| *** andreaf has quit IRC | 00:00 | |
| *** huslage has joined #openstack | 00:00 | |
| *** philippkueng has quit IRC | 00:00 | |
| *** cdub has quit IRC | 00:00 | |
| *** clauden_ has quit IRC | 00:01 | |
| *** clauden has joined #openstack | 00:01 | |
| *** rnorwood has left #openstack | 00:02 | |
| *** bsza has quit IRC | 00:04 | |
| *** dirakx1 has quit IRC | 00:11 | |
| *** dirakx2 has joined #openstack | 00:11 | |
| *** nerdstein has joined #openstack | 00:13 | |
| *** iRTermite has quit IRC | 00:15 | |
| *** jog0 has quit IRC | 00:19 | |
| *** nati2 has quit IRC | 00:21 | |
| *** vladimir3p has joined #openstack | 00:22 | |
| *** dtroyer has joined #openstack | 00:24 | |
| *** cdub has joined #openstack | 00:27 | |
| *** adjohn has quit IRC | 00:29 | |
| *** jakedahn has quit IRC | 00:29 | |
| *** jakedahn has joined #openstack | 00:29 | |
| *** adjohn has joined #openstack | 00:29 | |
| *** isaacfinnegan has quit IRC | 00:34 | |
| *** nati2 has joined #openstack | 00:36 | |
| *** nerdstein has quit IRC | 00:37 | |
| *** uncleofthestick has quit IRC | 00:41 | |
| *** dragondm has quit IRC | 00:48 | |
| *** nerdstein has joined #openstack | 00:48 | |
| *** cdub has quit IRC | 00:49 | |
| *** jakedahn has quit IRC | 00:56 | |
| *** cereal_bars has joined #openstack | 00:57 | |
| *** adjohn has quit IRC | 01:01 | |
| *** qazwsx has quit IRC | 01:04 | |
| *** obino has quit IRC | 01:13 | |
| *** nerdstein has quit IRC | 01:16 | |
| *** vidd-away is now known as vidd | 01:16 | |
| *** cereal_bars has quit IRC | 01:17 | |
| *** doorlock has joined #openstack | 01:20 | |
| *** cdub has joined #openstack | 01:21 | |
| *** jog0 has joined #openstack | 01:24 | |
| *** ron-slc has quit IRC | 01:28 | |
| *** herbster has joined #openstack | 01:29 | |
| herbster | hi guys | 01:29 |
|---|---|---|
| herbster | we have set up nova-volume and seems to be working well | 01:29 |
| herbster | however i have a permissions problem | 01:29 |
| herbster | as the admin user, i can create a volume and then create a snapshot of that volume in hybridfox/euca2ools/boto | 01:29 |
| herbster | however, one of the other users (who has the nova roles netadmin and sysadmin) | 01:30 |
| herbster | can only create volumes | 01:30 |
| herbster | when they try to create a snapshot from that volume | 01:30 |
| herbster | "Unauthorized request for controller=CloudController and action=CreateSnapshot" | 01:30 |
| herbster | (from nova-api.log) | 01:30 |
| *** jog0 has quit IRC | 01:31 | |
| *** huslage has quit IRC | 01:32 | |
| *** nci has joined #openstack | 01:33 | |
| *** doorlock has quit IRC | 01:34 | |
| *** Ryan_Lane has quit IRC | 01:38 | |
| nci | Hi all, I'm having some strange issues with Swift. Despite the operations succeeding (listing containers will show the new item), I am receiving 404 errors when I attempt to create or delete containers. Anyone have any ideas? | 01:38 |
| *** Ryan_Lane has joined #openstack | 01:38 | |
| herbster | what do the logs on swift-proxy say nci? | 01:39 |
| *** jakedahn has joined #openstack | 01:39 | |
| nci | swift-proxy proxy-server 150.203.248.116 150.203.248.116 03/Nov/2011/01/29/14 PUT /v1/AUTH_admin/blarggggg%253Cbuilt-in%2520method%2520now%2520of%2520type%2520object%2520at%25200x7f00ba295760%253E HTTP/1.0 404 | 01:40 |
| nci | - python-cloudfiles/1.7.9.2 admin%2CAUTH_tk2bfce33b5a344e278bd6c6d18a844dbe - - - tx5d0b6b94bbba406f8a69a1793ee4e6d5 - 0.0755 | 01:41 |
| *** adrianmoya has joined #openstack | 01:42 | |
| nci | Here's the 3 lines I get: | 01:42 |
| herbster | have you distributed the ring files from the proxy onto the storage nodes? | 01:42 |
| nci | Nov 3 12:42:02 swift-proxy tempauth - 150.203.248.116 03/Nov/2011/01/42/02 GET /auth/v1.0 HTTP/1.0 200 - python-cloudfiles/1.7.9.2 - - - - txe24ced0e04df4dd1b59a20ac7347062b - 0.0005 | 01:42 |
| nci | Nov 3 12:42:02 swift-proxy proxy-server 150.203.248.116 150.203.248.116 03/Nov/2011/01/42/02 GET /v1/AUTH_admin/%3Fformat%3Djson HTTP/1.0 200 - python-cloudfiles/1.7.9.2 admin%2CAUTH_tk2bfce33b5a344e278bd6c6d18a844dbe - 878 - txd92a7f0d8877467a805479a34c85a07c - 0.0081 | 01:42 |
| nci | Nov 3 12:42:03 swift-proxy proxy-server 150.203.248.116 150.203.248.116 03/Nov/2011/01/42/03 PUT /v1/AUTH_admin/test900 HTTP/1.0 404 - python-cloudfiles/1.7.9.2 admin%2CAUTH_tk2bfce33b5a344e278bd6c6d18a844dbe - - - txc0c672ef265143278ca516600c6a215b - 0.1137 | 01:42 |
| nci | root@swift-proxy:/etc/swift# md5sum container.ring.gz | 01:43 |
| nci | 70c4fcd83e811beb99d51cbdbf084546 container.ring.gz | 01:43 |
| nci | root@swift-dory:/etc/swift# md5sum container.ring.gz | 01:43 |
| nci | 70c4fcd83e811beb99d51cbdbf084546 container.ring.gz | 01:43 |
| nci | seems ok... | 01:43 |
| nci | I get no errors putting getting objects | 01:44 |
| nci | and the containers are being created and are usable - I can put objects in them | 01:44 |
| *** flamboz has joined #openstack | 01:44 | |
| nci | but the client is receiving a 404, which is a bit concerning | 01:45 |
| adrianmoya | Good night guys, I have a question: I have a two node install, and began launching instances but they seem to be all created at the controller. The weird thing is I have break the physical memory limit of the two servers and still the instances launch (a total of 12.3GB RAM in a 4GB ram physical server, and the other node has 8GB with 7GB free) | 01:46 |
| *** cdub has quit IRC | 01:47 | |
| adrianmoya | I'm not sure if I'm understanding what's happening... | 01:47 |
| herbster | nci: what exactly are you getting a 404 doing? deleting the container? | 01:47 |
| *** nid0 has quit IRC | 01:47 | |
| nci | herbster: creating a conatiner | 01:47 |
| herbster | adrianmoya: when you type nova-manage service list do the other nodes show up as nova-compute/available? | 01:47 |
| *** jog0 has joined #openstack | 01:48 | |
| herbster | nci: strange. you get a 404 when you create the container, but it does create it anyway, and you can upload/download files to it? | 01:48 |
| adrianmoya | herbster: let me check (I'm using dashboard+keystone, can I use nova-manage?) | 01:48 |
| herbster | in cactus or diablo | 01:48 |
| adrianmoya | herbster: diablo (controller installed via devstack, node via packages) | 01:49 |
| herbster | adrianmoya: sorry, the q was for nci | 01:49 |
| nci | herbster: yep, that's what I'm seeing. The functionality is all there, but the return code is a 404. I'm wondering if the replication is not working and the 404 is saying that the second copy is not being written. | 01:49 |
| nci | herbster: I'm running swift 1.4.3 | 01:50 |
| adrianmoya | herbster: yes, it appears as XXX :S | 01:51 |
| herbster | adrianmoya: i didnt ask if it said XXX, i asked if it showed as available | 01:51 |
| herbster | or enabled | 01:51 |
| adrianmoya | herbster: enabled | 01:51 |
| herbster | adrianmoya: tail the nova-api log on nova-api host and nova-compute log on nova-compute host | 01:52 |
| herbster | this should give you some diagnostics as to whats happening when you try to launch an instance | 01:52 |
| *** df1 has quit IRC | 01:52 | |
| herbster | nci: i would try tailing the logs on all your storage nodes and proxy at the same time | 01:53 |
| herbster | there has to be an error in there somewhere | 01:53 |
| herbster | personally in both cactus and diablo i found swift to be the most stable and "working as expected" part of openstack | 01:54 |
| nci | herbster: yep that's been my experience too, Swift has been very stable for us, I'm not sure what has happened. Thanks for your help | 01:54 |
| herbster | sorry i couldnt be more useful :/ | 01:54 |
| *** herbster has quit IRC | 01:55 | |
| *** Ryan_Lane has quit IRC | 01:58 | |
| adrianmoya | herbster: launched an extra 2GB instance successfully (14.4GB used of 11.6GB total). No errors in nova log in controller, nothing happened in nova-compute log in the other compute node | 01:58 |
| *** ccc1 has joined #openstack | 02:00 | |
| *** e1mer has joined #openstack | 02:03 | |
| adrianmoya | anybody willing to help me get a compute node from XXX to :-) ? | 02:05 |
| *** jog0 has quit IRC | 02:07 | |
| vidd | adrianmoya, compute node as in second machine? | 02:08 |
| adrianmoya | vidd: yes, it's a second machine | 02:09 |
| vidd | adrianmoya, this my sound stupid...but is the time in sync on both machines? | 02:09 |
| *** mandela123 has joined #openstack | 02:09 | |
| adrianmoya | let me check | 02:10 |
| *** shang has joined #openstack | 02:10 | |
| vidd | it is recommended that you set up an NTP server on the controller and have all nodes sync with it | 02:11 |
| adrianmoya | no :$ ok, I should install NTP right? let me do that | 02:11 |
| vidd | dont be embarrassed...time sync is a common issue =] | 02:12 |
| *** charless has joined #openstack | 02:12 | |
| vidd | and doc.openstack.org dont mention it | 02:12 |
| vidd | *docs.openstack.org | 02:13 |
| *** joeyang has joined #openstack | 02:13 | |
| charless | Can I get someone to verify my understanding of what openstack storage can do? | 02:13 |
| vidd | charless, storage like "swift" or storage like "nova-volume"? | 02:14 |
| vidd | does glance count as storage? =] | 02:15 |
| charless | vidd, Well, I don't know what nova-volume is. | 02:15 |
| charless | I want to storage a lot of documents, pdf, images, jpegs, etc. for my java based webserver to server when ever needed. | 02:16 |
| *** tokuzfunpi has quit IRC | 02:16 | |
| charless | I'd like to have some sort of automated redundancy too. | 02:16 |
| adrianmoya | vidd: I installed ntp on cloud-controller, but I'm not sure how to configure the node and what to install... was following a quick google result but package ntp-simple seems to have disapeared from ubuntu | 02:17 |
| *** jdurgin has quit IRC | 02:17 | |
| charless | so I can have my storage automatically replicate my documents to multiple devices so that I can shutdown a node without affecting the website. | 02:18 |
| vidd | charless, i dont think im qualified to answer your question =[ | 02:18 |
| charless | well, how about this? | 02:18 |
| vidd | but it sounds like you want swift | 02:18 |
| charless | Can I store files on it and it will automatically replicate it across multiple nodes? | 02:19 |
| *** crescendo has quit IRC | 02:19 | |
| vidd | charless, as i said...it sounds like you want swift...and i do not know enough about swift to answer your questions | 02:19 |
| charless | is there a flow chart of how swift works ? | 02:20 |
| vidd | adrianmoya, this doc is dated, but it should help:https://help.ubuntu.com/8.04/serverguide/C/NTP.html | 02:20 |
| vidd | adrianmoya, you of course want to set your node to use your controller instead of the ubuntu.com default | 02:21 |
| vidd | charless, the documentation for swift is here: http://docs.openstack.org/diablo/openstack-object-storage/admin/content/ | 02:21 |
| *** pixelbeat has quit IRC | 02:22 | |
| charless | thanks, vidd | 02:22 |
| vidd | adrianmoya, you dont have to install ntpdate...its pre-installed | 02:23 |
| vidd | adrianmoya, you just have to set up the cron and adjust the server | 02:23 |
| adrianmoya | vidd: Ok, for what I understood, installing ntp on both servers and configuring the second and adding the first as server could do the trick, right? and setting the cron | 02:24 |
| vidd | adrianmoya, http://cssoss.files.wordpress.com/2011/10/openstackbookv2-0_csscorp.pdf has step-by-steps of the complete setup process for both servers including the ntp info | 02:26 |
