quit IRC19:09
*** jaypipes has quit IRC19:09
*** brd_from_italy has quit IRC19:11
*** sirp1 has quit IRC19:21
*** BK_man has quit IRC19:24
*** jc_smith has joined #openstack19:24
*** sophiap has joined #openstack19:25
*** davidX- has quit IRC19:29
*** david has joined #openstack19:31
*** david is now known as davidX-19:31
*** burris has quit IRC19:32
*** jimbaker has quit IRC19:36
*** sirp1 has joined #openstack19:37
*** Podilarius has joined #openstack19:41
*** BK_man has joined #openstack19:49
*** lcfseth has joined #openstack19:53
lcfsethHello World!19:54
BK_manlcfseth: hi19:55
lcfseth:) hi19:55
*** Suh has left #openstack19:55
BK_manppl, how much spave should I have for a test instance from cloudfiles ?19:55
* BK_man have not so much space in /var/lib/nova/instances19:56
*** maplebed has quit IRC19:56
* BK_man and run out free space on dd bla-bla execution19:56
creihtbk_man: yeah if you want more space, then you will need to make the dd larger19:57
creihtusable space will be that value / 3 (roughly)19:58
BK_mancreiht: I have now 3.5G and this is not OK19:58
BK_mancreiht: what should I adjust in (image, code, conf files) so dd use less space?19:59
creihtjust use a smaller nuber when you do the dd20:00
creihtthat will make for a smaller file system20:00
BK_mancreiht: I'm using test images from cloudfiles - mentioned on wiki. What should I change in that image?20:04
BK_manI can't find any configuration option for abount disk size, memory size - etc for VMs20:04
*** MarkAtwood has quit IRC20:05
*** ppetraki has joined #openstack20:05
*** jc_smith has quit IRC20:08
creihtBK_man: oh... sorry, thought you were talking about swift20:09
BK_mancreiht: no, it's all about nova20:09
* creiht passes the buck to the nova guys20:09
*** gondoi has joined #openstack20:12
*** jc_smith has joined #openstack20:19
*** BK_man has quit IRC20:20
*** jc_smith has quit IRC20:22
*** BK_man has joined #openstack20:27
*** skrusty has quit IRC20:28
*** sirp1 has quit IRC20:28
*** rds__ has quit IRC20:30
alekibangoIm0: for now i install all new hosts and guests on fresh empty disks :)  But  this should answer you: https://blueprints.launchpad.net/nova/?searchtext=migration20:33
*** skrusty has joined #openstack20:33
*** skrusty has quit IRC20:38
*** GasbaKid has quit IRC20:39
vishyBK_man: use -t m1.tiny20:39
*** GasbaKid has joined #openstack20:39
*** skrusty has joined #openstack20:39
vishythat way it won't try to generate a multi-gig image from the base20:41
BK_manvishy: to which command?20:41
vishyeuca-run-instances -t m1.tiny20:41
BK_manvishy: here is my test case (a PLAN :-) http://paste.openstack.org/show/270/20:41
BK_manvishy: is it ok from you point of view?20:42
alekibangoBK_man: using perl for everything is evil20:43
alekibangoyou know, we have cut, sed, awk, grep20:43
BK_manalekibango: I know, but I'm old-style perlish guy :)20:44
alekibangoBK_man: i presonally preffer "$(something)"20:44
alekibangoBK_man: yeah, i had my portion of perl20:44
BK_manalekibango: sed have different regex syntax - it pisses me off each time20:44
vishyyes you can use -t m1.tiny in the run command20:44
BK_manvishy: thanks, will try20:45
alekibangoBK_man: still, try cut20:45
alekibangoits extremely easy to split line using cut20:45
* BK_man hope that having 3.5G is enough20:45
BK_manalekibango: thanks, will refactor later ;)20:45
alekibangoand btw, i love "$(run)" syntax  :) but i know perl guys will forever be in `here`20:46
BK_manI still have no solution for openssl-1.0.0 bug... Just workaroung in openssl.cnf20:47
alekibangoBK_man: you do not check values, sometimes you might get error like i did20:47
alekibangoi had to cd images/aki-lucid/20:47
BK_mankhm... libvirtError: unknown OS type hvm20:47
alekibangoouch :)20:47
