Thursday, 2011-06-23

*** ton_katsu has joined #openstack00:03
*** kashyap has quit IRC00:04
*** kashyap has joined #openstack00:05
*** mrjazzcat has quit IRC00:06
*** hadrian has joined #openstack00:08
*** RobertLaptop has quit IRC00:08
*** RobertLaptop has joined #openstack00:10
*** cp16net has joined #openstack00:17
*** PeteDaGuru1 has joined #openstack00:19
*** PeteDaGuru has quit IRC00:19
*** cp16net has quit IRC00:20
*** PW__ has quit IRC00:21
*** obino1 has quit IRC00:22
*** ton_katsu has quit IRC00:23
*** Xenith has quit IRC00:25
*** Xenith has joined #openstack00:25
*** rchavik has joined #openstack00:26
*** rchavik has quit IRC00:27
*** rchavik has joined #openstack00:28
*** kashyap has quit IRC00:34
*** ton_katsu has joined #openstack00:35
*** yui0 has joined #openstack00:46
*** heckj has quit IRC00:47
*** rminnear has quit IRC00:47
*** jdurgin has quit IRC00:48
*** mszilagyi has quit IRC00:49
*** dragondm has quit IRC00:49
*** kashyap has joined #openstack00:54
*** clauden has joined #openstack00:55
*** victor_lowther has joined #openstack00:55
*** clauden has quit IRC00:59
*** clauden___ has joined #openstack00:59
*** clauden___ has quit IRC01:02
uvirtbotNew bug: #800947 in glance "Incorrect content-type returned on GET '/v1/images/<id>'" [Undecided,New] https://launchpad.net/bugs/80094701:02
*** clauden has joined #openstack01:02
*** vernhart has joined #openstack01:08
*** cereal_bars has joined #openstack01:16
*** samba has left #openstack01:30
*** clauden has quit IRC01:31
*** gaveen has joined #openstack01:32
*** gaveen has joined #openstack01:32
*** vernhart has quit IRC01:36
*** obino has joined #openstack01:44
*** osier has joined #openstack01:49
*** jkoelker has quit IRC01:53
*** PeteDaGuru1 has quit IRC01:54
*** GeoDud has quit IRC01:55
*** dmshelton has quit IRC01:58
*** wilmoore has quit IRC02:01
*** hadrian has quit IRC02:14
*** ton_katsu has quit IRC02:19
*** fysa has joined #openstack02:23
*** tinton has joined #openstack02:23
tintonHi i started dual node openstack configuration but stucked in some point in network configuratrtion02:24
tintoncan some one help02:24
*** adjohn has joined #openstack02:27
*** andy-hk has quit IRC02:31
HugoKuo_tinton , may I ?02:35
*** andy-hk has joined #openstack02:39
*** andy-hk has quit IRC02:39
*** andy-hk has joined #openstack02:40
*** wilmoore has joined #openstack02:40
*** msivanes has joined #openstack02:42
*** msivanes has left #openstack02:42
tinton#HugoKuo_ thanks Hugo02:51
tintoni have two nodes02:51
tintonone is compute and rest of the all are in seperate node02:52
tintonwhen i was configuring i put 10.100.3.150 into eth0 which became the management IP02:53
tintonthen i provided eth1 as flat_interface. 172.10.0.0/16 as fixed ip range and some real ip range 203.140.40.0/29 as floating range.02:54
*** ton_katsu has joined #openstack02:54
tintonwhen instance is running i can see its assigned 172.10.0.102:54
tintonsorry 172.10.0.202:55
tintoncontroller node dummy interface having 172.10.0.1 ip02:55
tintonbut i cant ping to the guses OS running on compute node02:55
tintonThis is my problem Thanks in advance if you can help me02:56
HugoKuo_1. what's the network mode of your nova03:00
HugoKuo_2. plz show me your console out put of guest03:01
HugoKuo_btw , the instance should not be assigned 172.10.0.1 , this ip will be use for flat_network bridge eth1 ---> br100 on controller node for nova-network service.03:02
HugoKuo_To provision dhcp response03:03
*** mdomsch has joined #openstack03:03
tintonYes Hugo its 172.10.0.203:03
tinton172.10.0.1 is on controller node03:03
HugoKuo_ok03:03
HugoKuo_how's your console uot put03:03
HugoKuo_and which network mode in your deployment ?03:03
tintonconsole out put mean the "virsh list" ?03:04
HugoKuo_no03:04
HugoKuo_euca-get-consolexxxxxxxx03:04
*** cp16net has joined #openstack03:05
tintongive me a minit i am trying to connet03:06
HugoKuo_#euca-describe-instance       will show you instance ID  "i-00000101"   RESERVATION     r-rowssl3s      HUGO    hugo12303:08
HugoKuo_INSTANCE        i-00000101      ami-00000011    172.16.2.133    172.16.2.133    running hugok (HUGO, ubuntu3)   0               m1.tiny 2011-06-21T05:54:42Z    nova03:08
HugoKuo_# euca-get-console-output i-000001103:09
HugoKuo_you'll get guest console out put03:09
*** msivanes has joined #openstack03:11
tintonyes hugo its showing as it is03:12
tintonip is 172.16.0.203:12
tintonI am really sorry now i am at home but i set up the configuration in server room but unable to connect03:13
tintoni can conntct to controller node and kvm show   1 instance-00000001    running03:14
*** msivanes has left #openstack03:15
*** cp16net has quit IRC03:18
HugoKuo_did you install euca2ools on controller node ?03:19
tintonHi Hugo i have the output of Hybridfox do you need any information from it03:19
*** cereal_bars has quit IRC03:19
tintonyes03:19
HugoKuo_it's ok ... same stuff03:19
tintonit shows private and public instance 172.16.0.203:20
tintonstate running03:20
tintongroup default03:20
tintontype m1.small03:21
tintonalso Instance ID, AMI also there03:22
tintonnova-network.conf  # This is an automatically generated file by stackops # Please DO NOT CHANGE the parameters manually --dhcpbridge_flagfile=/etc/nova/nova-network.conf --verbose=True --rabbit_host=10.100.3.150 --fixed_range=172.16.0.0/16 --dhcpbridge=/var/lib/nova/bin/nova-dhcpbridge --sql_connection=mysql://root:nova@10.100.3.150:3306/nova --ec2_dmz_host=10.100.3.150 --lock_path=/tmp --network_size=64 --flat_interface=eth1 --aut03:23
tintonThis is the nova-network.conf file03:24
tinton--dhcpbridge_flagfile=/etc/nova/nova-network.conf --verbose=True --rabbit_host=10.100.3.150 --fixed_range=172.16.0.0/16 --dhcpbridge=/var/lib/nova/bin/nova-dhcpbridge --sql_connection=mysql://root:nova@10.100.3.150:3306/nova --ec2_dmz_host=10.100.3.150 --lock_path=/tmp --network_size=64 --flat_interface=eth1 --auth_driver=nova.auth.dbdriver.DbDriver --network_manager=nova.network.manager.FlatDHCPManager --ec2_host=10.100.3.150 --s303:25
tinton--s3_host=10.100.3.150 --public_interface=eth0 --use_project_ca=True --nodaemon=True03:26
*** maplebed has quit IRC03:26
*** Funnnny has joined #openstack03:27
tintonHi Hugo thank for the try now i am going to the office and i can connect to the server romm, I ll send you euca tools informations03:38
tintonthaks a lot03:38