| vidd | they recommend setting a "fuding" process in case your controller cannot access the internet to get time | 02:26 |
| adrianmoya | vidd: ok, both are sync... what can I check next? | 02:29 |
| vidd | check if it is :-) | 02:29 |
| adrianmoya | no, is still XXX... let me restart the compute node | 02:30 |
| mandela123 | does the latest trunk nova-api still need openstackx plugin? | 02:30 |
| vidd | if not, restart compute on second node | 02:30 |
| vidd | mandela123, are you using dashboard? | 02:30 |
| vidd | IIRC that is the only reason you would ever need openstackx | 02:31 |
| mandela123 | yes | 02:31 |
| vidd | dont quote me on it mandela123 but i do believe it does | 02:31 |
| mandela123 | but i cant find the flag any more | 02:31 |
| mandela123 | osapi_extensions_path | 02:32 |
| vidd | mandela123, i dont have an answer for you =[ | 02:32 |
| charless | adrianmoya: I came in at the middle of your conversation, Am I to understand your ntp is not syncing your time between your 2 servers? | 02:32 |
| * vidd uses stable-diablo | 02:32 | |
| mandela123 | now i use the trunk | 02:33 |
| vidd | charless, adrianmoya didnt HAVE ntp set up between the two | 02:33 |
| charless | oh, yeah, definite always good to have ntpo syncing your time. | 02:34 |
| vidd | =] | 02:34 |
| vidd | ppl dont realize 15 seconds is an eternity in computer-time =] | 02:34 |
| adrianmoya | vidd, charless: already :-), but only if I run it manually, it doesn't start, I'm not sure why, as it does not write the log | 02:34 |
| adrianmoya | sorry: I meant if I sudo service nova-compute start | 02:35 |
| adrianmoya | it doesn't work | 02:35 |
| *** dirakx2 has quit IRC | 02:35 | |
| vidd | adrianmoya, im not sure i understand..... | 02:35 |
| nci | charless: yes, swift replicates objects that you store in there and sounds suitable for what you're doing | 02:35 |
| charless | awesome, I'm looking at the documentation now. | 02:36 |
| vidd | are you saying that you restart compute, it becomes :-) for a minute then XXX again? | 02:36 |
| *** MarkAtwood has quit IRC | 02:36 | |
| *** df1 has joined #openstack | 02:36 | |
| *** charless has left #openstack | 02:37 | |
| *** mgius has quit IRC | 02:37 | |
| adrianmoya | no, I manually run sudo /usr/bin/nova-compute and it gets :-). But I'd like to start it as a service which doesn't work. | 02:37 |
| *** osier has joined #openstack | 02:38 | |
| vidd | so "service nova-compute start" does not work? | 02:38 |
| adrianmoya | no | 02:38 |
| vidd | adrianmoya, did you install via apt-get or via git? | 02:38 |
| adrianmoya | via apt-get | 02:39 |
| vidd | what does "ls -al /etc/init.d/ |grep nova" return? | 02:40 |
| uvirtbot | New bug: #885529 in glance "Use of _ function without importing it" [Undecided,New] https://launchpad.net/bugs/885529 | 02:41 |
| vidd | what does "ls -al /etc/init.d/ |grep nova" return? adrianmoya | 02:42 |
| adrianmoya | I found the problem, /etc/nova/nova.conf had wrong ownership... let me correct this, but in the meantime: Is it normal that openstack let me break the physical memory limit? | 02:42 |
| vidd | there is a physical memory limit? | 02:42 |
| vidd | are you saying you have 5 vm's running 2 gig ram each and you only have 8 gig ram? | 02:43 |
| vidd | or you have 2 gig ram and you try to start a vm with 4 gig ram? | 02:43 |
| adrianmoya | My controller has 4GB ram, and I'm running a total of 14.4GB on this node | 02:44 |
| adrianmoya | BTW: all happy faces now :-) thanks | 02:44 |
| vidd | but each vm is running less than 4 gb ram...right? | 02:44 |
| adrianmoya | yes, they seem to be doing something ala openvz | 02:45 |
| adrianmoya | because there is not such amount of memory in the two nodes | 02:45 |
| vidd | your are not breaking the "physical memory limit" | 02:45 |
| vidd | you can "switch-hit" the physical ram | 02:46 |
| uvirtbot | New bug: #885531 in devstack "nova-volume dependency" [Undecided,New] https://launchpad.net/bugs/885531 | 02:46 |
| adrianmoya | MMM, I thought kvm reserved memory for each instance | 02:46 |
| *** wariola has joined #openstack | 02:46 | |
| vidd | i thought xen did that...not kvm | 02:47 |
| * vidd could be wrong...it has happened before | 02:47 | |
| adrianmoya | so, with kvm, I can share physical memory between servers? :O | 02:48 |
| vidd | not between servers....between VMs | 02:48 |
| adrianmoya | correct, between vm.. but nova is not respecting ram quotas, I already view a bug reported on that issue | 02:49 |
| adrianmoya | So I ended up with -2.7 GIB AVAIL | 02:49 |
| vidd | adrianmoya, but all your VM's work | 02:50 |
| adrianmoya | yes. TBH, I've only ssh to the last of them, and everything seemed ok | 02:50 |
| adrianmoya | let me destroy them all, and recreate them to see if they get to the other node | 02:51 |
| vidd | IDK TBH i would call it a "feature" and not a bug .... but maybe i've worked with the software from a particular washingtom company for too many years =] | 02:52 |
| vidd | euca-describe-availability-zones shows no avalable aonesbut i have two VM's running =\ | 02:54 |
| *** hugokuo has joined #openstack | 02:54 | |
| *** helloworld has joined #openstack | 02:57 | |
| helloworld | hi,i install dashboard 46,after "sudo tools/with_venv.sh dashboard/manage.py syncdb",get this error "socket.error: [Errno 111] Connection refused",anyone help me ? | 02:59 |
| *** ldlework has quit IRC | 03:00 | |
| vidd | helloworld, is that all you get? no traceback? | 03:00 |
| vidd | helloworld, if you do have a traceback, please pastebin it | 03:00 |
| helloworld | Traceback (most recent call last): File "dashboard/manage.py", line 11, in <module> execute_manager(settings) File "/opt/osdb/openstack-dashboard/.dashboard-venv/lib/python2.6/site-packages/django/core/management/__init__.py", line 438, in execute_manager utility.execute() File "/opt/osdb/openstack-dashboard/.dashboard-venv/lib/python2.6/site-packages/django/core/management/__init__.py", line 379, in execute | 03:00 |
| helloworld | i use sqllite3 | 03:00 |
| vidd | helloworld, please post the traceback in a pastebin....chat does not have enough space to show it all | 03:01 |
| vidd | http://paste.openstack.org/ | 03:02 |
| helloworld | thanks | 03:02 |
| *** yeming has joined #openstack | 03:02 | |
| helloworld | hi,vidd ,http://paste.openstack.org/show/3050/ | 03:03 |
| *** ejat has quit IRC | 03:04 | |
| *** ejat has joined #openstack | 03:04 | |
| *** ejat has joined #openstack | 03:04 | |
| helloworld | i use setting in /var/lib/nova/bin/novarc | 03:04 |
| *** dosdawg has joined #openstack | 03:05 | |
| adrianmoya | vidd: now the vm stucks in build... it seems to be trying to create it in the new node | 03:06 |
| *** obino has joined #openstack | 03:06 | |
| adrianmoya | no errors in the log of both servers | 03:06 |
| vidd | helloworld, what do you mean you used the settings from novarc? | 03:07 |
| *** flamboz has quit IRC | 03:07 | |
| vidd | adrianmoya, have you looked in the -network, -api, and -schedular logs on the contoller? | 03:08 |
| *** hadrian has quit IRC | 03:08 | |
| helloworld | yes,@vidd | 03:08 |
| vidd | helloworld, it was not a "yes/no" question....please clarify what you maen by you used the settings in a credentials file for setting up a database | 03:09 |
| uvirtbot | New bug: #885540 in nova "stable/diablo didn't pass pep8" [Undecided,New] https://launchpad.net/bugs/885540 | 03:11 |
| *** rnorwood has joined #openstack | 03:12 | |
| adrianmoya | vidd: nothing in any log on controller | 03:13 |
| vidd | adrianmoya, im not sure where to point you to next =\ | 03:13 |
| *** joonwon has quit IRC | 03:14 | |
| helloworld | @vidd,yes-> username =‘NOVA_USERNAME="dev"’ from novarc,password is random "123456" | 03:14 |
| adrianmoya | and nothing in the node :S I'm expecting an ERROR on any log? | 03:14 |
| larissa | helloworld: Error: "vidd,yes->" is not a valid command. | 03:14 |
| *** mmetheny has joined #openstack | 03:15 | |
| vidd | does the nova.conf file on the node have the --my_ip and --server_ip flags? | 03:15 |
| vidd | LOL @ larissa | 03:15 |
| vidd | helloworld, none of that info is relevent to dashboard talking to its database | 03:16 |
| *** dysinger has quit IRC | 03:17 | |
| *** dendrobates is now known as dendro-afk | 03:17 | |
| adrianmoya | vidd: I think I found the error: nova-network is not running on the new node. I'm using multi_host=1 | 03:18 |
| vidd | or is this info you placed in the local settings py file? | 03:18 |
| *** flamboz has joined #openstack | 03:18 | |
| *** dendro-afk is now known as dendrobates | 03:18 | |
| vidd | adrianmoya, i think you found a winner =] | 03:19 |
| adrianmoya | Stderr: 'sh: /usr/lib/python2.7/dist-packages/bin/nova-dhcpbridge: not found\n\ndnsmasq: cannot run lease-init script /usr/lib/python2.7/dist-packages/bin/nova-dhcpbridge: No such file or directory\n' | 03:19 |
| vidd | adrianmoya, are you running flatdhcp on these nodes? | 03:20 |
| adrianmoya | yes | 03:20 |
| vidd | do you have 2 eths in the node? | 03:20 |
| adrianmoya | yes | 03:20 |
| vidd | did you set up br100 to point to the second eth in the compute-node? | 03:21 |
| helloworld | hi,@vidd, You just installed Django's auth system, which means you don't have any superusers defined. Would you like to create one now? (yes/no): yes Username (Leave blank to use 'root'): dev E-mail address: [email]nospam@nospam.com[/email] Password: Password (again): | 03:21 |
| vidd | adrianmoya, http://docs.openstack.org/diablo/openstack-compute/admin/content/configuring-multiple-compute-nodes.html has info on the proper network setup for you | 03:24 |
| adrianmoya | both have the same nova.conf | 03:24 |
| adrianmoya | with br100 on top of eth0 | 03:25 |
| vidd | helloworld, the connection error has to do with dashboard writing to its dtabase...not with the username/password you set up for the user interface | 03:25 |
| vidd | adrianmoya, please paste the /etc/network/interfaces of the compute node for me | 03:26 |
| adrianmoya | # The primary network interface auto eth1 iface eth1 inet static address 192.168.3.101 netmask 255.255.255.0 gateway 192.168.3.1 dns-nameserver 8.8.8.8 auto eth0 iface eth0 inet static address 192.168.5.101 netmask 255.255.255.0 | 03:27 |
| vidd | adrianmoya, http://docs.openstack.org/diablo/openstack-compute/admin/content/configuring-multiple-compute-nodes.html | 03:27 |
| vidd | http://paste.openstack.org/ @ adrianmoya | 03:27 |
| adrianmoya | notice that I didn't setup the bridge in /etc/network/interfaces as it seems compute does this automatically | 03:27 |
| vidd | compute does this on the controller automatically | 03:28 |
| vidd | the nodes need it | 03:28 |
| adrianmoya | http://paste.openstack.org/show/3051/ | 03:28 |
| adrianmoya | mmm, so I should manually set up my bridge on the node? let me try tha | 03:29 |
| adrianmoya | t | 03:29 |
| vidd | adrianmoya, which is your public and which is your private netmask? | 03:29 |
| vidd | also, eth0 has no gateway specified...did you cut it off in the paste? | 03:30 |
| helloworld | @vidd,"Connection refused" means connect sqlite3 refused? | 03:30 |
| larissa | helloworld: Error: "vidd,"Connection" is not a valid command. | 03:30 |
| vidd | helloworld, what larissa is saying is "do not start a line with the "@" symbol | 03:31 |
| helloworld | ok | 03:31 |