alekibangoBK_man: do you ahve hw virt support?20:47
alekibangoif no, use uml or qemu20:48
alekibangoor for best result, xen20:48
BK_mana think so. This is brand new Cisco UCS blade server20:48
BK_mans/a/I/ :)20:48
vishythat is either: no hw virt, virt turned off, or bad perms to /dev/kvm20:48
alekibangoi would bet on permissions20:48
alekibangoBK_man: which os?20:48
*** sirp1 has joined #openstack20:48
BK_manalekibango: you know my answer and my effort. P.S. RHEL620:49
BK_manmy HW have Intel Xeon E5620 cpus20:49
alekibangoah, i forgot that if you said it before...20:49
BK_manlet me check perms20:49
alekibangoyou might also want to modprobe kvm20:49
alekibangomaybe20:49
BK_manalekibango: yep. https://github.com/abrindeyev/openstack-nova-rhel620:49
alekibango:) i am working on debian branch20:50
alekibangobtw i cannot find good documentation of nova-volume20:50
BK_manalekibango: there is only one Linux distro for enterprise guys here in US - RHEL :-(20:50
*** skrusty has quit IRC20:50
alekibangoand that distro is made in czech republic, in my town20:51
alekibango:)20:51
*** skrusty has joined #openstack20:51
BK_manouch! in dmesg: kvm: disabled by bios20:51
*** Abd4llA has joined #openstack20:51
alekibangolol classic one20:52
alekibangohi Abd4lla20:52
BK_manis that meaning that Intel VM is disabled?20:52
alekibangoBK_man: reboot, go to bios20:52
alekibangoenable20:52
Abd4llAhi alekibango20:52
alekibangoAbd4llA: what is new?20:52
BK_manalekibango: let me figure out how to do that on stateless cisco blade...20:52
alekibangoso you are on virtual host?20:52
alekibangoor what is blade20:53
alekibangostateless? :) really?20:53
hadriandoes nova.sh run as expected on ubuntu 10.10 ?20:53
alekibangoAbd4llA: docs are growin fast now :)20:53
hadrianinstructions only mention 10.0420:53
alekibangohadrian: the same20:53
hadrianalekibango: thx20:54
vishyhadrian: as far as I know it does20:54
alekibango(i hope)20:54
hadrianvishy: thx too20:54
hadrianwill know in a few mins...20:54
alekibangohadrian: whats teh human readable name for 10.10? maverick?20:55
hadrianalekibango: ubuntu maverick meerkat20:56
hadrianbut 10.10 is more readable to me :)20:56
alekibango:D  i fail to remember those numbers20:56
hadrianalekibango: first one is the year: 2010, second is the month 10=oct20:59
alekibangoaaah20:59
alekibangothanks man20:59
hadrianubuntu releases like clockwork in apr and oct, so it's either 04 or 10 (second number)20:59
*** dirakx has joined #openstack21:00
alekibangohadrian: i suspected something fishy about those numbers... thanks21:00
hadriannp21:00
hadrianone more q: must nova.sh be executed as root?21:01
*** dirakx has quit IRC21:01
termievishy: what else do i need to do to verify that the eventlet branch is reliable?21:04
*** jimbaker has joined #openstack21:07
uvirtbotNew bug: #688734 in nova "registering an image doesn't use kernel and ramdisk" [Medium,In progress] https://launchpad.net/bugs/68873421:07
*** skrusty has quit IRC21:12
Abd4llAalekibango: yea I noticed, that's really great21:12
Abd4llAalekibango: and I c you really took off :)21:12
alekibangonova-volume docs still missing21:13
alekibangoAbd4llA: i am not flying yet21:13
vishyhadrian: yes21:16
vishyor sudo'ed21:16
hadrianvishy: yeah, i noticed that install needs root privileges for aptitude, branch probably not, but doesn't hurt21:17
hadrianthanks21:17
*** dirakx has joined #openstack21:17
*** skrusty has joined #openstack21:17
vishynova workers need a lot of sudo perms, so unless you want to hack up sudoers, best to just sudo nova.sh run21:17
*** skrusty has quit IRC21:27
BK_manwhat should be permissions for /dev/kvm ?21:31
* BK_man finally enabled Intel VT on my blade21:31