*** joearnold has joined #openstack03:40
*** AimanA is now known as HouseAway03:47
*** joearnold has quit IRC03:52
*** cp16net has joined #openstack04:02
*** mikhail has joined #openstack04:04
*** mikhail has quit IRC04:09
*** joearnold has joined #openstack04:10
*** ewindisch has joined #openstack04:15
*** andy-hk has quit IRC04:18
*** mikhail has joined #openstack04:21
*** ryker has quit IRC04:22
*** andy-hk has joined #openstack04:24
*** medberry is now known as med_out04:24
*** heckj has joined #openstack04:28
*** PW__ has joined #openstack04:29
*** tinton has quit IRC04:33
*** joearnold has quit IRC04:34
*** shang_ has quit IRC04:42
*** shang has quit IRC04:42
*** heckj has quit IRC04:57
*** cbeck has quit IRC04:58
*** cbeck has joined #openstack04:59
*** victor_lowther has quit IRC05:00
*** mikhail has quit IRC05:00
*** ton_katsu has quit IRC05:04
*** kennethkalmer has joined #openstack05:06
*** ton_katsu has joined #openstack05:06
*** ton_katsu has joined #openstack05:07
*** ton_katsu has quit IRC05:08
*** fysa has quit IRC05:10
*** PW__ has quit IRC05:11
*** mikhail has joined #openstack05:13
*** gregp76 has joined #openstack05:14
*** gregp76 has left #openstack05:14
*** mdomsch has quit IRC05:16
*** arun has quit IRC05:18
*** shang has joined #openstack05:18
*** cloudgroups has joined #openstack05:22
*** cloudgroups has left #openstack05:22
*** Binbin has joined #openstack05:27
*** mikhail has quit IRC05:31
*** joearnold has joined #openstack05:39
*** vernhart has joined #openstack05:52
*** cp16net has quit IRC06:05
*** kennethkalmer has quit IRC06:05
*** ton_katsu has joined #openstack06:22
*** worstadmin has joined #openstack06:23
*** andy-hk has quit IRC06:28
*** andy-hk has joined #openstack06:30
*** rsingh has joined #openstack06:31
*** nerens has joined #openstack06:34
*** joearnold has quit IRC06:39
rsinghHello All, I am a python developer and new to openstack. I am looking forward to extend openstack webservices. I followed http://wiki.openstack.org/PythonDevelopmentEnvironment  but it doesn't explain how to extend or write new Webservice for openstack. Is there any link which can help me in kickoff06:39
*** wariola has joined #openstack06:46
*** ton_katsu has quit IRC06:52
*** jedi4ever has joined #openstack07:03
*** CloudChris has joined #openstack07:03
*** wariola_ has joined #openstack07:06
*** mikhail has joined #openstack07:09
*** wariola has quit IRC07:09
*** reidrac has joined #openstack07:11
*** JKERZN has quit IRC07:16
*** guigui has joined #openstack07:17
*** guigui has left #openstack07:17
*** JKERZN has joined #openstack07:19
*** CloudChris has left #openstack07:23
*** JKERZN has quit IRC07:25
*** JKERZN has joined #openstack07:25
*** mikhail has quit IRC07:29
*** ziyadb has joined #openstack07:33
*** wariola_ has quit IRC07:33
HugoKuo_does nova-obj /glance / nova-api must on same host ?07:34
*** wariola_ has joined #openstack07:34
*** CloudChris has joined #openstack07:40
*** CloudChris has left #openstack07:40
*** rsingh has quit IRC07:46
*** nacx has joined #openstack07:47
*** p0 has joined #openstack07:48
*** nRy has quit IRC07:51
*** adjohn has quit IRC08:09
*** cloudgroups has joined #openstack08:12
*** Binbin has quit IRC08:12
*** cloudgroups has left #openstack08:12
*** johnpur has quit IRC08:19
*** rsingh has joined #openstack08:21
rsinghHello All, I am a python developer and new to openstack. I am looking forward to extend openstack webservices. I followed http://wiki.openstack.org/PythonDevelopmentEnvironment  but it doesn't explain how to extend or write new Webservice for openstack. Is there any link which can help me in kickoff08:22
*** _william_ has joined #openstack08:22
*** _william_ has quit IRC08:25
*** dobber has joined #openstack08:27
*** daysmen has joined #openstack08:30
*** kashyap has quit IRC08:34
*** deepakn has joined #openstack08:36
*** 17WAAN603 has joined #openstack08:38
*** tudamp has joined #openstack08:40
*** worstadmin has quit IRC08:41
*** worstadmin has joined #openstack08:41
*** ziyadb has quit IRC08:41
*** ziyadb has joined #openstack08:42
*** ziyadb has joined #openstack08:42
*** chemikadze has joined #openstack08:47
*** jedi4ever has quit IRC08:48
*** ewindisch has quit IRC08:50
*** ewindisch has joined #openstack08:51
*** Linex has quit IRC08:53
*** Capashen has joined #openstack08:53
*** Linex has joined #openstack08:54
*** Linex has quit IRC08:55
*** Linex has joined #openstack08:55
doudeHi all, someone use the inject file with OSAPI with XenSever driver ?09:05
doudeI get try this command 'nova boot --image 65 --flavor 1 --file /tmp=tmp.sh "Test VHD image"' but I get this error: http://paste.openstack.org/show/1734/09:05
ziyadbreading the nova cactus admin guide, I think this is a typo, "A ‘volume’ is a detachable block storage device. You can think of it as a USB hard drive. It can only be attached to one instance at a time, so it does not work like a SAN." it should be NAS instead of SAN. Can anyone confirm?09:05
worstadminA NAS can be attached to more than one device as well09:07
*** kashyap has joined #openstack09:07
bonzayziyadb: no, a nas _can_ be attached to more than one instance (it normally has some sort of service that manages the connections)09:07
*** zobel has left #openstack09:08
*** allsystemsarego has joined #openstack09:12
*** tudamp has quit IRC09:15
*** miclorb has quit IRC09:16
*** RoAkSoAx has joined #openstack09:19
*** irahgel has joined #openstack09:24
*** winston-d has quit IRC09:29
*** openpercept_ has joined #openstack09:31
*** wilmoore has quit IRC09:31
*** jedi4ever has joined #openstack09:47
*** ton_katsu has joined #openstack09:53
*** shang has quit IRC09:56
ziyadbI'm confused. You create LUNs on SANs and mount them to instances as see fit, so it's not "shared" i.e. the volumes are dedicated for the instances while operating. But when doing a NAS (I believe NAS is file storage) so multiple users can access it simulataneously.09:57
*** shang has joined #openstack10:00
*** reidrac has left #openstack10:03
*** reidrac has joined #openstack10:03
*** shang has quit IRC10:05
*** ziyadb_ has joined #openstack10:08
*** brendan_ has joined #openstack10:09
*** brendan_ is now known as Guest6596210:09
*** ziyadb has quit IRC10:10
*** andy-hk has quit IRC10:13
*** Guest65962 has quit IRC10:15
*** brendan1495 has joined #openstack10:16
*** Funnnny has quit IRC10:17
*** Funnnny has joined #openstack10:18
brendan1495hi all just a quick question, I am attempting to set up openstack for a university project, do I need 2 ethernet interfaces or can I connect nodes to the server using a switch Cheers10:21
*** arun has joined #openstack10:24
*** osier has quit IRC10:28
*** wariola_ has quit IRC10:28
*** ziyadb has joined #openstack10:32
*** ziyadb has joined #openstack10:32
*** ziyadb_ has quit IRC10:33
*** ziyadb has quit IRC10:37
*** ziyadb has joined #openstack10:45
*** miclorb_ has joined #openstack11:01
*** ziyadb_ has joined #openstack11:12
*** ziyadb has quit IRC11:14
*** shang has joined #openstack11:14
*** morfeas has quit IRC11:17
*** morfeas has joined #openstack11:18
*** ziyadb_ has quit IRC11:21
*** markvoelker has joined #openstack11:23
*** ctennis has quit IRC11:25
*** rsingh has quit IRC11:27
*** jaypipes has quit IRC11:29
*** gaveen has quit IRC11:38
*** p0 has quit IRC11:43
*** arun_ has joined #openstack11:52
*** Razique has joined #openstack11:54
*** bsa has joined #openstack12:00
*** brendan1495 has quit IRC12:01
*** openpercept_ has quit IRC12:04
*** ctennis has joined #openstack12:05
*** ctennis has joined #openstack12:05
*** Funnnny has quit IRC12:06
*** lionel has quit IRC12:09
*** lionel has joined #openstack12:09
*** Funnnny has joined #openstack12:17
*** katkee has joined #openstack12:22
*** miclorb_ has quit IRC12:23
*** Funnnny has quit IRC12:44
*** ryker has joined #openstack12:49
*** p0 has joined #openstack12:51
*** hggdh has quit IRC12:55
*** hggdh has joined #openstack13:00
*** hadrian has joined #openstack13:00
*** PeteDaGuru has joined #openstack13:00
*** cereal_bars has joined #openstack13:01
*** lorin1 has joined #openstack13:04
*** imsplitbit has joined #openstack13:12
*** bcwaldon has joined #openstack13:18
*** mrjazzcat has joined #openstack13:19
*** amccabe has joined #openstack13:30
*** nid02 has joined #openstack13:31
*** nid0 has quit IRC13:34
*** msivanes has joined #openstack13:38
*** coleman has joined #openstack13:40
*** msivanes has quit IRC13:40
*** coleman has quit IRC13:41
*** mdomsch has joined #openstack13:42
*** jkoelker has joined #openstack13:43
RaziqueHi all13:44