| vidd | helloworld, yes, sqlite is having some difficulty | 03:32 |
| adrianmoya | my public network is 192.168.3.x , and for openstack I have a subnet from 3.160 to 3.191. It has no gateway as my access is from eth1 (eth0 is connected to a simple switch with the other node) | 03:32 |
| vidd | adrianmoya, it needs the gateway to the private network | 03:33 |
| vidd | so the .5.xxx is your private range for the VM's | 03:34 |
| *** troya has joined #openstack | 03:35 | |
| adrianmoya | no, they are in 10.0.0.x | 03:36 |
| troya | hey All | 03:36 |
| errr | howdy! | 03:36 |
| adrianmoya | hey, I could create the bridge, it's already created (I think nova-compute did it) | 03:37 |
| troya | anyone have tried swift ? | 03:37 |
| adrianmoya | I couldn't I mean | 03:38 |
| errr | troya: sure, its great.. one of the most stable parts of the project i think | 03:38 |
| notmyname | troya: do you have specific questions about swift? | 03:38 |
| troya | wow, it's look interesting, so after nova, i want try it :D | 03:38 |
| troya | notmyname: mm, can you explains to me, swift usually what for ? | 03:39 |
| notmyname | troya: most simply, it's for unstructured data that can grow without bounds | 03:39 |
| vidd | adrianmoya, try setting up compute node as such: http://dpaste.com/646390/ | 03:39 |
| troya | implementation of it i mean | 03:39 |
| troya | what do you mean unstructured data, notmyname ? | 03:40 |
| notmyname | troya: documents, backups, etc. just blobs of bytes. unstructured as compared to a rdbms | 03:41 |
| notmyname | troya: use cases include storing backups and storing static web content. pretty much anything where you need large amounts of data reliably and cheaply stored | 03:42 |
| notmyname | troya: it's very similar to amazon s3, so if you're familiar with their product, swift is directly comparable (except with swift you get all the code to run it yourself for free) | 03:42 |
| notmyname | troya: swift could be a key part of document management, ISP caching layers, even useful for things like storing medical images from an mri or data from a space telescope | 03:43 |
| *** rods has quit IRC | 03:44 | |
| adrianmoya | vidd: tryed but not getting anywhere | 03:45 |
| adrianmoya | I still can't run nova-network | 03:45 |
| vidd | adrianmoya, im out of ideas =[ | 03:45 |
| notmyname | troya: swift isn't a filesystem. it doesn't meet POSIX requirements. it doesn't support ACID. it's a large-scale distributed storage system designed for high aggregate throughput, availability, durability, and price | 03:45 |
| vidd | adrianmoya, do you have nova-network installed on compute node? | 03:46 |
| *** adjohn has joined #openstack | 03:46 | |
| *** troya has quit IRC | 03:47 | |
| adrianmoya | vidd: thanks for your help, I'm also tired so I think I'll leave this for tomorrow... | 03:48 |
| vidd | adrianmoya, im sure its something small we are both overlooking =] | 03:49 |
| *** clopez has quit IRC | 03:49 | |
| *** hezekiah_ has joined #openstack | 03:49 | |
| *** helloworld has quit IRC | 03:50 | |
| *** hezekiah_ is now known as isaacfinnegan | 03:51 | |
| *** troya has joined #openstack | 03:52 | |
| *** neogenix has joined #openstack | 03:52 | |
| *** adrianmoya has quit IRC | 03:52 | |
| troya | hi all, can i install swift on single node? | 03:55 |
| vidd | troya, yes you can | 03:56 |
| notmyname | troya: there are instructions on how to set up an all-in-one VM with swift (http://swift.openstack.org/development_saio.html) | 03:56 |
| troya | oh..thanks all, i'll install it soon | 03:57 |
| troya | :) | 03:57 |
| notmyname | troya: but be aware that the saio is desinged for limited POC and (mainly) for dev work. not all of the parts of swift are exposed cleanly when running in one VM (for example, mounting virtual drives as loopback devices rather than having real hardware) | 03:57 |
| notmyname | troya: I'd certainly recommend running an instance of SAIO (swift all-in-one) to get your feet wet. but, depending on your needs, it may be a loose approximation of your final product | 03:58 |
| *** negronjl has joined #openstack | 04:00 | |
| *** nati2_ has joined #openstack | 04:00 | |
| vidd | does glance acually use a database? | 04:01 |
| isaacfinnegan | sqlite3 | 04:01 |
| *** nati2 has quit IRC | 04:02 | |
| vidd | i set it up to use a mysql database, but nothing has been added ti it =\ | 04:02 |
| vidd | my instances (seem to) work fine | 04:03 |
| *** worstadmin has quit IRC | 04:05 | |
| *** dendrobates is now known as dendro-afk | 04:08 | |
| *** vladimir3p has quit IRC | 04:08 | |
| *** dendro-afk is now known as dendrobates | 04:08 | |
| *** rsampaio has quit IRC | 04:09 | |
| *** nati2_ has quit IRC | 04:11 | |
| *** wariola has quit IRC | 04:11 | |
| troya | notmyname. i try to following instruction from http://swift.openstack.org/development_saio.html#partition-section, is partition which used for swift must LVM as when i configure nova volume ? | 04:11 |
| *** nati2 has joined #openstack | 04:12 | |
| *** wariola has joined #openstack | 04:12 | |
| notmyname | troya: will you be here in the channel tomorrow? | 04:13 |
| troya | hm..may be | 04:13 |
| notmyname | troya: a partition in swift is not the same thing as a filesystem partition. it's an unfortunately overloaded term | 04:14 |
| notmyname | but (in my timezone) it's late, and I was just heading to bed | 04:14 |
| isaacfinnegan | anyone try to disable the openstack nat to be able to route to instances directly? | 04:14 |
| notmyname | troya: I'd be happy to explain in more detail tomorrow | 04:14 |
| troya | notmyname: ok, i'll in channel tomorrow, see you :) | 04:14 |
| vidd | isaacfinnegan, like trying to ssh from the controller to the private ip? | 04:15 |
| isaacfinnegan | no. | 04:15 |
| isaacfinnegan | I want the fixed IP subnet to be routable. To be able to access instance directly on their fixed ip, and not have to assign a floating | 04:16 |
| *** nati2 has quit IRC | 04:16 | |
| isaacfinnegan | I have a router and have added an interface onthe backend 'fixed ip' subnet. I can route to it fine. but openstack sets up a bunch of iptables rules and basically forces all instance traffic to be nat'd | 04:17 |
| isaacfinnegan | whether or not a floating is used | 04:17 |
| *** jog0 has joined #openstack | 04:19 | |
| *** supriya has joined #openstack | 04:19 | |
| troya | vidd, can i installed swift on server which not supported VT ? | 04:20 |
| vidd | troya, i dont see why not | 04:22 |
| troya | so i can install Swift on it ? | 04:22 |
| vidd | i was able to install the entire openstack stack on machines that had no VT | 04:23 |
| vidd | troya, i assume you mean VT==Virtualization | 04:24 |
| *** troya_ has joined #openstack | 04:26 | |
| *** troya has quit IRC | 04:27 | |
| vidd | you can put all the non-VM-related services of openstack on machines without virtualization abbilities.... nova-volume, glance, keystone, MySQL server.....these dont actually do any virtualization, so they can be run on machines that dont have virtualiation cababilities | 04:27 |
| troya_ | vidd, what's better schema to install openstack between glance, nova, swift.I have idea to installed Nova + glance on Server 1, then installed Swift on Server 2.How about you ? in this case i have only 2 server | 04:28 |
| vidd | troya_, right now....i have only one server | 04:29 |
| troya_ | so all in one on it ? | 04:29 |
| *** jog0 has quit IRC | 04:29 | |
| vidd | i will put everything but swift on one server | 04:29 |
| *** supriya has quit IRC | 04:29 | |
| troya_ | so you will separate swift to another server, right ? | 04:30 |
| vidd | migrate as much of the data center infrastructure as i possibly can onto that one machine, and repurpose a few freed up machines to run additional compute nodes to free up the remaining machines, and then set up a swift array | 04:30 |
| *** negronjl has quit IRC | 04:31 | |
| troya_ | eh..vid, have you try Amazon ? | 04:32 |
| troya_ | i mean Aws ?, is it free trial ? | 04:32 |
| vidd | no...i dont want to do that....im trying to get un-laid-off! | 04:32 |
| troya_ | what do you mean un-laid-off? | 04:33 |
| vidd | i show the bosses tht amazon can do all this for less then i can do this...they wont need me =] | 04:33 |
| vidd | currently, im not getting a paycheck | 04:34 |
| troya_ | :)) | 04:34 |
| *** rnorwood has quit IRC | 04:35 | |
| troya_ | actually i want to make User layer presentation for use, so that they can build custom Cloud Infrastructure for their requirement, do you have suggestion to create it ? | 04:37 |
| vidd | im not sure what you mean...so no...no idea | 04:38 |
| troya_ | i mean, for example i am customer, i want to build my custom cloud infrasctructure.So i customize some parts as memory, storage, then others else.For that customization i wlll use Web Interface, right ?because it can make me easy t configure them | 04:42 |
| *** flamboz has quit IRC | 04:43 | |
| troya_ | then, how i can put that Web Interface for customiztion on above nova? | 04:43 |
| vidd | install dashboard i suppose | 04:44 |
| *** gyee has quit IRC | 04:44 | |
| troya_ | yes user dashboard, but with dashboard openstack can i custom for example memory or Hardisk storage for my cloud ? | 04:45 |
| vidd | have you looked at the dashboard project for openstack? | 04:46 |
| troya_ | i have looked at devstack | 04:47 |
| vidd | so you have seen the dashboard | 04:47 |
| vidd | is that what you are looking for? | 04:47 |
| *** MarkAtwood has joined #openstack | 04:48 | |
| *** worstadmin has joined #openstack | 04:49 | |
| troya_ | please wait some moment vid, i must to go now | 04:49 |
| *** thickskin has left #openstack | 04:56 | |
| *** miclorb_ has quit IRC | 04:56 | |
| vidd | do euca commands work with keystone? im getting "NotAuthorized: 401 Unauthorized" when trying to list images | 04:57 |
| *** chomping has joined #openstack | 04:58 | |
| vidd | do i need to add something? | 04:58 |
| *** v0id has quit IRC | 04:59 | |
| *** hugokuo has quit IRC | 05:01 | |
| *** worstadmin has quit IRC | 05:04 | |
| *** supriya has joined #openstack | 05:05 | |
| *** miclorb_ has joined #openstack | 05:10 | |
| *** obino has quit IRC | 05:12 | |
| *** vladimir3p has joined #openstack | 05:15 | |
| vidd | paste.filter_factory = keystone.middleware.ec2_token:EC2Token.factory ....where do i find/verify-i-have ec2_token? | 05:16 |
| vidd | 2011-11-03 01:10:19,340 CRITICAL nova [-] No module named ec2_token | 05:16 |
| vidd | i installed keystone via apt-get | 05:17 |
| uvirtbot | New bug: #885558 in devstack "apts and pips should not install if respective services are disabled" [Undecided,New] https://launchpad.net/bugs/885558 | 05:21 |
| *** wariola has quit IRC | 05:25 | |
| isaacfinnegan | it's werid | 05:27 |
| *** wariola has joined #openstack | 05:28 | |
| *** dysinger has joined #openstack | 05:28 | |
| vidd | isaacfinnegan, ? | 05:28 |
| isaacfinnegan | I want to enable routing to instances. directly to their fixed ip's | 05:29 |
| isaacfinnegan | It's not clear to me what's preventing packets from getting to the instances | 05:30 |
| isaacfinnegan | oh | 05:30 |
| isaacfinnegan | shoot | 05:30 |
| isaacfinnegan | hmmm | 05:30 |
| *** vladimir3p has quit IRC | 05:31 | |
| *** superjudge has joined #openstack | 05:33 | |
| *** flamboz has joined #openstack | 05:35 | |
| vidd | 2011-11-03 01:10:19,340 CRITICAL nova [-] No module named ec2_token | 05:37 |