*** dirakx has quit IRC21:31
*** Messanger has joined #openstack21:34
Messangerworld leaders setting you all up right now to get microchipped by RFID microchips with identity/healthcare/credit information into their new world order why they are folding the economy , their solution to the problem they are causing is the RFID microchip  http://www.scribd.com/doc/44997148/The-Mark-of-the-Beast    Romans 10:13 all who call upon the name of the Lord shall be saved.21:34
*** Messanger has left #openstack21:34
*** hggdh has quit IRC21:35
notmynameseems like they will need a good storage system for all that data21:36
*** hggdh has joined #openstack21:40
*** sirp1 has quit IRC21:42
dendrobatesha21:43
*** sirp1 has joined #openstack21:43
*** gondoi has left #openstack21:44
*** skrusty has joined #openstack21:44
sorenvishy: Can you explain the symptoms of  bug 688734 ?21:45
uvirtbotLaunchpad bug 688734 in nova "registering an image doesn't use kernel and ramdisk" [Medium,In progress] https://launchpad.net/bugs/68873421:45
vishysure21:45
vishyeuca-register doesn't pull kernel and ramdisk from the manifest21:45
sorenI'm reasonably sure it does.21:46
sorenI fixed that in r337.21:46
vishyhmm21:46
sorenI'm succesfully running euca-run-instances with just an ami.21:46
sorenNo kernel or ramdisk specified.21:46
vishyafter registering a new one?21:46
sorenEh?21:47
vishyis it magically being done in some other piece of code?21:47
vishythe code path in register image21:47
vishydoes not set kernelId and ramdiskId21:47
vishyin the image.json21:47
vishy(afaict)21:47
sorenCan you perhaps glance at r337?21:49
*** dkocher has joined #openstack21:49
*** jdarcy_ has joined #openstack21:50
*** jdarcy has quit IRC21:50
*** jdarcy_ is now known as jdarcy21:51
*** lcfseth has quit IRC21:52
vishydamn internet21:52
vishysoren: it sets it properly in the add() method but register_aws_image is used21:53
sorenvishy: That's what I fixed.21:53
sorenvishy: Ok, so looking at trunk now.21:54
sorennova/objectstore/image.py21:54
*** jc_smith has joined #openstack21:54
sorenline 21021:54
sorenThe kernel_id there is from line 189.21:54
soren189 extracts the kernel id from the manifest.21:54
sorenLine 210 shoves it into the machine image's info.json.21:54
vishysoren: ah21:55
sorenI'm not saying there's nothing left to fix. I just wonder what it is :)21:55
vishysoren: you are correct, I was porting it from deploy21:55
vishydidn't notice the change21:55
sorenr337 is ancient.21:55
*** jaypipes has joined #openstack21:55
vishyarchitecture isn't set21:55
BK_manvishy: crw-rw---- 1 root root /dev/kvm - is it ok permissions?21:56
sorenWhen did you branch?21:56
vishya long time ago21:56
vishy:)21:56
sorenvishy: No kidding :)21:56
* BK_man got kvm working but error still the same - libvirtError: unknown OS type hvm21:56
sorenvishy: r337 is from Oct 11th.21:56
sorenBK_man: pastebin the output of "virsh capabilities" (as the use that runs nova-compute)21:57
sorenas the *user* that runs nova-compute.21:57
vishysomewhere around 30721:58
sorenvishy: Right, architecture is missing. Good point. I've completely stopped considering 32 bit environments. :(21:58
vishy:) ok i'll mark the bug invalid21:58
vishyand delete my branch21:58
sorenvishy: Or adjust it to mention architecture.21:58
BK_mansoren: http://paste.openstack.org/show/271/21:58
sorenBK_man: Err... Is qemu or kvm installed at all?21:59
*** miclorb has quit IRC21:59
BK_mansoren: libvirtd is running. how to check?21:59
sorenBK_man: Don't know. Which OS?21:59
sorenBK_man: Then I still don't know :) rpm -qXXXX where XXXX is something cleverer thatn XXXX.22:00
sorenWhoops.22:00
sorenWow, weird.22:00
BK_mansoren: how to run under KVM and not with Qemu?22:00
sorenI switched windows a couple of times, could have sworn I saw you say Fedora.22:00