Raziqueare the rules applied via euca-authorize apply instantely ?13:45
*** ameade has joined #openstack13:45
*** nid02 is now known as nid013:47
*** ameade has quit IRC13:55
*** 17WAAN603 has quit IRC13:56
doudeRazique: Hello, tes it'a apply instantly14:04
doudeRazique: yes*14:05
Raziquedoude: thanks14:05
Raziquehi doude :)14:05
Raziquecan I change the sg for a running instance ?14:05
*** ameade has joined #openstack14:05
*** ton_katsu has quit IRC14:06
doudeRazique: No didn't find any way to do that. I think we only can associate an instance to a sg at the boot time14:06
Raziqueerf :/14:06
doudeRazique: Are you French ?14:06
Raziqueyup14:06
doudeRazique: me, I am14:06
Razique^^Yup I figured14:06
uvirtbotRazique: Error: "^Yup" is not a valid command.14:06
*** aliguori has joined #openstack14:06
doudeRazique: :)14:07
doudeRazique: Did you try to use rescue commands ?14:08
Raziquewhich ones ?14:08
*** rchavik has quit IRC14:08
doudeRazique: with OSAPI, it's possible to use commands 'nova rescue <id>' or 'nova resume <id>'14:09
RaziqueI've never used OSAPI, what is it exactly ?14:10
doudeRazique: When I try it, i get this error in 'nova-api': http://paste.openstack.org/show/1746/14:11
*** dkim1987 has joined #openstack14:12
doudeRazique: OpenStack API, API develop specialy for the project. A first version v1.0 was done for cactus release and a new one v1.1 is drafting14:12
Raziquedoude: Oh, yes, I know it14:13
RaziqueI didn't know it OSAPI was that :)14:13
doudeRazique: it's an alternative of EC2 API14:13
Raziqueyup14:13
doudeRazique: I don't think EC2 API supports the rescue commands14:14
*** ameade has quit IRC14:20
*** ameade has joined #openstack14:21
*** cp16net has joined #openstack14:22
*** ewindisch has quit IRC14:25
doudeRazique: so, did you try a rescue command ?14:25
*** ewindisch has joined #openstack14:26
*** mattray has joined #openstack14:26
Raziquedoude: not yet, do you have a link for those commands ?14:26
RaziqueI don't know what they are used for actually ^^14:27
*** ziyadb has joined #openstack14:28
*** zul has quit IRC14:28
*** dendro-afk is now known as dendrobates14:29
*** bkkrw has joined #openstack14:29
doudeRazique: I don't have documentation about that, just http://wiki.openstack.org/RescueMode14:30
RaziqueOh, nice I was not aware of these setting14:31
Raziquegreat14:31
doudeNormaly, we just need to define flags rescue_image_id rescue_aki_id rescue_ari_id and execute 'nova rescue <server id>'14:32
doudei don't now if we must specifed the 3 flags, perhaps the rescue_image_id is enough14:33
*** dgags has joined #openstack14:33
*** cereal_bars has quit IRC14:35
*** deepakn has quit IRC14:41
doudeRazique: Do you know how the libvirt driver purge the VM image cache it created in '/var/lib/nova/instances/_base' ?14:48
Raziquedoude: I don't really know "how"14:49
doudeRazique: i cannot find any purge mechanisme14:50
*** gregp76 has joined #openstack14:59
Raziquedoude: I had this issue14:59
*** zul has joined #openstack14:59
Raziquethe cached instances took all the space ; so I run out of space14:59
Raziquebut manually ; someone here gave me a command in order to find used instances cache15:00
*** msinhore has joined #openstack15:00
*** dragondm has joined #openstack15:01
*** med_out is now known as med15:05
*** med is now known as medberry15:06
*** wilmoore has joined #openstack15:08
*** gregp76 has quit IRC15:09
*** reidrac has quit IRC15:10
*** dendrobates is now known as dendro-afk15:12
*** jj0hns0n has joined #openstack15:12
*** msivanes has joined #openstack15:16
doudeRazique: another question, did you try to inject file through the OSPI in FS of a VM ?15:16
Raziquenot yet no ^^15:17
Raziqueyou ?15:17
*** gaveen has joined #openstack15:17
*** msivanes has quit IRC15:18
*** msivanes has joined #openstack15:18
doudeRazique: I tried to set '--file' in the command 'nova boot' but nothing happend with libvirt driver and I get an error with XenAPI driver15:20
*** rminnear has joined #openstack15:21
uvirtbotNew bug: #801195 in nova "Rescue OSAPI command fails" [Undecided,New] https://launchpad.net/bugs/80119515:22
*** RJD22 has left #openstack15:22
*** heckj has joined #openstack15:27
*** wilmoore has quit IRC15:27
*** alandman has joined #openstack15:29
*** Shentonfreude has joined #openstack15:30
*** obino has quit IRC15:32
Raziquedoude: gtg home, i'll be back then15:35
Raziquecya later15:35
*** Razique has quit IRC15:35
*** rnirmal has joined #openstack15:36
*** bkkrw has quit IRC15:40
*** chemikadze has quit IRC15:44
*** dendro-afk is now known as dendrobates15:48
*** pothos is now known as pothos_15:49
*** pothos_ is now known as pothos15:49
*** guigui1 has joined #openstack15:51
*** Ryan_Lane has joined #openstack15:53
*** p0 has quit IRC16:00
*** hallyn has quit IRC16:00
NashTrashHello, Openstack'ers16:01
*** NashTrash has quit IRC16:02
*** NashTrash has joined #openstack16:02
NashTrashAnyone have any experience bringing a Win7 instance up on Openstack?16:02
*** clauden_ has joined #openstack16:02
NashTrashI have a nicely working WIn7 image on kvm-qemu, but I am not certain on what else is needed to run it under OS.16:03
*** clauden_ has quit IRC16:03
*** clauden_ has joined #openstack16:03
*** obino has joined #openstack16:03
*** MarkAtwood has quit IRC16:04
NashTrashI have some instructions for a WinXP image for UEC, but I am not certain if they apply here.16:04
*** cereal_bars has joined #openstack16:05
*** Capashen has quit IRC16:08
*** FallenPegasus has joined #openstack16:08
*** FallenPegasus is now known as MarkAtwood16:08
*** guigui1 has quit IRC16:09
*** jedi4ever has quit IRC16:10
*** guigui has joined #openstack16:13
*** pguth66 has joined #openstack16:14
*** daysmen_ has joined #openstack16:16
*** aliguori has quit IRC16:17
*** jedi4ever has joined #openstack16:18
*** maplebed has joined #openstack16:21
*** guigui has quit IRC16:22
*** Xenith has quit IRC16:23
*** arun has quit IRC16:23
*** arun_ has quit IRC16:24
*** Xenith has joined #openstack16:24
*** evil_e has joined #openstack16:25
*** tomh_ has joined #openstack16:28
*** hallyn has joined #openstack16:29
evil_ehey can someone help me with an install problem?  I've added the ppa:nova-core/truck repo and then I get a bad signature error when trying to perform a apt-get update16:29
evil_ethis is on a fresh ubuntu 10.10 system16:29
*** nacx has quit IRC16:30
uvirtbotNew bug: #801231 in nova "Too much log with VMware driver in verbose mode" [Undecided,New] https://launchpad.net/bugs/80123116:31