| vidd | anyone know how to fix this...or where i should look to see where the keystone middleware is housed? | 05:37 |
| *** adjohn has quit IRC | 05:40 | |
| *** clayg has quit IRC | 05:41 | |
| *** 31NAADY1U is now known as clayg | 05:41 | |
| *** MarkAt2od has joined #openstack | 05:56 | |
| *** openpercept has joined #openstack | 05:57 | |
| *** MarkAtwood has quit IRC | 06:00 | |
| *** dysinger has quit IRC | 06:03 | |
| *** tokuz has joined #openstack | 06:03 | |
| *** joeyang has quit IRC | 06:06 | |
| *** nerens has joined #openstack | 06:09 | |
| *** cdub has joined #openstack | 06:10 | |
| *** jog0 has joined #openstack | 06:13 | |
| *** jog0 has quit IRC | 06:15 | |
| *** miclorb_ has quit IRC | 06:16 | |
| *** obino has joined #openstack | 06:17 | |
| *** vidd has quit IRC | 06:19 | |
| *** katkee has joined #openstack | 06:23 | |
| *** hugokuo has joined #openstack | 06:24 | |
| troya_ | vidd | 06:31 |
| *** vidd has joined #openstack | 06:38 | |
| *** jingizu_ has quit IRC | 06:39 | |
| *** kaigan_ has joined #openstack | 06:45 | |
| vidd | i have images taht i stored with euca commands...where are they stored? | 06:47 |
| vidd | i have glance set up and glance cant seem to find them and euca-describe-images cant find them anymore either | 06:48 |
| troya_ | vidd, must i create LVM partition to install swift ? | 06:51 |
| vidd | troya_, i do not know...i have not tried to install swift....what does the documentation tell you to do? | 06:52 |
| troya_ | i following this instruction >> http://swift.openstack.org/development_saio.html | 06:52 |
| troya_ | on that reference i must create new partition, but can i use filesystem partition? | 06:53 |
| vidd | do you not understand the instructions? | 06:56 |
| vidd | first you create a virtual machine | 06:56 |
| vidd | then you attach a volume to that virtual machine....just like any other time you attach a volume to a virtual machine | 06:57 |
| vidd | you can use whatever filesystem your volume can support i dont know if you can make a virtual lvm to be honest | 06:58 |
| *** andreaf has joined #openstack | 06:59 | |
| vidd | troya_, i think you should direct your questions to someone who understands swift | 06:59 |
| *** TheOsprey has joined #openstack | 07:01 | |
| *** Ryan_Lane has joined #openstack | 07:03 | |
| *** qwe890 has joined #openstack | 07:04 | |
| qwe890 | Hi everyone. Can anyone shed a light on why stackops.com/stackops.org is down and how much time it might take? I'm trying to reach it for about an hour now... | 07:07 |
| GheRivero | morning | 07:07 |
| GheRivero | qwe890, it looks like a power outage, they have one a couple of weeks ago | 07:08 |
| *** crescendo has joined #openstack | 07:08 | |
| *** crescendo has joined #openstack | 07:08 | |
| GheRivero | qwe890, if you just want to download the iso, the project is in sourceforge, http://sourceforge.net/projects/stackops/files/ | 07:09 |
| qwe890 | Too bad. I need their "Smart Installer". | 07:11 |
| vidd | qwe890, they are located in spain if thats any help =] | 07:11 |
| qwe890 | :) | 07:11 |
| vidd | does anyone know where euca puts images by default (when glance is not set up)? | 07:11 |
| wittwerch | vidd: I guess here under /var/lib/nova/images | 07:12 |
| *** jedi4ever has joined #openstack | 07:13 | |
| vidd | =\ nope | 07:13 |
| vidd | found them..... thanks wittwerch | 07:14 |
| vidd | you got me looking in the right direction =] | 07:14 |
| wittwerch | where was it? | 07:14 |
| vidd | /var/lib/nova/buckets | 07:14 |
| wittwerch | okey | 07:15 |
| wittwerch | I'm running nova diablo and I need a basic accounting. Does diablo has anything like that built in? | 07:15 |
| vidd | built in? no | 07:16 |
| wittwerch | is it possible to get that information from the api? | 07:16 |
| vidd | there is a nice drupal package though | 07:16 |
| vidd | im assuming you mean "this instance has been up this long and that image has been up that long...etc? | 07:17 |
| *** mnour has quit IRC | 07:19 | |
| vidd | wittwerch, check out http://drupal.org/project/cloud | 07:20 |
| vidd | euca-get-console-output might also be useful | 07:21 |
| *** hingo has joined #openstack | 07:23 | |
| wittwerch | thx for the link, looks interesting | 07:25 |
| wittwerch | I found the api which returns what I need | 07:25 |
| wittwerch | openstackx.extras.usage | 07:26 |
| wittwerch | but I have no clue how to use it.. | 07:26 |
| vidd | wittwerch, from what i understand, dashboard uses that | 07:30 |
| vidd | ok...i have untar'd an image that i like to use....its in 3 parts.... | 07:32 |
| vidd | how do i get it into glance and ...more importantly....how do i get glance to see the three parts are a package | 07:32 |
| vidd | the docs.openstack.org info reads like stereo instructions =\ | 07:33 |
| wittwerch | by 3 parts you mean kernel, disk and ramdisk? | 07:33 |
| vidd | yes... | 07:34 |
| vidd | vmlinuz, initrd, and blank.img | 07:34 |
| wittwerch | nova-manage image all_register --image=ami-maverick/image --kernel=aki-maverick/image --ram=ari-maverick/image --owner=myuser --name=simpleImage --public=T | 07:34 |
| vidd | wittwerch, i assume replace -maverick/image with cirros-0.3.0-x86_64-blank.img/image? | 07:36 |
| *** andreaf has quit IRC | 07:38 | |
| vidd | vmlinuz is the kernel,,,,is initrd the ram? | 07:39 |
| *** guigui has joined #openstack | 07:39 | |
| *** miclorb_ has joined #openstack | 07:40 | |
| *** nerens has quit IRC | 07:44 | |
| wittwerch | vidd, yes | 07:48 |
| vidd | wittwerch, i tried to do the command you stated, adjusting for my info, but it says they are not directories =\ | 07:49 |
| vidd | do i add them to glance individually, and then run nova-manage? | 07:51 |
| wittwerch | nova-manage uses glance as a backend | 07:52 |
| vidd | i feel like each piece i add puts me 3 weeks behind in my grasping the entire setup =\ | 07:53 |
| zykes- | hey vidd, don't you sleep ? ;p | 07:54 |
| zykes- | what have you added now vidd ? | 07:54 |
| vidd | everything worked fine until i put keystone on this | 07:54 |
| zykes- | :d | 07:54 |
| zykes- | that's the standard :d | 07:54 |
| vidd | zykes-, keystone | 07:54 |
| zykes- | what you broke now then ? :p | 07:54 |
| vidd | i actually turned glance on | 07:54 |
| zykes- | and you get which error ? | 07:54 |
| vidd | euca wont work | 07:55 |
| zykes- | k | 07:55 |
| *** crescendo has quit IRC | 07:55 | |
| vidd | the images i added with euca are no longer accessable | 07:55 |
| zykes- | if you do | 07:56 |
| vidd | i dont know if i broke nova by turning on keystone auth, or if euca just isnt supposed to work with keystone | 07:56 |
| zykes- | glance index is_public=False @ your glance host | 07:56 |
| zykes- | what do you get? | 07:56 |
| *** Ryan_Lane has quit IRC | 07:56 | |
| vidd | Not authorized to make this request. Check your credentials (OS_AUTH_USER, OS_AUTH_KEY, ...). | 07:57 |
| zykes- | ok then, glance -A <admin token> details | 07:58 |
| vidd | same thing | 07:58 |
| *** ejat has quit IRC | 07:59 | |
| vidd | now there is supposed to be a osauth script somewhere to "fix" this | 07:59 |
| zykes- | k | 08:00 |
| zykes- | then you know more then me ;) | 08:00 |
| *** ejat has joined #openstack | 08:01 | |
| *** ejat has joined #openstack | 08:01 | |
| *** crescendo has joined #openstack | 08:04 | |
| *** crescendo has joined #openstack | 08:04 | |
| *** isaacfinnegan has quit IRC | 08:07 | |
| *** jdag has quit IRC | 08:10 | |
| *** crescendo has quit IRC | 08:10 | |
| *** crescendo has joined #openstack | 08:11 | |
| *** crescendo has joined #openstack | 08:11 | |
| *** jdag has joined #openstack | 08:11 | |
| *** jedi4ever has quit IRC | 08:11 | |
| *** datajerk has quit IRC | 08:13 | |
| *** reidrac has joined #openstack | 08:14 | |
| *** krow has quit IRC | 08:14 | |
| *** jdag has left #openstack | 08:17 | |
| *** dobber has joined #openstack | 08:19 | |
| *** stewart has quit IRC | 08:25 | |
| vidd | zykes-, add https://github.com/openstack/glance/blob/master/tools/nova_to_os_env.sh info to your novarc (or .bashrc...whichever you source) | 08:25 |
| zykes- | ah | 08:26 |
| vidd | it is SUPPOSED to fix the auth issue | 08:26 |
| zykes- | but still vidd glance -A <adm token> should work as well | 08:26 |
| vidd | of course, in my situation, that is a major NOVEMBER-GOLF | 08:26 |
| zykes- | if the user + pw doesn't work | 08:26 |
| vidd | no go | 08:26 |
| vidd | so my keystone must be broken | 08:27 |
| zykes- | checked glance logs ? | 08:27 |
| *** datajerk has joined #openstack | 08:29 | |
| vidd | zykes-, glance/api is blank...never been touched | 08:32 |
| vidd | registry has no error | 08:33 |
| *** miclorb_ has quit IRC | 08:33 | |
| *** Hakon|mbp has quit IRC | 08:34 | |
| *** javiF has joined #openstack | 08:36 | |
| vidd | zykes-, any ideas? | 08:40 |
| zykes- | not sure vidd, working on my dns baby atm ;p | 08:42 |
| *** jedi4ever has joined #openstack | 08:48 | |
| *** Razique has joined #openstack | 08:49 | |
| *** supriya has quit IRC | 08:49 | |
| vidd | zykes-, wgat ya doiing with DNS? | 08:50 |
| *** nacx has joined #openstack | 08:51 | |
| zykes- | vidd: alla reddwarf service for dns | 08:52 |
| *** mandela123 has quit IRC | 08:53 | |
| *** stewart has joined #openstack | 08:54 | |
| *** katkee has quit IRC | 08:55 | |
| *** mnour has joined #openstack | 08:56 | |
| vidd | now why the heck does my nova network keep crashing?!? | 08:58 |
| *** joeyang has joined #openstack | 08:59 | |
| *** uksysadmin has joined #openstack | 08:59 | |
| uksysadmin | hello all | 09:00 |
| Razique | hi all :) | 09:00 |
| Razique | zykes-: http://docs.openstack.org/diablo/openstack-compute/admin/content/migrating-from-cactus-to-diablo.html | 09:00 |
| Razique | We just updated the migration process ! | 09:01 |
| uksysadmin | Quick q (isn't it always?) on floating IPs... do I always have to issue a command like "euca-associate-address" to assign a floating IP to my instances or is there a way of assigning this all the time? | 09:01 |
| Razique | http://docs.openstack.org/trunk/openstack-compute/admin/content/migrating-from-cactus-to-diablo.html | 09:01 |
| Razique | uksysadmin: the novaclient tool does make directly the association | 09:01 |
| Razique | while euca2ools don't | 09:01 |
| *** nid0 has joined #openstack | 09:02 | |
| troya_ | hi razique | 09:02 |
| Razique | hi ! | 09:02 |
| uksysadmin | cheers Razique - will check it out (and bring my toolset into the 21st Century at the same time eh?) ;-) | 09:02 |
| *** tokuz has quit IRC | 09:03 | |
| *** Hakon|mbp has joined #openstack | 09:03 | |
| *** yeming has quit IRC | 09:03 | |
| vidd | Razique, you use keystone? | 09:03 |
| troya_ | razique: if i publish image .tar.gz with command for example >> uecpublishtarball ubuntu11.04servercloudimg amd64.tar.gz imagebucket , so if i have .iso file how do it ? | 09:04 |
| uksysadmin | OK, I give up - where do I find nova tools? | 09:05 |
| uksysadmin | and how do I get juju and the like to do this? (did I say it was a quick question?) | 09:06 |
| troya_ | vidd: if i publish image .tar.gz with command for example >> uecpublishtarball ubuntu11.04servercloudimg amd64.tar.gz imagebucket , so if i have .iso file how do it ? | 09:06 |
| *** datajerk has quit IRC | 09:06 | |
| vidd | troya_, from what i understand, you need to have glance configured to use .iso's | 09:07 |
| Razique | vidd: I don't use it today | 09:07 |