* BK_man going to luch...22:01
*** Cybodog has quit IRC22:01
sorenBK_man: Which OS are you running?22:01
*** abecc has quit IRC22:01
vishysoren: ok i'll change to arch.  Do you mind the different code, or should I change it to use the same style you were using?22:03
sorenvishy: I haven't actually looked at your patch. I started by looking at the bug and didn't understand the problem it was meant to solve.22:05
sorenLet me have a look.22:05
*** rds__ has joined #openstack22:05
sorenIt's what Friday night is for, isn't it?22:06
sorenvishy: Oh, yours is fine, too. Same approach, just different ordering.22:06
sorenvishy: If you fix up the unit tests to check for architecture, I'm super happy.22:07
*** dendrobates is now known as dendro-afk22:08
*** burris has joined #openstack22:11
*** zhengli has quit IRC22:14
*** allsystemsarego has quit IRC22:17
vishyyay internet22:25
*** sophiap has quit IRC22:25
*** jfluhmann has quit IRC22:31
*** jdarcy has quit IRC22:31
*** jimbaker has quit IRC22:34
*** jaypipes has quit IRC22:35
*** bwalk_57 has joined #openstack22:37
*** jimbaker has joined #openstack22:39
*** jakedahn has joined #openstack22:41
*** jakedahn has quit IRC22:41
*** sophiap has joined #openstack22:43
bwalk_57hey all, I just installed nova on Ubuntu 10.04, and have it running, but after opening port 22 and starting an instance, any attempt to ssh into it times out22:47
bwalk_57and euca-get-console-output only returns the instance id with a timestamp22:47
bwalk_57has anyone encountered this recently; I just downloaded a branch of code yesterday22:48
*** ivan has quit IRC22:50
*** Cybodog has joined #openstack22:51
vishydoesn't sound like your instance is actually running successfully22:52
vishylook in the instances/instance-xxxxxxx/ folder22:52
vishyif you have no console output22:53
*** filler has quit IRC22:54
bwalk_57vishy: yep, the console.log file is empty22:54
*** filler has joined #openstack22:54
vishyif you describe instances does your instance show as running?22:55
bwalk_57yes22:55
vishyhow about virsh list22:55
bwalk_57yes, also shows as running22:55
*** ivan has joined #openstack22:58
*** dendro-afk is now known as dendrobates22:59
vishyfile instance-xxxx/*23:00
vishyare you using uml / kvm / or qemu?23:00
bwalk_57qemu23:03
termieeday, vishy: SOMEBODY SET US UP THE BOMB23:04
vishytermie: WHAT YOU SAY?23:04
*** ppetraki has quit IRC23:06
*** jimbaker has quit IRC23:08
*** Podilarius has left #openstack23:11
*** aimon has joined #openstack23:19
*** dkocher has quit IRC23:19
*** dfg_ has quit IRC23:21
*** maplebed has joined #openstack23:30
*** EdwinGrubbs has quit IRC23:32
*** miclorb has joined #openstack23:35
hadrianthe default content of the networks table is:23:36
hadrian2010-12-10 21:21:34.097514|||0|1|0|10.0.0.0/27|255.255.255.224|br100|10.0.0.1|10.0.0.31||100||1000|10.0.0.2|10.0.0.3||23:36
hadrianis there a tool that changes the cidr and ip settings?23:36
hadrianor just edit the database manually?23:36
hadriannova-manage network create only adds to this table23:38
hadrianfrom what i could gather23:38
*** daleolds has quit IRC23:39
hadrianhmm, looks like this entry was added by nova.sh run23:41
*** littleidea has joined #openstack23:42
*** GasbaKid has quit IRC23:43
*** sophiap has quit IRC23:47
*** sophiap has joined #openstack23:47
*** dendrobates is now known as dendro-afk23:53
*** daleolds has joined #openstack23:54
*** dendro-afk is now known as dendrobates23:54
hadriannova.sh line 174 invokes on vblade-persist which does not exist on my system23:59
*** JuanPerez has quit IRC23:59
hadrianrun.sh install should probably make sure vblade vblade-persist aoetools are installed23:59

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!