*** kashyap has quit IRC16:35
*** mszilagyi has joined #openstack16:38
*** Shentonfreude has quit IRC16:39
*** clauden_ has quit IRC16:39
*** clauden has joined #openstack16:39
*** Shentonfreude has joined #openstack16:40
*** aliguori has joined #openstack16:40
*** arun has joined #openstack16:41
katkeehi, when i use an instance type with 80G (m1.large) i only get 10G in the instance... any ideas??16:42
*** bsa has quit IRC16:43
uvirtbotNew bug: #801239 in nova "XenServer file inject fails" [Undecided,New] https://launchpad.net/bugs/80123916:47
*** wilmoore has joined #openstack16:47
*** kashyap has joined #openstack16:52
*** qhead has left #openstack16:52
*** fysa has joined #openstack16:53
*** clauden__ has joined #openstack16:55
*** clauden has quit IRC16:55
*** clauden__ is now known as clauden16:55
*** ziyadb_ has joined #openstack16:56
*** clauden_ has joined #openstack16:56
*** ziyadb has quit IRC16:56
*** Xenith has quit IRC16:59
*** Xenith has joined #openstack16:59
*** clauden___ has joined #openstack17:00
*** joearnold has joined #openstack17:00
*** worstadmin has quit IRC17:00
*** clauden_ has quit IRC17:00
*** clauden_ has joined #openstack17:01
*** clauden___ has quit IRC17:01
*** llang629 has joined #openstack17:02
katkeei notice that qemu-img does not specifies the size of the file being created for the instance in function _cache_image17:02
*** llang629 has left #openstack17:03
*** clauden_ has quit IRC17:07
*** fabiand__ has joined #openstack17:08
*** clauden_ has joined #openstack17:08
*** worstadmin has joined #openstack17:08
*** jdurgin has joined #openstack17:10
*** joearnold has quit IRC17:11
*** ohnoimdead has joined #openstack17:12
*** zenmatt has joined #openstack17:13
*** ohnoimdead has left #openstack17:20
*** jsm has joined #openstack17:20
*** ohnoimdead has joined #openstack17:21
*** vladimir3p has joined #openstack17:25
*** kbringard has joined #openstack17:26
kbringardahoy17:27
kbringardanyone know off the top of their head the flag for setting the instances directory?17:27
kbringardor is it inherently tied to the --state_path flag?17:27
*** daysmen_ has quit IRC17:28
*** daysmen has quit IRC17:28
kbringardis it --instances_path17:29
kbringard?17:29
*** katkee has quit IRC17:31
vishyaye17:33
*** clauden_ has quit IRC17:33
vishykatkee: it uses qemu-img to make a cow snapshot.  it is the same size as the source image17:34
kbringardvishy: does _base and stuff need to be in instances_path already, or will it create all that if it doesn't exist?17:34
vishyit will create it17:34
kbringardrad17:34
kbringardthanks17:34
*** clauden_ has joined #openstack17:39
*** jedi4ever has quit IRC17:40
*** joearnold has joined #openstack17:40
*** Xenith has quit IRC17:43
*** joearnold has quit IRC17:45
*** JuanPerez has joined #openstack17:46
*** PW__ has joined #openstack17:52
*** dmshelton has joined #openstack17:59
*** Xenith has joined #openstack18:00
*** imsplitbit has quit IRC18:00
*** bonzay is now known as zz_bonzay18:03
*** donald650 has joined #openstack18:06
redkilianhello - quick question. little fuzzy on this. if a XCP set up, do I need to install anything in dom0? nova-compute?18:06
*** Juan_Perez has joined #openstack18:09
*** JuanPerez has quit IRC18:10
*** Juan_Perez has quit IRC18:12
*** freeflying has quit IRC18:13
*** Ryan_Lane has quit IRC18:13
*** freeflying has joined #openstack18:14
*** PW__ has quit IRC18:20
*** antonyy has joined #openstack18:21
*** GeoDud has joined #openstack18:23
*** Ryan_Lane has joined #openstack18:25
*** rminnear has quit IRC18:27
donald650anyone have any luck running the tests provided in the "test" directory of swift18:29
donald650when i try to run them I get the following: UNABLE TO READ FUNCTIONAL TESTS CONFIG FILE18:29
donald650UNABLE TO READ FUNCTIONAL TESTS CONFIG FILE18:29
donald650SKIPPING FUNCTIONAL TESTS DUE TO NO CONFIG18:29
donald650the unit tests run fine18:29
*** Xenith has quit IRC18:31
*** clauden_ has quit IRC18:35
vishyredkilian: nova-compute needs to run in domU.  There are some xen plugins that need to be installed in the guide i believe18:36
*** Xenith has joined #openstack18:37
redkilianvishy: thanks, I think I saw something about the plugins18:37
redkilianso, do I need a nova-compute dumU on each dom0 in the cluster, or will just one work for the say 3-6 nodes18:38
*** ewindisch has quit IRC18:40
*** letterj has quit IRC18:40
*** ewindisch has joined #openstack18:40
*** PW__ has joined #openstack18:40
*** BK_man has joined #openstack18:42
*** ziyadb has joined #openstack18:42
*** ziyadb_ has quit IRC18:44
*** ziyadb_ has joined #openstack18:46
*** ziyadb has quit IRC18:47
*** ziyadb has joined #openstack18:48
*** ziyadb_ has quit IRC18:50
*** ziyadb_ has joined #openstack18:50
*** ziyadb has quit IRC18:52
*** antonyy has quit IRC18:53
*** ziyadb_ has quit IRC18:56
*** ziyadb has joined #openstack18:56
*** ziyadb has joined #openstack18:56
*** vernhart has quit IRC18:59
heckjI'm installing the latest ppa:nova-core/milestone to do some testing. When I install "nova-objectstore" from packages, I'm getting an error: No module named "image.service" from "import nova.image.service" (http://paste.openstack.org/show/1748/)18:59
heckjAh, nevermind - found relevant bug: https://bugs.launchpad.net/ubuntu/+source/nova/+bug/66042819:01
*** ziyadb has quit IRC19:04
*** cp16net has quit IRC19:04
*** rnirmal has quit IRC19:05
*** cp16net has joined #openstack19:07
*** rnirmal has joined #openstack19:07
*** msinhore1 has joined #openstack19:12
*** msinhore has quit IRC19:12
*** Martz has left #openstack19:13
*** aliguori has quit IRC19:18
*** aliguori has joined #openstack19:18
*** rods has quit IRC19:19
mattrayanyone have some time for a few nova/glance integration questions? glance index lists a couple of AMIs I pushed into the system, but I'm not sure how to launch them in nova19:22
mattrayeuca-describe-images isn't happy19:22
kbringardyou may not have the metadata set right19:23
kbringardhow did you upload the image(s)?19:23
mattraychances are very good :)19:23
mattrayglance-upload --host=192.168.11.10 --disk-format=ami --container-format=ami --type=machine --kernel=lucid-server-uec-amd64-vmlinuz-virtual lucid-server-uec-amd64.img lucid-server-uec-amd64.img19:23
kbringardhmmm, that looks right19:24
mattrayglance index returns the ami19:24
mattray1                lucid-server-uec-amd64.img     ami                  ami                      147639500819:24
kbringardare you getting any errors in the nova-api.log when you run euca-describe-images19:24
uvirtbotNew bug: #801297 in nova "OSAPI: GET flavors results in 500 error" [Undecided,New] https://launchpad.net/bugs/80129719:26
*** clauden_ has joined #openstack19:26
*** johnpur_ has joined #openstack19:26
*** ChanServ sets mode: +v johnpur_19:26
mattraykbringard: yes indeed: https://gist.github.com/104340519:26
zulhey does glance really need python-daemon installed it doesnt look like it19:26
kbringardmattray: ah, it's setting the kernel to the name, not the id19:27
mattraykbringard: fyi, I'm finally back on the glance stuff you pointed me to a few weeks ago. Velocity ate 2 weeks19:27
kbringardah, I heard velocity was a good time19:27
mattrayyeah, I gave a presentation with Ron and xtoddx19:28