| Razique | because I run an upgrade from cactus | 09:07 |
| *** flamboz has quit IRC | 09:07 | |
| Razique | troya_: is it a distro ISO ? | 09:07 |
| troya_ | yes, like ubuntu 11.04 server.iso, like that | 09:08 |
| Razique | troya_: if so, you need to install it via KVM first an "cloudify" it :p | 09:08 |
| Razique | troya_: ok, then here is what you need :http://cssoss.wordpress.com/2011/04/27/openstack-beginners-guide-for-ubuntu-11-04-image-management/ | 09:08 |
| vidd | Razique, i was wondering if nova.conf needs a replacement for the --use-depreciated-auth to something about keystone | 09:08 |
| *** flamboz has joined #openstack | 09:08 | |
| Razique | vidd: if you remove the flag, you'll also make sure that Keystone is connected to nova :) | 09:10 |
| *** dendrobates is now known as dendro-afk | 09:10 | |
| vidd | Razique, the --use-depriciated-auth is turned off | 09:10 |
| Razique | vidd: if you simply remove the line, nova-api assumes you are using Keystone then :) | 09:11 |
| vidd | uksysadmin, you might want to look into the --auto_assign_floating_ip flag for nova http://docs.openstack.org/diablo/openstack-compute/admin/content/reference-for-flags-in-nova-conf.html | 09:11 |
| Razique | so don't forget to install the middleware and pipelines (see api-paste.ini) | 09:11 |
| vidd | so the --auth_driver (if missing) is set to keystone? | 09:12 |
| uksysadmin | cheers vidd - sounds like what I'm looking for | 09:12 |
| *** nikolaij has joined #openstack | 09:12 | |
| vidd | Razique, yes...i went to the cloudbuilder's git site to get the pipelines and middleware (ubuntu repo's don include the examples dir) | 09:13 |
| Razique | vidd: so read the last update ; I describe the implementation | 09:14 |
| vidd | Razique, it seems like im missing something because ever since i installed keystone, my nova-network keeps crashing | 09:14 |
| Razique | vidd: here http://docs.openstack.org/trunk/openstack-compute/admin/content/migrating-from-cactus-to-diablo.html | 09:14 |
| Razique | check "D- Why is Keystone not integrated ?" | 09:14 |
| vidd | =] | 09:14 |
| *** datajerk has joined #openstack | 09:14 | |
| vidd | loading page now | 09:14 |
| *** e1mer has quit IRC | 09:16 | |
| *** MarkAt2od has quit IRC | 09:21 | |
| *** e1mer has joined #openstack | 09:22 | |
| *** pixelbeat has joined #openstack | 09:23 | |
| *** miclorb_ has joined #openstack | 09:26 | |
| *** jedi4ever has quit IRC | 09:26 | |
| *** jedi4ever has joined #openstack | 09:26 | |
| *** Charz has joined #openstack | 09:26 | |
| *** LanceHaig has left #openstack | 09:28 | |
| *** dendro-afk is now known as dendrobates | 09:29 | |
| uksysadmin | vidd, am I correct in assuming that there must be some sort of dhcp service listening on my "public" interface to hand out these auto_assigned_floating_ips ? (does seem a dumb question, but wanted to make sure before I spin a dhcp service up on there...) | 09:29 |
| *** darraghb has joined #openstack | 09:31 | |
| vidd | Razique, is the api-paste correct? your example has them set to "noauth" while the examples folder has totoken authtoken keystonecontext" and the like | 09:31 |
| Razique | vidd: in my examples Keystone is bypasses | 09:31 |
| vidd | uksysadmin, if i understand it correctly, that flag IS the dhcp server | 09:31 |
| vidd | in the section D | 09:32 |
| Razique | but I think you are right, I should correct it in order to activate it | 09:32 |
| *** e1mer has quit IRC | 09:33 | |
| vidd | Razique, while you are making changes...you have "nova-api.conf" instead of "nova.conf" | 09:33 |
| uksysadmin | ok - I'll do some digging. I presume its dnsmasq that does this job? | 09:34 |
| vidd | for a second i thought i needed another config file | 09:34 |
| uksysadmin | I get exceptions, ultimately Floating ip not found for address <nova.db.sqlalchemy.models.FloatingIp object at 0x40e3e50>. | 09:34 |
| *** siwos has joined #openstack | 09:35 | |
| vidd | uksysadmin, i think our frend nova-network that does this | 09:35 |
| uksysadmin | yes, ta - that's where the exception is logged | 09:35 |
| vidd | uksysadmin, you still have to allacate the IP's | 09:35 |
| Razique | vidd: thanks, I'll update that | 09:36 |
| *** miclorb__ has joined #openstack | 09:36 | |
| *** stevegjacobs_ has joined #openstack | 09:37 | |
| uksysadmin | yeah I can associate manually | 09:38 |
| *** miclorb_ has quit IRC | 09:38 | |
| *** adjohn has joined #openstack | 09:40 | |
| *** adjohn has quit IRC | 09:40 | |
| *** e1mer has joined #openstack | 09:41 | |
| *** e1mer has joined #openstack | 09:41 | |
| vidd | uksysadmin, so the ip's are alrady allocated? | 09:41 |
| vidd | have you tried to deallocate a few? | 09:42 |
| uksysadmin | vidd, I have no instances running. Prior to adding the auto flag to true I can spin up instances and manually associate addresses. Since adding in the flag I get the following: http://paste.openstack.org/show/3053/ | 09:42 |
| vidd | so you have a pool of "ready" and a pool of "reserve" ips...so you can see which nova allocutes | 09:43 |
| uksysadmin | I set my quota to 1024 (ruling out a previous error of exceeded allocation) | 09:43 |
| uksysadmin | I have allocated public IPs | 09:43 |
| uksysadmin | if I remove the flag and restart my services I can spin up machines ok and allocate addresses manually | 09:44 |
| vidd | your quota is what? 10? | 09:44 |
| uksysadmin | 1024 | 09:44 |
| uksysadmin | to avoid any contentions | 09:44 |
| uksysadmin | (artificially high) | 09:44 |
| Razique | uksysadmin: does nvoa-manage floating list shows the ip ? | 09:45 |
| vidd | set 5 to allocated and 5 to not-allocated, put up the flag and see if it pulls an IP and if it does, see if it pulls one you have allocated or one you have not-allocated | 09:45 |
| Razique | otherwise create them via nova-manage floating create IP/CIDR | 09:45 |
| uksysadmin | it shows private as my private ip, public as the private ip and state pending because of the error | 09:45 |
| uksysadmin | I've no problems allocating addresses manually - do I need to do that command again? | 09:46 |
| uvirtbot | New bug: #885595 in swift "swift-init use /etc/swift without a way to change it" [Undecided,New] https://launchpad.net/bugs/885595 | 09:46 |
| vidd | uksysadmin, no...DONT allocate | 09:46 |
| uksysadmin | yes - I'm saying I CAN before I added in the flag to suggest I've no issues manually doing it | 09:46 |
| uksysadmin | I can't do it now because the box is stranded | 09:46 |
| vidd | i think the error is saying it is trying to allocate for you and has none to pull from | 09:46 |
| uksysadmin | well in pending | 09:46 |
| uksysadmin | I've allocated addresses | 09:47 |
| uksysadmin | its not auto associating | 09:47 |
| vidd | deallocate them | 09:47 |
| *** katkee has joined #openstack | 09:47 | |
| uksysadmin | I've nothing allocated | 09:47 |
| uksysadmin | I have no machines running at all | 09:47 |
| vidd | what does euca-describe-addresses return? anything...or a whole slue? | 09:47 |
| Razique | uksysadmin: you mean by updating the database ? | 09:47 |
| Razique | or via euca-associate | 09:48 |
| *** adrian17od has joined #openstack | 09:48 | |
| *** _ulli has joined #openstack | 09:49 | |
| uksysadmin | its shows a few but with "None (myproject)" | 09:49 |
| uksysadmin | I'm deallocating | 09:49 |
| uksysadmin | and starting again | 09:50 |
| *** e1mer has quit IRC | 09:50 | |
| *** ulli has quit IRC | 09:51 | |
| Razique | uksysadmin: are you using keysotne ? | 09:51 |
| *** bonzay is now known as zz_bonzay | 09:53 | |
| uksysadmin | Razique, no, not for this set up at the mo. | 09:54 |
| *** e1mer has joined #openstack | 09:55 | |
| *** e1mer has joined #openstack | 09:55 | |
| *** chomping has quit IRC | 09:56 | |
| *** clopez has joined #openstack | 09:56 | |
| vidd | why is nova-network running this command "Command: sudo ip addr add 192.168.15.224/32 scope global dev br100"? | 09:56 |
| uksysadmin | ahh, vidd and Razique - when I launch my instance, a new address public IP pops up (when I view the addresses) - its the associating stage its failing. What relationship does it need between the this allocated address to what its just spun up? | 09:58 |
| vidd | what is the new error? | 09:59 |
| *** obino has quit IRC | 09:59 | |
| *** chomping has joined #openstack | 10:00 | |
| uksysadmin | same as before: FloatingIpNotFoundForAddress: Floating ip not found for address <nova.db.sqlalchemy.models.FloatingIp object at 0x4524f10>. | 10:00 |
| *** clauden has quit IRC | 10:00 | |
| uksysadmin | I know I'm missing a stage - but what I'm expecting is that an instance spins up, it gets a private IP and it auto assigns a public IP so I can then access it | 10:00 |
| uksysadmin | as the story says behind this request: just like AWS EC2 | 10:01 |
| *** obino has joined #openstack | 10:01 | |
| *** supriya has joined #openstack | 10:05 | |
| *** po has joined #openstack | 10:05 | |
| *** alekibango has quit IRC | 10:06 | |
| *** qwe890 has quit IRC | 10:07 | |
| *** alekibango has joined #openstack | 10:08 | |
| vidd | SUCCESS! i fixed my "nova-network keeps crashing" issue | 10:09 |
| *** Oneiroi has joined #openstack | 10:11 | |
| *** GheRivero has quit IRC | 10:14 | |
| *** GheRivero has joined #openstack | 10:17 | |
| *** jedi4ever has quit IRC | 10:26 | |
| *** jedi4ever has joined #openstack | 10:26 | |
| *** troya_ has quit IRC | 10:27 | |
| Razique | vidd: great | 10:28 |
| Razique | what happended ? | 10:28 |
| Razique | vidd: it creates the bridge IP I think | 10:28 |
| vidd | i didnt have the --host flag set | 10:29 |
| vidd | and either keystone or dashboard demands it | 10:29 |
| Razique | uksysadmin: the difference here is that nova makes a difference between private and public ip | 10:29 |
| vidd | i installed them both at the same time so not sure which one broke it | 10:29 |
| uksysadmin | yes - so does ec2 - it has a private IP known to the AZ, and a public one that's routeable - which genuinely is public | 10:30 |
| Razique | uksysadmin: could you paste your nova.conf and show me both SQL requests result : SELECT* FROM floatings_ips and SELECT * FROM fixed_ips | 10:30 |
| uksysadmin | I'm doing the same - a non-routeable private IP I can't connect to, but the public one is a routed address on my LAN | 10:30 |
| uksysadmin | will do | 10:31 |
| Razique | uksysadmin: ohhh ok, sorry I wasn't aware of that | 10:31 |
| Razique | uksysadmin: from the nova server can you reach the private ip ? | 10:31 |
| uksysadmin | not at this stage - it fails to launch because it bombs out trying to associate the public ip | 10:32 |
| uksysadmin | the table entries looks odd but could be holding the clue... | 10:33 |
| uksysadmin | http://paste.openstack.org/show/3055/ | 10:34 |
| uksysadmin | lots of NULLs | 10:34 |
| vidd | anyone know what the endpoint template for dashboard is supposed to be? | 10:35 |
| *** stevegjacobs_ has quit IRC | 10:35 | |
| *** stevegjacobs has quit IRC | 10:35 | |
| Razique | vidd: you sure Dashboard has it's own endpoint ? | 10:39 |
| Razique | Dashboard should point to the Keystone one | 10:39 |
| Razique | uksysadmin: that the issue here :) | 10:40 |
| Razique | fixed ip are not linked to ur current project | 10:40 |
| Razique | sorry | 10:41 |
| Razique | the floating don't have an ip | 10:41 |
| Razique | fixed_ip_id, you should have IDs here | 10:41 |