kbringardif you glance index, what id # is lucid-server-uec-amd64-vmlinuz-virtual19:28
kbringard?19:28
kbringardand can you show me the output of glance show 119:29
*** dendrobates is now known as dendro-afk19:29
mattrayID               Name                           Disk Format          Container Format     Size19:29
mattray---------------- ------------------------------ -------------------- -------------------- --------------19:29
mattray1                lucid-server-uec-amd64.img     ami                  ami                      147639500819:29
kbringardheh, don't wanna paste in here... use paste.openstack.org ;-)19:29
mattrayhttp://paste.openstack.org/show/1749/19:30
kbringardand 1 is the only image in glance index?19:30
kbringardor is the kernel in there somewhere too?19:31
mattrayhttp://paste.openstack.org/show/1750/ is 'glance details'19:32
kbringardah, OK19:32
kbringardso, the problem is that the kernel isn't uploaded19:32
kbringardI thought glance-upload did it, but in truth I've not used it in long enough that I don't recall19:32
mattrayis glance-upload deprecated?19:33
mattrayI'm using Cactus and I was following through your helper script19:33
kbringardI don't think it's deprecated19:33
kbringardone sec19:34
*** _vinay has joined #openstack19:34
kbringardyea, OK19:34
kbringardso you need to do a glance-upload for each image19:34
kbringardso do19:34
_vinayHi, I am following the swift install instructions19:34
*** daysmen has joined #openstack19:34
_vinayhttp://swift.openstack.org/development_saio.html19:34
*** johnpur_ has quit IRC19:35
*** daysmen_ has joined #openstack19:35
kbringardglance-upload --disk-format=aki --container-format=aki --type=kernel lucid-server-uec-amd64-vmlinuz-virtual lucid-server <some name>19:35
kbringardthat'll return an ID19:35
kbringardthen, rerun your other command19:35
_vinaytowards the end it asks for "recreateaccounts"19:35
_vinayBut I dont find it19:35
_vinaynowhere did I create any such script19:35
notmyname_vinay: probably a holdover from an earlier version. you are using tempauth, right?19:36
kbringardglance-upload --host=192.168.11.10 --disk-format=ami --container-format=ami --type=machine --kernel=<id returned by the command above> l lucid-server-uec-amd64.img lucid-server-uec-amd64.img19:36
kbringardmattray: sorry, add the --host in the first one too19:36
_vinaytempauth ??19:36
mattraykbringard: and for 'ari's if they exist?19:36
kbringardyessir19:36
_vinayyes I am using tempauth19:36
kbringardmake sure you do aki and ari before you do images19:36
mattraykbringard: I think I follow19:37
kbringardbecause you need the id #s19:37
notmyname_vinay: you should be good to ignore that line referencing recreateaccounts. I'll put in a bug to remove it19:37
kbringardmy script basically consolidates all the commands down into one19:37
_vinaycoo lthanks19:37
mattraykbringard: ugh, so I need to capture the output of commands?19:37
mattraykbringard: (I'm automating this)19:37
kbringardwell, what my script does is19:37
kbringardramdisk_id=$( $glance index | awk '{ print $1, $2 }' | egrep "(${ramdisk_name}$)" | awk '{ print $1}' )19:37
mattrayyeah19:38
kbringardyou should be able to pass it a host option19:38
mattraylooking now19:38
NashTrashkbringard: Have you ever created a Windows VM image?19:38
kbringardif it works, you're welcome to just use the .bash script19:38
kbringardNashTrash: nope, sorry :-/19:38
*** timrc has quit IRC19:38
kbringardor, if you're not averse to running ruby, the glance-uploader.rb is better, imo19:39
kbringardhttps://github.com/kevinbringard/OpenStack-tools/blob/master/glance-uploader.rb19:39
NashTrashkbringard: Hmmm.  I have a nicely running VM image but I am not certain of how to get nova to properly use it19:39
notmyname_vinay: https://bugs.launchpad.net/swift/+bug/80130319:39
NashTrashAnyone have windows running on Openstack?19:39
kbringardvishy may know, I think he did some testing of that functionality19:40
NashTrashOr how about this: I have a windows disk image that I can start through KVM.  Do I have to have a kernel or ramdisk?  If so, what would they contain (in theory)19:40
kbringardhmmm, I don't think so... what I/O driver are you using when you run it in KVM?19:41
NashTrashvirtio19:41
kbringardhmmm19:41
NashTrashkvm -m 1024 -boot c -drive file=win7_x64.img,boot=on,if=virtio -nographic -vnc :0 -redir tcp:3389::338919:41
vishyno19:41
kbringardhmmm, do you get any kind of error in the compute log when you launch it?19:41
vishyno kernel ramdisk19:41
NashTrashhave not tried uploading it yet.  I wanted to see if there was any knowledge out there19:42
NashTrashhappy to give it a try though19:42
NashTrashVishy: Thanks19:42
kbringardyea, I mean... if KVM launches it, then I don't see why openstack wouldn't19:42
NashTrashYeah, it picks up a DHCP address just fine and with the 3389 redirect I can Remote Desktop into it.19:43
NashTrashOk, lets try this…to glance-upload we go...19:43
kbringardhehe19:43
kbringardI'd be curious about licensing issues and WGA crap with windows images19:44
NashTrashHey, if I get it running I can cross that bridge.19:44
kbringardhaha, agreed19:44
kbringardI was more musing out loud than anything19:44
NashTrashWell, this image will eventually have Matlab/Simulink and a whole bunch of other licensed software on it, so we will have to figure it out at some point.19:45
Jbainif you sysprep an image it prompts you for a new key doesn't it?19:45
NashTrashWhat is this sysprep you speak of?19:45
Jbainthough, that kinda blows automated launching of instances out of the water =\19:46
kbringardJbain: I think so, but depending on what you're doing you don't want it... yea19:46
kbringardwhat you said19:46
kbringardhaha19:46
kbringardyou'd probably have to have a site license with a VLK19:46
NashTrashYeah.  I need Remote Desktop enabled and a default user account.19:46
JbainNashTrash: sysprep is a tool for windows and removes all unique identifiers like hostname, license key etc19:47
Jbainwhen you reboot it prompts you to enter the new info19:47
kbringardwe've already run into some interesting problems with vendors for whose software we have site licenses, but it's always been a "call when you install it and we'll gen you a unique license for that workstation"19:47
NashTrashHmm.  Could some of that be done through User Info?  Can you run User Info on a Windows image?19:47
NashTrashSorry, User Data19:47
kbringardwhich obviously doesn't work when I do a euca-run-instances -n 2019:48
kbringardhehe19:48
NashTrashha19:48
NashTrash-n 20 works for you?19:48
kbringardyea19:48
kbringardat least.. it did a few weeks ago19:48
kbringardit's how I tested specific compute instances before the updated scheduler, lol19:48
NashTrashMust be nice.  I tried -n 10 once and it launched a few and then hung.  That was way before Diablo-1 though19:48
*** lionel has quit IRC19:49
kbringardlaunch 20 or 30 instances, and one or 2 is bound to land on the node I want19:49
NashTrashha ha ha19:49
NashTrashNice19:49
kbringardit's a blunt, yet effective' mothod19:49
kbringard:-)19:49
kbringarddangit, typing is hard19:49