| vidd | Razique, nothing else makes sence with this keystone =\ | 10:41 |
| Razique | vidd: when you configure the dashbord, the URL to set is http://KEYSTONE-IP:5000 | 10:42 |
| Razique | 500/v2.0/ would be better | 10:42 |
| vidd | Razique, i need something production-grade with a GUI so the standard dashboard-in-a-venv isnt going to cut it =\ | 10:43 |
| vidd | im going to have to hack it into place =] | 10:43 |
| Razique | vidd: you mean update the dashboard for your needs ? | 10:44 |
| vidd | yeah...and give the improvements back to the community =] | 10:45 |
| Razique | ok ok | 10:46 |
| Razique | great | 10:46 |
| Razique | so basically DASH ---> KEYSTONE ENDPOINT ----> SERVICES | 10:46 |
| *** dolphm has joined #openstack | 10:47 | |
| vidd | yes...with a billing plug-in into keystone so we can suspend people that dont pay thier bills on time | 10:47 |
| vidd | that should be relatively easy...set a user or a tenent to "inactive"should kill the credentials they us, but i also need to set up some way to have the dashboard role get switch from "member" to "suspended" | 10:49 |
| vidd | which is why i will need [to create if need be] an endpoint for dashboard | 10:50 |
| vidd | would not do to have a user who is suspended be able to access the tools to un-suspend themselves =] | 10:51 |
| *** dolphm has quit IRC | 10:51 | |
| uksysadmin | Yeah Razique as soon as I saw that it didn't make sense - its saying for this project I have no IPs to assign. I'll try fudging the database and then see why its not populating the correct fields | 10:51 |
| Razique | uksysadmin: would you run that for me ? nova-manage floating create 192.168.1.1/32 | 10:52 |
| uksysadmin | sure | 10:53 |
| Razique | and select * from floating_ips where fixed_ip_id != NULL | 10:53 |
| uksysadmin | Empty set | 10:54 |
| Razique | ok | 10:54 |
| Razique | nova-manage project list | 10:54 |
| Razique | nova-manage network list | 10:54 |
| *** stewart has quit IRC | 10:55 | |
| *** adrian17od has quit IRC | 11:01 | |
| *** stewart has joined #openstack | 11:05 | |
| *** Razique_ has joined #openstack | 11:07 | |
| *** stewart has quit IRC | 11:08 | |
| uksysadmin | nova-manage project list... myproject | 11:08 |
| *** stewart has joined #openstack | 11:09 | |
| uksysadmin | network list | 11:09 |
| uksysadmin | id IPv4 IPv6 start address DNS1 DNS2 VlanID project uuid | 11:09 |
| uksysadmin | 1 10.0.0.0/26 None 10.0.0.3 None None 100 myproject None | 11:09 |
| *** Razique has quit IRC | 11:09 | |
| *** Razique_ is now known as Razique | 11:09 | |
| uksysadmin | my public ip range is 192.168.241.0/24 | 11:09 |
| uksysadmin | (0,1 and 2 are unusable - that's just a shortcoming of my environment at the moment but doesn't affect the associating of addresses beyond this) | 11:10 |
| uksysadmin | now I remember having an issue with floating ips a while back in that the project column wasn't being populated - back then the create floating command accepted a project | 11:11 |
| uksysadmin | it doesn't now | 11:11 |
| uksysadmin | (running Diablo on Ocelot) | 11:12 |
| Razique | uksysadmin: ok nova nova-manage floating create 192.168.241.50/32 && tail -100 /var/log/nova-manage.log | 11:14 |
| *** dolphm has joined #openstack | 11:17 | |
| uksysadmin | successful - only 2 lines created in nova-manage.log | 11:17 |
| uksysadmin | 2011-11-03 11:17:09,267 DEBUG nova.utils [-] backend <module 'nova.db.sqlalchemy.api' from '/usr/lib/python2.7/dist-packages/nova/db/sqlalchemy/api.pyc'> from (pid=4193) __get_backend /usr/lib/python2.7/dist-packages/nova/utils.py:450 | 11:17 |
| uksysadmin | 2011-11-03 11:17:09,324 INFO nova.db.sqlalchemy [-] Using mysql/eventlet db_pool. | 11:17 |
| *** stewart has quit IRC | 11:18 | |
| * uksysadmin is appreciating the help from Razique here - if I get over this I can demonstrate some really interesting things here | 11:18 | |
| *** naehring has joined #openstack | 11:18 | |
| *** stewart has joined #openstack | 11:19 | |
| uksysadmin | looking at the db, project isn't being populated - reckon that's my issue? | 11:20 |
| *** bsza has joined #openstack | 11:20 | |
| *** stewart has quit IRC | 11:21 | |
| uksysadmin | Razique, I'm busy now most of today - think I've a couple of things I can start investigating and trying. | 11:23 |
| uksysadmin | cheers for your help | 11:23 |
| *** uksysadmin has quit IRC | 11:24 | |
| *** ahasenack has joined #openstack | 11:24 | |
| *** jedi4ever has quit IRC | 11:26 | |
| *** jedi4ever has joined #openstack | 11:26 | |
| *** markvoelker has joined #openstack | 11:33 | |
| uvirtbot | New bug: #885634 in nova "VM state of guest invalid when host shutdown" [Undecided,New] https://launchpad.net/bugs/885634 | 11:36 |
| uvirtbot | New bug: #885635 in nova "Host reboot renders host status invalid" [Undecided,New] https://launchpad.net/bugs/885635 | 11:36 |
| *** nerdstein has joined #openstack | 11:36 | |
| *** miclorb__ has quit IRC | 11:40 | |
| vidd | Razique, you still here? | 11:41 |
| *** ninkotech has joined #openstack | 11:41 | |
| *** ninkotech has quit IRC | 11:46 | |
| *** dnjaramba has joined #openstack | 11:47 | |
| *** dolphm has quit IRC | 11:47 | |
| *** hadrian has joined #openstack | 11:50 | |
| *** Razique has quit IRC | 11:53 | |
| *** ejat has quit IRC | 11:58 | |
| *** livemoon has joined #openstack | 12:00 | |
| *** osier has quit IRC | 12:06 | |
| *** qwe890 has joined #openstack | 12:07 | |
| *** e1mer has quit IRC | 12:13 | |
| *** msivanes has joined #openstack | 12:16 | |
| *** cereal_bars has joined #openstack | 12:17 | |
| *** e1mer has joined #openstack | 12:18 | |
| *** ogelbukh has joined #openstack | 12:19 | |
| *** npmapn has joined #openstack | 12:19 | |
| *** dolphm has joined #openstack | 12:22 | |
| *** jedi4ever has quit IRC | 12:26 | |
| *** jedi4ever_ has joined #openstack | 12:26 | |
| vidd | zykes-, you still around? | 12:27 |
| *** hggdh has joined #openstack | 12:28 | |
| *** lorin1 has joined #openstack | 12:28 | |
| *** cdub_ has joined #openstack | 12:29 | |
| *** crescendo has quit IRC | 12:29 | |
| *** cdub has quit IRC | 12:29 | |
| livemoon | hi,vidd | 12:35 |
| livemoon | nice to see you | 12:35 |
| vidd | hello livemoon | 12:35 |
| vidd | i have gotten the thumbs up for purchasing equiptment | 12:36 |
| livemoon | congratulaion | 12:36 |
| livemoon | congratulation | 12:37 |
| vidd | going to convert as much of the data center to openstack compute as possible on one server...then repurpose freed machines to virtulize the rest =] | 12:37 |
| livemoon | :) | 12:37 |
| livemoon | just one server? | 12:37 |
| vidd | to start | 12:38 |
| vidd | we have a few under-utilized machines i want to frr up | 12:38 |
| *** PeteDaGuru has joined #openstack | 12:38 | |
| vidd | then make them into compute nodes | 12:38 |
| livemoon | good | 12:39 |
| livemoon | good luck | 12:39 |
| livemoon | I will go to bed | 12:39 |
| vidd | hopefully, i can free up enough machines to get everything virtualized and have machines left over to do a decent swift array =] | 12:39 |
| livemoon | these two days I have a traning outdoor | 12:39 |
| zykes- | hey vidd | 12:41 |
| zykes- | yeah in between | 12:41 |
| *** crescendo has joined #openstack | 12:42 | |
| vidd | so zykes- i have this issue.... | 12:42 |
| *** supriya has quit IRC | 12:42 | |
| vidd | with keystone installed glance seams to repond but euca wont co-operate | 12:42 |
| vidd | do your euca tools work with keystone? | 12:43 |
| *** e1mer has quit IRC | 12:43 | |
| *** livemoon has left #openstack | 12:43 | |
| *** dolphm has quit IRC | 12:45 | |
| *** iRTermite has joined #openstack | 12:45 | |
| *** iRTermite has quit IRC | 12:45 | |
| *** hggdh has quit IRC | 12:47 | |
| *** dprince has joined #openstack | 12:47 | |
| *** qwe890 has left #openstack | 12:48 | |
| *** e1mer has joined #openstack | 12:48 | |
| *** e1mer has joined #openstack | 12:48 | |
| vidd | ===========-oTTo^| | 12:49 |
| vidd | zykes-, any idea what i should be looking for? | 12:51 |
| zykes- | yeah | 12:51 |
| zykes- | what version of ks you running ? | 12:51 |
| vidd | keystone-manage 0.9 | 12:52 |
| zykes- | hmm, i run a more recent one then that | 12:52 |
| zykes- | though i don't think it should affect it | 12:52 |
| zykes- | do you get any access denied errors in your glance-api.log ? | 12:52 |
| zykes- | or in the nova-api | 12:52 |
| zykes- | when you use euca* it goes to nova which auths first with keystone which then goes to glance which also auths you with keystone fyi | 12:53 |
| vidd | 2011-11-03 07:48:27,176 INFO nova.api.openstack.wsgi [-] http://192.168.15.199:8774/favicon.ico returned with HTTP 300 | 12:53 |
| *** e1mer has quit IRC | 12:53 | |
| zykes- | so that's not the EC2 API, look for requests on port 8773 | 12:53 |
| zykes- | euca* doesn't use the OSAPI vidd | 12:54 |
| vidd | zykes-, that might-could explain it =\ | 12:55 |
| *** e1mer has joined #openstack | 12:57 | |
| *** beekhof_muc has joined #openstack | 12:58 | |
| *** lorin1 has quit IRC | 12:58 | |
| beekhof_muc | sandywalsh: will the orchestrator meetings be at a regular time? | 12:58 |
| *** lborda has joined #openstack | 12:58 | |
| zykes- | basically vidd a euca image req for example would be: client > keystone middleware in front of nova > nova api > glance keystone middleware > glance | 12:59 |
| zykes- | so your euca command for getting images can fail at either the nova api point of auth or at the glance middleware | 12:59 |
| vidd | export EC2_URL="http://192.168.15.199:5000/services/Cloud" | 12:59 |
| vidd | that is pointing to keystone | 12:59 |
| zykes- | nope, EC2_URL="http://192.168.15.199:8773/services/Cloud | 13:00 |
| vidd | im not trying to get an image just yet | 13:00 |
| *** dolphm has joined #openstack | 13:00 | |
| zykes- | that's how it should be | 13:00 |
| vidd | im just trying to allocate an ip (so i can just test nova) | 13:00 |
| *** supriya has joined #openstack | 13:00 | |
| zykes- | it's only for swift if I recall it correct where you need to pass the keystone url to the client | 13:00 |
| zykes- | then use the url i gave you ;p | 13:00 |
| zykes- | think that's correct at least | 13:01 |
| vidd | now its taking too long | 13:02 |
| *** hggdh has joined #openstack | 13:03 | |
| *** lorin1 has joined #openstack | 13:04 | |
| *** supriya has quit IRC | 13:06 | |
| vidd | Warning: failed to parse error message from AWS: <unknown>:1:0: not well-formed (invalid token) | 13:09 |
| *** huslage has joined #openstack | 13:11 | |
| *** Oneiroi has quit IRC | 13:11 | |
| *** Razique has joined #openstack | 13:11 | |
| Razique | vidd: | 13:11 |
| Razique | yup back from lunch | 13:11 |
| Razique | is uksysadmin around ? | 13:11 |
| vidd | zykes-, port for nova is 8774 | 13:12 |
| vidd | Warning: failed to parse error message from AWS: <unknown>:1:0: not well-formed (invalid token) | 13:12 |
| vidd | Razique, no, uksysadmin when home | 13:12 |
| vidd | or work, or someplace =] | 13:12 |
| Razique | vidd: seems like the endpoint is not well written :) | 13:12 |
| Razique | vidd: u are using euca2ools, right? | 13:13 |
| vidd | yes | 13:13 |
| Razique | ok can I see the .bashrc ? | 13:13 |
| sandywalsh | beekhof, :) regular for who? | 13:13 |
| sandywalsh | beekhof, or do you mean 'will they be regular events?' | 13:14 |
| *** iRTermite has joined #openstack | 13:14 | |
| *** bcwaldon has joined #openstack | 13:16 | |
| vidd | Razique, http://dpaste.com/646702/ | 13:16 |