*** lionel has joined #openstack19:49
uvirtbotNew bug: #801303 in swift "remove recreateaccounts reference" [Low,Confirmed] https://launchpad.net/bugs/80130319:51
NashTrashDisk format = AMI??19:51
kbringardI'd have to assume so :-/19:51
NashTrashtype=machine, right?19:51
kbringardyea, should be19:51
NashTrashuploading...19:52
kbringardI'll be curious to see how this goes19:52
*** hggdh has quit IRC19:52
*** vernhart has joined #openstack19:52
NashTrashFailed to add new image. Got error:  Unable to connect to server. Got error: [Errno 5] Input/output error19:52
mattraykbringard: what's the license on your OpenStack-tools? I'm looking to just distribute that with my OpenStack automation with Chef19:53
kbringardgo for it19:53
mattraykbringard: apache?19:53
kbringarduhm, in truth I never thought about it19:53
mattrayheh19:54
mattrayI don't really want to change it, I just want to make sure I can distribute it19:54
NashTrashNo jaypipes when you need him.  Got a glance upload error: http://paste.openstack.org/show/1751/19:54
kbringardGPL 2 lets you do what you want, yea?19:54
mattrayprobably, lemme ask around19:55
kbringardor creative commons19:55
kbringardI dunno, you're welcome to distribute it19:55
kbringardbut I guess we want to protect us all from some other jackass down the line19:55
*** msinhore1 has quit IRC19:55
*** hggdh has joined #openstack19:55
NashTrashkbringard: Looks like glance got about 2.5GB into the upload before it puked.19:56
kbringardthat's very odd19:57
*** fabiand__ has left #openstack19:57
kbringardhow big is the image?19:57
kbringardor more directly, how far into the upload did it get?19:57
NashTrash15GB19:57
kbringardlike, was it done with the data and then barfed doing something else19:57
NashTrashNope.19:58
kbringardah, OK, so it's def an issue during transmit19:58
kbringardweird19:58
kbringardI have never seen that one before19:58
NashTrashI am uploading on the machine that has glance-api and glance-registry on it.19:58
donald650hi anyone use the tests that come with the swift and nova source code?19:58
NashTrashglance-upload --disk-format=ami --container-format=ami --type=machine win7_x64.img win7_x64.img19:58
kbringardI've uploaded huge images before, so there isn't some size limit that I'm aware of19:58
NashTrashLet's try again...19:59
kbringarddo you have some kind of proxy or something running in front of it?19:59
*** mdomsch has quit IRC19:59
NashTrashNope.  Straight and simple all on one machine19:59
kbringardI had setup nginx to do authentication and it had a file upload limit set19:59
kbringardhmmm, weird19:59
NashTrash415GB available on the drive19:59
kbringardyea, it looks like it was an issue reading the http stream (chunks), but I have no idea why20:00
NashTrashIt is trying again.20:00
NashTrashGo baby go!20:00
NashTrashYou can do it!20:00
kbringardhehe20:00
NashTrashNope.  Same error.20:00
kbringardsame place too?20:00
NashTrashls -all20:01
NashTrashYea.  To the byte.20:01
kbringardgotta be something in the image messing up the stream then... I'd wager20:01
kbringardno kind of clue what would do such a thing though20:01
NashTrashThe image is just a qcow2 format image.  Do I have to do anything to prep it before I upload?20:02
kbringarddon't think so20:02
kbringardI've never done anything special with *nix images20:02
*** perestrelka has quit IRC20:02
kbringardyou're in new territory, I really have no idea what is going on there, hah20:03
*** rnirmal has quit IRC20:03
NashTrashWhere is that JayPipes?20:03
*** rnirmal has joined #openstack20:03
*** msinhore has joined #openstack20:04
notmyname_vinay: http://swift.openstack.org/development_saio.html is fixed now20:05
*** Davor` has joined #openstack20:05
Davor`Hi all. I'm trying to continue troubleshooting my configuration. I started yesterday, but didn't get very far...20:07
*** vernhart has quit IRC20:10
*** perestrelka has joined #openstack20:11
*** victor_lowther has joined #openstack20:11
kbringardmattray: let's call it MIT, I'll add a license file to the repo20:12
NashTrashOk.  Checksum'ed the image before upload.  Here we go.20:12
*** MarkAtwood has quit IRC20:13
Davor`Logs: http://eavesdrop.openstack.org/irclogs/%23openstack/%23openstack.2011-06-22.log (starts at 22:12:34)20:14
*** daysmen_ has quit IRC20:14
*** daysmen has quit IRC20:14
*** zz_bonzay is now known as zz_bonzay|away20:14
*** zz_bonzay|away is now known as bonzay20:14
NashTrashOk.  Got a good upload.  Now we run instance.20:16
mattraykbringard: you want to update the headers on your files? MIT should be fine for redistributing unchanged20:16
kbringardsure, I added a LICENSE file to the repo20:17
kbringardlet me update the individual files20:17
mattrayI'm probably just going to grab the .bash20:18
Davor`If I can't ping a VM instance, but it is shown as 'running' by euca-describe-instances, is there another way for me to check it's current state? Where would the instance's console log be?20:23
*** imsplitbit has joined #openstack20:24
kbringardDavor`: did you euca-authorize ICMP and is the VM in the group/project you authorized?20:24
kbringardyou should also be able to do euca-get-console-output <instance-id>20:25
kbringardbut, it physically lives on the compute node the instance is on (you can get this by running euca-describe-instances <instance-id> as an admin user)20:25
kbringardin /var/lib/nova/instances/<instance-id>/console.log20:25
kbringardassuming you didn't change the instance location in your config file20:25
*** Eyk has joined #openstack20:26
NashTrashkbringard: Well, the Win7 instance spawns but I can't seem to connect to it via Remote Desktop.  I did enable port 3389 for its group.20:28
*** nerens has quit IRC20:28
kbringardhmmm20:28
*** gaveen has quit IRC20:29
kbringardcan you get to it via the VNC console?>20:29
NashTrashDon't think so.  I don't have easy access to the node it is running on.20:29
kbringardhrmm20:29
kbringarddo you have the nova-vncproxy stuff setup?20:29
NashTrashNo, I haven't done that yet.20:30
kbringardwell... if it doesn't have anything you don't mind sharing, I'd be happy to grab a copy of it and try it here20:30
kbringardI can get VNC console pretty easily20:30
NashTrashNope nothing special on the image.  Where can I send the 15GB file?20:31
NashTrashPM if you prefer20:31
kbringardlemme fire up an instance and give you a login, one sec20:31
kbringardyea, lets go there20:31
*** Vasichkin has joined #openstack20:35
Davor`kbringard: thanks, euca-get-console-output worked.20:35
kbringardnp20:35
mattraykbringard: cool, it seems to be working. Does the kernel version really matter?20:35
kbringardit's a useful command to be sure20:35
mattraykbringard: s/seems to be working/is working/20:36
kbringardnaw, we just have a set of metadata that we want to be required for our environment20:36
kbringardfeel free to hack it up as necessary20:36
*** dgags has left #openstack20:36
kbringardI think you can do that with MIT20:36
kbringardhaha20:36
Davor`kbringard: yes, it is in the pingable group: euca-describe-groups says "testprj1 default ALLOWS icmp" and euca-describe-instances has "RESERVATION r-0heoo43f testprj1 default20:38
Davor`INSTANCE i-00000001 ami-21f1271f 10.0.1.2"20:38