| *** supriya has joined #openstack | 13:16 | |
| Razique | I think it should be http://192.168.15.199:5000/v2.0/ | 13:17 |
| Razique | and into the nova.conf : -keystone_ec2_url=http://172.16.40.11:5000/v2.0/ec2tokens | 13:17 |
| *** lborda is now known as lborda_otp | 13:18 | |
| vidd | Razique, still a no-go ...http://dpaste.com/646707/ | 13:21 |
| Razique | vidd: so let the line into nova.conf | 13:21 |
| Razique | but redo the changes into the .bashrc | 13:21 |
| Razique | I had the same issues | 13:22 |
| Razique | issue | 13:22 |
| Razique | it's definitely linked to that :) | 13:22 |
| vidd | ok...kill bash.... | 13:22 |
| vidd | edit novarc.... | 13:23 |
| vidd | source novarc | 13:25 |
| vidd | restart everything | 13:25 |
| *** e1mer has quit IRC | 13:26 | |
| vidd | >-|O | | 13:26 |
| *** jedi4ever_ has quit IRC | 13:26 | |
| vidd | still same issue | 13:26 |
| *** jedi4ever has joined #openstack | 13:26 | |
| *** rods has joined #openstack | 13:27 | |
| vidd | Razique, could it be the endpoint in keystone DB? | 13:27 |
| Razique | oh wait | 13:27 |
| Razique | check that file = /usr/local/lib/python2.6/dist-packages/keystone-1.0-py2.6.egg/keystone/middleware/ec2_token.py | 13:28 |
| *** mmetheny has quit IRC | 13:28 | |
| *** mmetheny_ has joined #openstack | 13:28 | |
| *** jtanner has joined #openstack | 13:28 | |
| Razique | do u have that line o = urlparse(FLAGS.keystone_ec1_url) | 13:28 |
| vidd | i have no keystone eggs | 13:29 |
| vidd | ' File: /usr/share/pyshared/keystone/middleware/ec2_token.py | 13:30 |
| *** e1mer has joined #openstack | 13:31 | |
| vidd | Razique, is that in the paste api? or in the ec2_token | 13:31 |
| Razique | ec2_token | 13:32 |
| Razique | run updatedb && locate ec2_token.py | 13:32 |
| *** dolphm has quit IRC | 13:32 | |
| vidd | got it...and yes....i have that line | 13:32 |
| vidd | lemme guess...change it to ec2? | 13:33 |
| *** supriya has quit IRC | 13:35 | |
| zykes- | change what vidd ? | 13:36 |
| vidd | the 0= flag in ec2_token file | 13:37 |
| *** kieron has joined #openstack | 13:37 | |
| *** hggdh has quit IRC | 13:37 | |
| Razique | vidd: nope | 13:37 |
| Razique | here | 13:37 |
| Razique | # o = urlparse(FLAGS.keystone_ec1_url) | 13:38 |
| Razique | o = urlparse(FLAGS.keystone_ec2_url) | 13:38 |
| *** jwalcik has joined #openstack | 13:38 | |
| Razique | and | 13:38 |
| Razique | # token_id = result['auth']['token']['id'] | 13:38 |
| Razique | token_id = result['access']['token']['id'] | 13:38 |
| *** javiF has quit IRC | 13:39 | |
| *** dolphm has joined #openstack | 13:39 | |
| *** e1mer has quit IRC | 13:41 | |
| *** katkee has quit IRC | 13:41 | |
| *** jeh has joined #openstack | 13:43 | |
| vidd | Razique, still no goode | 13:44 |
| Razique | service nova-api restart after that | 13:44 |
| Razique | you have left the directive into nova.conf ? | 13:44 |
| *** imsplitbit has joined #openstack | 13:44 | |
| vidd | that "updatedb" is that a keystone-manage command? | 13:44 |
| Razique | vidd: nope | 13:44 |
| vidd | Razique, yes...keytone directive is left in | 13:44 |
| Razique | keystone-manage populates the database if it's empty | 13:44 |
| vidd | so just updatedb? | 13:45 |
| Razique | ok so now put back http://nova-api:8773/services/Cloud | 13:46 |
| Razique | and I think it should work :) | 13:46 |
| *** e1mer has joined #openstack | 13:46 | |
| *** e1mer has joined #openstack | 13:46 | |
| vidd | but back....where? | 13:46 |
| vidd | the novarc? | 13:46 |
| *** dolphm has quit IRC | 13:46 | |
| Razique | yup | 13:47 |
| Razique | EC2_URL | 13:47 |
| *** dolphm has joined #openstack | 13:48 | |
| vidd | now an error 500 =\ | 13:49 |
| vidd | must be the capital C in cloud | 13:49 |
| vidd | nope...no good either....both trow syntax errors | 13:50 |
| Razique | vidd: nope, that means nova doesn't seem able to forward the request to keystone | 13:51 |
| *** e1mer has quit IRC | 13:51 | |
| Razique | tail /var/log/nova/nova-api.log | 13:51 |
| vidd | http://dpaste.com/646717/ | 13:52 |
| vidd | thats with lowwrcase c | 13:52 |
| vidd | and Capital C gives a 500 responce | 13:53 |
| vidd | waiting to complte error out so i can paiste the tai;l | 13:53 |
| vidd | http://dpaste.com/646719/ capital C | 13:54 |
| *** e1mer has joined #openstack | 13:55 | |
| Razique | weird, since I use myself the capital C and it's working | 13:55 |
| Razique | do you see keystone logs moving ? | 13:56 |
| Razique | that would means nova-api forwards the request to keystone | 13:56 |
| Razique | also, into keystone DB do u have the EC2 entry ? | 13:56 |
| Razique | (under credentials table) | 13:56 |
| vidd | http://dpaste.com/646724/ | 13:57 |
| vidd | yes...made the EC2 entry | 13:57 |
| vidd | the script in your docs put ' marks around the so i had to take them out....and the values didnt match the database structure | 13:58 |
| *** PiotrSikora has quit IRC | 13:58 | |
| *** PiotrSikora has joined #openstack | 13:59 | |
| *** katkee has joined #openstack | 13:59 | |
| vidd | is there a keystone.conf setting we are missing? | 14:00 |
| *** benrodrigue has joined #openstack | 14:00 | |
| *** huslage has quit IRC | 14:00 | |
| *** jamespage_ has joined #openstack | 14:02 | |
| Razique | error 401, that means an auth error | 14:03 |
| Razique | what euca2ools outputs on it's side ? | 14:03 |
| vidd | nova-api? | 14:04 |
| *** katkee has quit IRC | 14:04 | |
| vidd | http://dpaste.com/646727/ | 14:05 |
| uvirtbot | New bug: #885696 in glance "glance checked_iter missing variable in error case, mixed type of expected_size" [Undecided,New] https://launchpad.net/bugs/885696 | 14:06 |
| Razique | vidd: nope, what's the error when you run for instance euca-describe-instances ? | 14:08 |
| vidd | well right now its timing out because last tim i had the capital C | 14:10 |
| vidd | Warning: failed to parse error message from AWS: <unknown>:1:0: syntax error | 14:10 |
| *** imsplitbit has quit IRC | 14:11 | |
| *** mnour has quit IRC | 14:12 | |
| vidd | http://dpaste.com/646729/ with the lowecase c (with --debug flag) | 14:12 |
| *** catintheroof has joined #openstack | 14:13 | |
| Razique | vidd: ok weirk | 14:13 |
| Razique | $ | 14:13 |
| Razique | would you let me start my VM and reactivate keystone + ec2 | 14:13 |
| *** Madkiss has quit IRC | 14:13 | |
| *** Madkiss_ has joined #openstack | 14:13 | |
| *** Madkiss_ is now known as Madkiss | 14:13 | |
| Razique | right now I'm kinda busy, writting docs :D and I'd like to stay focused | 14:13 |
| Razique | since I finish that, 'ill check | 14:13 |
| *** benrodrigue has quit IRC | 14:14 | |
| vidd | Razique, sure...i need to get some sleep anyway | 14:14 |
| *** tryggvil_ has joined #openstack | 14:14 | |
| vidd | been at this for over 19 hours | 14:14 |
| Razique | vidd: gime your city, 'ill add you into my world clock | 14:14 |
| Razique | so I know when I can reach you | 14:14 |
| vidd | new york time zone.... | 14:14 |
| vidd | but im usually here between 6pm to 9am | 14:15 |
| vidd | local time | 14:15 |
| Razique | ok, good to know | 14:15 |
| vidd | im a night person =] | 14:16 |
| Razique | I see that ^^ | 14:16 |
| *** joesavak has joined #openstack | 14:16 | |
| vidd | maybe i live on tokeyo time =] | 14:16 |
| Razique | I'm trying not to be | 14:16 |
| Razique | hehe | 14:16 |
| vidd | me too...but this project will get me un-laid-off so ive been putting in some 36-hour days on it | 14:17 |
| Razique | vidd: ok 'ill look for the integration | 14:17 |
| Razique | update the doc accordingly | 14:18 |
| Razique | and let u know when it's ok! :) | 14:18 |
| *** e1mer has quit IRC | 14:18 | |
| vidd | the wife actually told me to go to the casino a couple weeks ago to get a break | 14:18 |
| vidd | im using version 0.9 if that helps any | 14:18 |
| vidd | the one that comes in 11.10 repos | 14:19 |
| *** sandywalsh_ has joined #openstack | 14:19 | |
| *** lorin1 has joined #openstack | 14:20 | |
| *** Barakas has joined #openstack | 14:20 | |
| Razique | 0.9 ? | 14:20 |
| vidd | root@vidd-host:~# keystone-manage --version | 14:21 |
| vidd | keystone-manage 0.9 | 14:21 |
| vidd | its the one packaged in ubuntu repos | 14:21 |
| *** jsavak has joined #openstack | 14:22 | |
| *** imsplitbit has joined #openstack | 14:23 | |
| *** e1mer has joined #openstack | 14:23 | |
| *** jedi4ever_ has joined #openstack | 14:25 | |
| * vidd is off to see the sand-man | 14:25 | |
| *** jedi4ever has quit IRC | 14:25 | |
| *** jedi4ever_ is now known as jedi4ever | 14:25 | |
| *** vidd is now known as vidd-away | 14:26 | |
| *** joesavak has quit IRC | 14:26 | |
| fjay | when bringing up openstack for the first time.. does it need its own network(s)? | 14:26 |
| zykes- | Razique: care to read a etherpad on something ? | 14:26 |
| Razique | zykes-: sure | 14:26 |
| Razique | fjay: nope, it could be integrated into an existing network | 14:27 |
| fjay | Razique: can it leverage an exisiting dhcp server? | 14:27 |
| fjay | Razique: the doc's kinda say it wants to run its own and don't really say much | 14:27 |
| zykes- | fjay: nova nodes and "service" nodes can, instances is maybe a bit more tricky in some cases | 14:27 |
| *** mattray has joined #openstack | 14:28 | |
| fjay | i have a small ~4 server setup i am trying to roll | 14:28 |
| fjay | just to get our dev env more like production | 14:29 |
| *** zul has joined #openstack | 14:29 | |
| fjay | and i don't really need (or want) the instances on their own network | 14:29 |
| fjay | but i could live w/ it. | 14:29 |
| fjay | just alot more work | 14:29 |
| *** sugar_punkin has joined #openstack | 14:29 | |
| fjay | are there any doc's or references to running it 'flat' like that? | 14:29 |
| *** sugar_punkin has quit IRC | 14:30 | |
| *** sugar_punkin has joined #openstack | 14:30 | |
| *** imsplitbit has quit IRC | 14:31 | |
| *** supriya has joined #openstack | 14:31 | |
| *** katkee has joined #openstack | 14:32 | |
| fjay | i guess worst case i add another vlan to the network and put eth1 on that vlan or just trunk to the machines | 14:34 |
| gnu111 | Can someone look at my settings for Diablo in RHEL? I was finally able to get a centos image bootup and ssh to it, but can't ping outside from it: http://paste.openstack.org/show/3056/ | 14:37 |
| *** gmih has joined #openstack | 14:37 | |
| *** neogenix has quit IRC | 14:38 | |
| *** dubsquared has joined #openstack | 14:38 | |
| *** e1mer has quit IRC | 14:39 | |
| *** rnirmal has joined #openstack | 14:41 | |
| *** jfluhmann has joined #openstack | 14:41 | |
| *** cp16net has joined #openstack | 14:42 | |
| *** e1mer has joined #openstack | 14:44 | |
| *** code_franco has joined #openstack | 14:47 | |
| *** clayg_ has joined #openstack | 14:47 | |
| *** supriya has quit IRC | 14:47 | |
| gnu111 | I thought this bug was fixed in Diablo: https://bugs.launchpad.net/nova/+bug/844935 | 14:48 |
| *** siwos has quit IRC | 14:48 | |
| *** zul has quit IRC | 14:48 | |
| gmih | has anybody here managed to get Nova+Dashboard+Keystone working in a multi-node setup running on Ubuntu 11.10 from the official Ubuntu repo? | 14:48 |
| annegentle | gmih: as far as I know, it won't work from the official Ubuntu repo | 14:50 |
| *** dnjaramba has quit IRC | 14:50 | |
| *** dnjaramba_ has joined #openstack | 14:50 | |
| *** dendrobates is now known as dendro-afk | 14:50 | |