Davor`But instance's console output doesn't look so good: DataSourceEc2.py[WARNING]: waiting for metadata service at http://169.254.169.254/2009-04-04/meta-data/instance-id -- network is unreachable"20:39
Vasichkinhello.20:39
*** PW__ has quit IRC20:39
Vasichkincan someone help me&20:39
Vasichkinhttp://paste.openstack.org/show/1754/20:39
Davor`(this could be just the image I'm using)20:39
NashTrashDavor`: Nope, that means you don't have good connectivity between your instance and the nova-api machine.20:40
*** gaveen has joined #openstack20:41
Davor`NashTrash: why does it even look for 169.254.169.254?20:42
Davor`NashTrash: can you tell me whether my network interfaces, nova.conf, and routing tables look reasonable?20:44
VasichkinRHEL 6.1, Diablo-1, can't start instance.. http://paste.openstack.org/show/1755/20:44
NashTrashDavor`: The instance needs some metadata to get started (mostly looking for its instance ID)20:44
Vasichkinsetup was made using http://wiki.openstack.org/NovaInstall/RHEL6Notes-Cactus20:46
*** HouseAway is now known as AimanA20:46
*** hadrian_ has joined #openstack20:49
*** hadrian has quit IRC20:51
*** hadrian_ is now known as hadrian20:51
Davor`NashTrash: Hm, I was just following the install instructions from the wiki.20:51
NashTrashDavor`: No worries.  This stuff gets complicated easily.  How many hardware nodes are you using and what services are on each?  Plus, what type of networking mode?20:52
Davor`NashTrash: but I probably made a mistake in trying to get fancy with my cluster's networking. Should have got the basic setup working first.20:52
*** aliguori has quit IRC20:52
Davor`NashTrash: it's supposed to be a dual-node cluster, but right now I can't even get VMs on the controller to work.20:52
NashTrashDavor`: I am a big fan of FlatDHCP.  It has generally worked pretty well for me.20:52
NashTrashDavor`: And the other node is running nova-network and nova-api together?20:53
Davor`I went with FlatDHCPManager too.20:53
*** ctennis has quit IRC20:55
NashTrashDavor`: Assuming you do have network and api on one machine and compute on the other, can you please paste the iptables for your network machine.20:55
Davor`The controller is running all services (api, compute, network, objectstore, scheduler); the other node will be just a compute node, but isn't installed yet.20:55
*** katkee has joined #openstack20:55
NashTrashOk.  In the iptables for the node do you see an entry talking about 169.254.169.254?20:56
Davor`Controller's iptables: http://pastebin.com/9eqLhCJC; ifconfig: http://pastebin.com/7GYPWBa920:56
Davor`NashTrash: No, nothing about 169.254.169.25420:57
NashTrashOk.  So you are missing that.  If you do ip addr show, can you see the 169 address on the lo?20:57
Davor`NashTrash: Yes, I do: http://pastebin.com/QuBg4uXd20:59
Davor`1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN20:59
Davor`    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:0020:59
Davor`    inet 127.0.0.1/8 scope host lo20:59
Davor`    inet 169.254.169.254/32 scope link lo20:59
NashTrashOk.  Half the battle.  Try restarting nova-network and see if you get the 169 entry to appear.  Without it the meta data requests will never make it to the nova-api service.  We can hack this, but if you can get nova-network to do it properly it would be better21:00
NashTrashIf this doesn't work, you can hack it by running this: iptables -t nat -A PREROUTING -d 169.254.169.254/32 -p tcp -m tcp --dport 80 -j DNAT --to-destination <<nova-api IP address>>:877321:01
*** alandman has quit IRC21:01
NashTrashThis is what the entry nova-network should create looks like.21:01
Davor`restarted, but still no 169.25.4169.254 in iptables.21:02
*** ameade has quit IRC21:04
Davor`NashTrash: the controller node has two NICs, one public and the other on the private subnet that I want to use as the OpenStack management LAN. Plus 10.0.1.1 on the VM subnet.21:04
NashTrashOk.  I have to head out, but try the hack and see if it works.21:04
Davor`So what should I have in iptables and nova.conf where IPs/NICs are needed?21:05
NashTrashYou may need to seek the council of truly knowledgable people like Vishy or notmyname.21:05
Davor`What's the best way to reach them? Email, Launchpad Answers, or here?21:06
kbringardhere generally21:06
* notmyname moves into the shadows....21:06
kbringardthese aren't the devs you're looking for21:06
* kbringard waves hand21:06
Davor`:-)21:07
Vasichkinhello...21:07
notmynameI'm happy to help with swift. I don't know much about nova21:07
*** lorin1 has quit IRC21:08
*** bcwaldon has quit IRC21:08
Davor`notmyname: "nova" being the controller and "swift" the object store?21:09
donald650notmyname: have you ran the functional tests that come with swift21:09
notmynamedoh! I was hoping to avoid donald650 ;-)21:10
*** ctennis has joined #openstack21:10
*** ctennis has joined #openstack21:10
notmynamedonald650: no offense, just not sure what's going on there :-)21:10
notmynamedonald650: some config file missing?21:10
notmynamedonald650: yes21:10
notmynameerr21:11
notmynameDavor`: yes21:11
notmynamedonald650: do you have a swift.conf in /etc/swift/?21:12
notmynamedonald650: ?21:15
kbringardsoren: you about mate?21:15
_vinay+notmyname thanks for fixing the docs.. . appreciate it21:15
notmyname_vinay: np. that was a simple fix. <highlight>+<delete>+push+merge :-)21:16
Davor`Adding the rule for 169.256.169.254 to my iptables didn't help. Instance console still says "network is unreachable".21:18
*** _vinay has quit IRC21:20
Davor`Although, if I "curl http://169.254.169.254/2009-04-04/meta-data/instance-id" from the host, I at least get a 404 response from the web server.21:20
donald650notmyname: yea i have that file21:21
Davor`But that's not the same as "curl localhost:8773/2009-04-04/meta-data/instance-id", which prints out a Python stack trace for "FloatingIpNotFound" exception for fixed address 127.0.0.121:21
*** msivanes has quit IRC21:21
notmynamedonald650: hmm...that was my first guess. will you be in later this evening? or tomorrow? something came up at home and I have to take of early21:21
donald650notmyname: yes I will be on tommorrow21:23
donald650notmyname; there is little documenation on these tests do you have any links to where I can learn more21:23
*** nvdm has joined #openstack21:24
*** nvdm has left #openstack21:26
*** chalupa has quit IRC21:26
*** h1nch has quit IRC21:26
*** msivanes has joined #openstack21:28
*** msivanes has quit IRC21:32
*** jsm has quit IRC21:34
vishyDavor` you shouldn't need any manual rules if things are set up properly21:37
*** GeoDud has quit IRC21:43
*** _vinay has joined #openstack21:45
_vinayOK .. so in order for my nova installation to talk to swift.. I have to go through glance .. correct?21:46
*** pothos_ has joined #openstack21:46
*** kashyap has quit IRC21:48
*** pothos has quit IRC21:48
*** pothos_ is now known as pothos21:48
*** imsplitbit has quit IRC21:51
*** bonzay is now known as zz_bonzay21:51
*** msinhore has quit IRC21:52
*** zz_bonzay is now known as zz_bonzay|away21:53
*** zz_bonzay|away is now known as bonzay21:53
*** bsa has joined #openstack21:56
katkeehi, when i create a m1.large instance i only get an instance with 10G of disk space instead of 80G. any ideas?22:00
kbringardkatkee: check /dev/vdb22:01
kbringardit's unformatted and unpartitioned22:01
kbringardso you'll have to do like22:01
kbringardfdisk /dev/vdb22:01
kbringardpartition it out22:01
kbringardsave it, etc22:01
kbringardthen mkfs.ext{2,3,4} or whatever22:02
kbringardon /dev/vdb122:02
kbringardthen mount it somewhere and a df should show that mountpoint as being 80G22:02
kbringardthe reason the root is only 10G is because that's the size of the root filesystem of image you're loading22:03
kbringardthe disk space associated with the flavor (machine size) gets allocated out and loaded as /dev/vdb, but doesn't get partitioned, formatted, etc22:03
*** bsa has quit IRC22:03
kbringardit should be pretty easy to write a user_data script to load to do that though22:03
kbringardanyway, hope that helps22:04
kbringardI need to call it a day, have a good one cloudies22:04
*** kbringard has quit IRC22:04
*** h1nch has joined #openstack22:06
*** allsystemsarego has quit IRC22:06
uvirtbotNew bug: #797211 in swift "Need archived doc site for Swift 1.3 documentation built from RST" [Undecided,New] https://launchpad.net/bugs/79721122:06
*** Eyk has quit IRC22:07
*** markvoelker has quit IRC22:07
*** katkee has quit IRC22:09
Davor`vishy: I now wonder if the rule had been there, but not shown with "iptables --list"22:10
*** rnirmal has quit IRC22:11
Davor`after adding it by hand, I had multiple copies of it with "iptables -t nat -n -L"22:11
*** libspeed has quit IRC22:11
*** katkee has joined #openstack22:11
Davor`And if I read the rule correctly, it forwards traffic directed to 169.254.169.254:80 to API_host:877322:12
Davor`But "curl" to 169.254.169.254 gets me the web server on API_host:80, not the same as API_host:877322:13
Davor`And the instance console shows that it can't reach even 169.254.169.254 during startup22:14
vishyyes22:15
vishyyou need to do iptables -t nat22:15
vishy--list22:15
vishyto see it22:15
vishydo you have network and api on the same host?22:16
*** Eyk has joined #openstack22:18
*** cowmix has quit IRC22:20
*** aliguori has joined #openstack22:28
Davor`vishy: yes22:31
Davor`Everything is on the one host. I want to set up an additional compute node, but right now I can't get even a single host to work.22:32
*** vernhart has joined #openstack22:33
Davor`vishy: let me know if this would be better done on Launchpad.22:35
*** martin has quit IRC22:38
vishyeverything should be working on one host22:38
vishyunless you have setup some odd ip addresses22:38
*** martin has joined #openstack22:39
Davor`vishy: that's probably where the problem is... My host has two NICs, and I'm using both of them. So what should the parameters in my nova.conf be for IPs and interfaces?22:39
vishywell first of all you should make sure you can get it running without using any interfaces22:40
vishyjust br10022:40
Davor`Sorry, what do you mean by that? I do have br100 present. Its "link/ether" is the same as eth0's, so I assume it's bound to that NIC.22:42
Davor`br100 also has two "inet" entries: 10.0.1.1/27 (VM network) and 192.168.11.1/24 (management subnet for OpenStack hosts)22:43
Davor`eth1 has the public IP, and lo has 169.254.169.254/32 bound to it (all info based on the output of "ip addr show")22:44
*** GeoDud has joined #openstack22:45
*** mattray has quit IRC22:46
vishyyes22:48
vishywell i was suggesting to not bind it to eth0, but that is fine22:48
vishyare you getting dhcp properly?22:48
Davor`vishy: I'm not sure how (not) to do that. Is that because of "flat_interface=eth0" in nova.conf?22:50
vishyare yes that does it22:51
vishybut let me repeat22:51
Davor`How do I check DHCP's status?22:51
vishydo your vms get ip addresses?22:51
vishyit is usually in the console log22:51
*** cp16net has quit IRC22:51
Davor`They have an address in euca-describe-instance.22:52
vishythat means nothing :)22:52
vishyyou should also get dhcp messages22:52
vishyin syslog22:52
vishyor /var/log/daemon.log22:53
Davor`No mention of IP address assignment in instance's console.log. It ends with "waiting for metadata service" and about a dozen "Network is unreachable" warnings that come from DataSourceEc2.py22:54
Davor`and re DHCP, this is probably a bad sign: "dnsmasq-dhcp[1495]: no address range available for DHCP request via br100"22:55
worstadminMeans you need to edit your config22:55
worstadminYour either out of addresses to assign or you havent specified a range22:55
*** zul has quit IRC22:55
Davor`I do have "fixed_range=10.0.0.0/12" in nova.conf. And "network_size=6000" and only one instance running, so I don't see how it could run out of addresses.22:56
Davor`dnmasq process is running as: dnsmasq --strict-order --bind-interfaces --conf-file= --domain=novalocal --pid-file=/var/lib/nova/networks/nova-br100.pid --listen-address=10.0.1.1 --except-interface=lo --dhcp-range=10.0.0.2,static,120s --dhcp-lease-max=32 --dhcp-hostsfile=/var/lib/nova/networks/nova-br100.conf --dhcp-script=/usr/bin/nova-dhcpbridge --leasefile-ro22:57
*** MarkusWPHRorg has joined #openstack22:58
MarkusWPHRorghi all22:58
MarkusWPHRorggood morn/eve/aftrnoon all22:58
Davor`Ah, I see something: dnsmasq's dhcp-range is 10.0.0.2, but the instance is 10.0.1.2.22:58
*** Eyk has quit IRC23:00
Davor`On second thought, that shouldn't be an error. nova.conf's "fixed_range=10.0.0.0/12" does include 10.0.1.2.23:01
MarkusWPHRorgGotta question about basic funtionality in Openstack Networking23:01
MarkusWPHRorgWhen you do a 'nova-manage floating create <ipaddress>'23:01
Davor`And /var/lib/nova/networks/nova-br100.conf has an entry for the instance: "02:16:3e:33:6f:a0,i-00000001.novalocal,10.0.1.2"23:02
MarkusWPHRorgdoes nova set up an aliased interface or something for that ip, or does it have to already exsit on an interface(aliased or otherwise)?23:02
*** dkim1987 has quit IRC23:03
MarkusWPHRorgSame goes for "euca-allocate-address" or wherever it is the "new" ip is actually brought into being23:03
*** amccabe has quit IRC23:05
*** katkee has quit IRC23:06
*** Shentonfreude has quit IRC23:06
*** jj0hns0n has quit IRC23:07
*** Vasichkin has quit IRC23:08
*** aliguori has quit IRC23:18
*** dragondm has quit IRC23:19
MarkusWPHRorgI'm just tryign to figure out what the default behavior is since the allocate/assoicate thing isn't working for me23:20
*** ryker has quit IRC23:20
*** chilts has left #openstack23:21
*** _vinay has quit IRC23:21
*** miclorb_ has joined #openstack23:25
*** Ryan_Lane has quit IRC23:31
*** jkoelker has quit IRC23:33
*** Ryan_Lane has joined #openstack23:34
*** katkee has joined #openstack23:37
*** vernhart has quit IRC23:38
*** vernhart has joined #openstack23:44
*** nelson____ has joined #openstack23:44
*** zul has joined #openstack23:45
adam_gdoes nova-network or some other component now add an iptables rule to redirect metadata server requests to the api server?23:46
*** katkee has quit IRC23:46
*** osier has joined #openstack23:49
*** gaveen has quit IRC23:49
*** Cromulent has joined #openstack23:51
*** mszilagyi has quit IRC23:57
*** fysa has quit IRC23:59

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