| Razique | gmih: annegentle is right here, you'll need at least a github keystone and dashboard | 14:52 |
| *** jamespage_ has quit IRC | 14:53 | |
| gmih | Razique: thanks for reply; I would like to use the official github, not the one maintained by cloudbuilders, would it work? | 14:55 |
| *** huslage has joined #openstack | 14:55 | |
| Razique | gmih: yup | 14:57 |
| Razique | we published a doc for that : http://docs.openstack.org/trunk/openstack-compute/admin/content/migrating-from-cactus-to-diablo.html | 14:57 |
| Razique | section D | 14:57 |
| Razique | i'm updating it, since vidd gave me some feedbacks, but you can start with that :) | 14:58 |
| *** ayates83 has left #openstack | 14:58 | |
| Razique | with that, you'll just need after that to follow the standard procedure for the Dash | 14:58 |
| *** adjohn has joined #openstack | 15:00 | |
| gmih | unfortunatelly, I already blown away my Cactus cluster and started from scratch with Diablo on Ubuntu 11.10 | 15:03 |
| *** hezekiah_ has joined #openstack | 15:03 | |
| *** hggdh has joined #openstack | 15:04 | |
| *** dragondm has joined #openstack | 15:05 | |
| *** javiF has joined #openstack | 15:05 | |
| uvirtbot | New bug: #885721 in horizon "RAM quota incorrect" [Undecided,New] https://launchpad.net/bugs/885721 | 15:05 |
| *** clayg has quit IRC | 15:06 | |
| *** Gordonz has joined #openstack | 15:06 | |
| *** clayg_ is now known as clayg | 15:06 | |
| *** clayg_ has joined #openstack | 15:06 | |
| *** f-yang has joined #openstack | 15:09 | |
| *** fred_ has joined #openstack | 15:09 | |
| *** fred_ is now known as Guest89527 | 15:09 | |
| Glacee | Is there a way to get more logging in swift? Debug level logs? | 15:10 |
| *** neogenix has joined #openstack | 15:12 | |
| *** MarkAtwood has joined #openstack | 15:13 | |
| *** f-yang has quit IRC | 15:13 | |
| *** Guest89527 has quit IRC | 15:13 | |
| *** f-yang has joined #openstack | 15:14 | |
| Glacee | nm found it | 15:16 |
| *** jdg has joined #openstack | 15:16 | |
| uvirtbot | New bug: #885723 in nova "Xen failure on resize should set instance to ERROR" [Undecided,In progress] https://launchpad.net/bugs/885723 | 15:16 |
| Glacee | What are in other options of log level though? I just see INFO in the docs | 15:17 |
| Glacee | I am gessing that DEBUG would be an option? | 15:18 |
| *** jdg has quit IRC | 15:19 | |
| *** ldlework has joined #openstack | 15:19 | |
| *** rnorwood has joined #openstack | 15:20 | |
| *** superjudge has quit IRC | 15:21 | |
| *** lionel has quit IRC | 15:21 | |
| *** e1mer has quit IRC | 15:21 | |
| *** lionel has joined #openstack | 15:21 | |
| *** jdg__ has joined #openstack | 15:22 | |
| *** rsampaio has joined #openstack | 15:23 | |
| *** jedi4ever has quit IRC | 15:24 | |
| *** jedi4ever has joined #openstack | 15:24 | |
| *** Gordonz has quit IRC | 15:25 | |
| *** Gordonz has joined #openstack | 15:25 | |
| *** adjohn has quit IRC | 15:27 | |
| *** dolphm has quit IRC | 15:27 | |
| *** hggdh has quit IRC | 15:29 | |
| *** lorin1 has quit IRC | 15:29 | |
| *** lorin1 has joined #openstack | 15:29 | |
| *** dolphm has joined #openstack | 15:30 | |
| doude | Hi stackers, I try to set up a multi-region deployment with OpenStack | 15:31 |
| *** hezekiah_ is now known as isaacfinnegna | 15:31 | |
| *** isaacfinnegna is now known as isaacfinnegan | 15:31 | |
| *** anon_os has joined #openstack | 15:31 | |
| *** benrodrigue has joined #openstack | 15:31 | |
| doude | I get an error when I add a zone to another one. The parent zone failed to pull added child. | 15:32 |
| *** dtroyer has quit IRC | 15:33 | |
| doude | Parent zone asked keystone with URL http://keystone.url:5000/v2.0/tokens with admin acount but without tenant ID | 15:33 |
| *** cdub_ has quit IRC | 15:33 | |
| *** sgg has joined #openstack | 15:34 | |
| *** mnour has joined #openstack | 15:34 | |
| dolphm | doude: does it look like a nova or keystone issue? | 15:34 |
| doude | So it fails to parse the service catalog because Keystone respond that http://paste.openstack.org/show/3058/ (No service catalog) | 15:35 |
| *** sgd2 has joined #openstack | 15:35 | |
| doude | I don't know if it's keystone issue or Nova issue | 15:35 |
| doude | dolphm: I don't know if it's keystone issue or Nova issue | 15:35 |
| dolphm | doude: i believe that's expected on keystone's behalf -- let me double check... what versions of nova & keystone are you running? | 15:37 |
| doude | I found a way to fix that by edditing the file 'nova/scheduler/zone_manager.py' and specifie the tenant ID to the novaclient request (line 103) | 15:37 |
| *** beekhof_muc has quit IRC | 15:38 | |
| dolphm | doude: that's what i'm thinking -- i'm not sure a service catalog is expected if no tenant is provided... the token keystone responds with is basically a token so you can go back to keystone and call GET /tenants with it, choose a tenant, POST /tokens again (this time with a tenant in mind), and you should get back a service catalog | 15:38 |
| doude | dolphm: I use the trunk version rev11312 | 15:38 |
| dolphm | doude: so perhaps nova is expecting to bypass that process by assuming you know your tenant? | 15:38 |
| dolphm | doude: which is fine, if you do | 15:38 |
| doude | dolphm: so it's a nova issue. The zone manager should find a tenant ID. | 15:39 |
| *** sgd21 has joined #openstack | 15:39 | |
| doude | dolphm: or sepecifed it | 15:40 |
| *** marrusl has joined #openstack | 15:40 | |
| dolphm | doude: i believe that's correct | 15:40 |
| *** clopez has quit IRC | 15:40 | |
| *** lborda_otp has quit IRC | 15:40 | |
| doude | dolphm: should I open a bug ? | 15:41 |
| *** clopez has joined #openstack | 15:41 | |
| *** sgd2 has quit IRC | 15:41 | |
| *** mnour has quit IRC | 15:42 | |
| *** lborda has joined #openstack | 15:43 | |
| doude | dolphm: Other question, I don't understand why admin account need to be linked to a tenant ? There is a specific reason for that ? | 15:44 |
| *** lorin1 has quit IRC | 15:45 | |
| *** lorin1 has joined #openstack | 15:46 | |
| dolphm | doude: always feel free to open bugs :) | 15:46 |
| dolphm | doude: that shouldn't be the case -- and the idea of a "default tenant" is being completely removed | 15:47 |
| *** marrusl has quit IRC | 15:51 | |
| jeh | What package contains the stack command referred to at http://docs.openstack.org/trunk/openstack-compute/admin/content/getting-an-instances-vnc-console.html | 15:51 |
| *** clauden_ has joined #openstack | 15:52 | |
| dolphm | doude: i say default tenant is being removed, and now i'm reading a review that improves support for default tenants.. | 15:54 |
| *** dendro-afk is now known as dendrobates | 15:55 | |
| *** dtroyer has joined #openstack | 15:57 | |
| vishy | jeh: stack is not packaged | 15:57 |
| *** openpercept has quit IRC | 15:59 | |
| *** dendrobates is now known as dendro-afk | 15:59 | |
| jeh | vishy: can I get a vnc_console_id without it? | 16:00 |
| jeh | vishy: and the obvious next question, where can I get the stack command source? | 16:01 |
| *** sgd21 is now known as sgd2 | 16:02 | |
| *** sugar_punkin has quit IRC | 16:02 | |
| vishy | jeh: stack is in the nova source tree | 16:02 |
| vishy | jeh: there is an extension for getting a vnc console | 16:02 |
| vishy | right now it is in extras controller in openstackx | 16:02 |
| vishy | it hasn't been moved into an extension in the nova tree yet | 16:02 |
| jeh | vishy: ok, thanks | 16:02 |
| *** juddm has joined #openstack | 16:03 | |
| vishy | github.com/cloudbuilders/openstackx | 16:03 |
| sgd2 | Is there a reason why I shouldn't specify my public IPs as nova fixed IPs? It seems simpler to use one network. | 16:03 |
| vishy | sgd2: if you can set up routing to work that way and you have enough public ips, go for it | 16:04 |
| *** dirakx1 has joined #openstack | 16:04 | |
| vishy | you should probably make nova give out an external gateway in that case | 16:04 |
| vishy | and you might have to set up some forwards on your gateway to get your metadata to work | 16:05 |
| *** dirakx2 has joined #openstack | 16:05 | |
| *** dirakx1 has quit IRC | 16:06 | |
| sgd2 | vishy, For a testbed environment, where I wouldn't need to access instances from outside, does that simplify routing? | 16:06 |
| *** TheOsprey has quit IRC | 16:06 | |
| vishy | zykes-: yes dnsaas would be great | 16:06 |
| vishy | sure | 16:06 |
| vishy | sgd2: just remember nova-network is acting as a gateway by default | 16:07 |
| *** dirakx1 has joined #openstack | 16:07 | |
| *** dirakx2 has quit IRC | 16:07 | |
| sgd2 | vishy: Does it provide DHCP and DNS itself? | 16:07 |
| vishy | yes | 16:08 |
| vishy | it uses host dns resolver | 16:08 |
| vishy | dnsmasq | 16:08 |
| *** dirakx1 has quit IRC | 16:09 | |
| *** dirakx1 has joined #openstack | 16:09 | |
| sgd2 | It seems I would want nova to run inside a vlan, so as not to conflict with our existing DHCP. | 16:09 |
| *** reed has joined #openstack | 16:09 | |
| uvirtbot | New bug: #885750 in keystone "Support CRUD calls for ec2credentials as per contract." [Medium,Confirmed] https://launchpad.net/bugs/885750 | 16:11 |
| *** zul has joined #openstack | 16:12 | |
| *** freeflyi1g has quit IRC | 16:13 | |
| *** jog0 has joined #openstack | 16:13 | |
| *** negronjl_mobile has joined #openstack | 16:13 | |
| *** cloudgeek has joined #openstack | 16:14 | |
| *** reidrac has quit IRC | 16:14 | |
| *** freeflying has joined #openstack | 16:15 | |
| *** jedi4ever has quit IRC | 16:18 | |
| *** jedi4ever has joined #openstack | 16:19 | |
| uvirtbot | New bug: #885755 in nova "Nova and Keystone don't work in multi-zone mode" [Undecided,New] https://launchpad.net/bugs/885755 | 16:21 |
| *** foexle has joined #openstack | 16:21 | |
| dolphm | doude: thanks for the feedback! | 16:21 |
| cloudgeek | hey all | 16:22 |
| foexle | heyho guys, ive a problem with euca-describe-images .... i get EC2ResponseError: 300 Multiple Choices from the openstack api .... how can i deactivate the depricated version | 16:22 |
| foexle | ? | 16:22 |
| *** obino has quit IRC | 16:23 | |
| dolphm | foexle: do you know which call is returning a 300? | 16:23 |
| foexle | my EC2_URL="http://xxx.xxx.xxx.xxx:8774/services/Cloud" and nova url http://xxx.xxx.xxx.xxx:8774/v1.1/ | 16:24 |
| *** datajerk has quit IRC | 16:25 | |
| foexle | hi dolphm : http://pastebin.com/SccLeeW4 | 16:25 |
| doude | dolphm: Ok, thanks for your answers. Can you point me the review ? A bug was open for zone problem: https://bugs.launchpad.net/nova/+bug/885755 | 16:26 |
| foexle | dolphm: i get the xml response with 2 choices | 16:26 |
| dolphm | foexle: two choices for nova, i assume? | 16:26 |
| dolphm | doude: review for..? | 16:27 |
| foexle | dolphm: yep | 16:27 |
| *** negronjl_mobile has quit IRC | 16:29 | |
| foexle | dolphm: i set the correct version in keystone and in my sysvars | 16:29 |
| vishy | foexle: use port 8773 | 16:30 |
| foexle | dolphm: so i need to resolve this issue to upload uec image to glance :=) | 16:30 |
| dolphm | doude: foexle: i'm not a nova guy so i'd just be guessing :/ there are several active nova devs in #openstack-dev at the moment if ya'll want to try there | 16:30 |
| vishy | for ec2 url | 16:30 |
| foexle | vishy: i'll try | 16:30 |
| dolphm | or ask vishy :) | 16:30 |
| *** datajerk has joined #openstack | 16:31 | |
| foexle | vishy: thx :) .... ok now i get other error .... but i think thats an other problem.... i'm so stupid -.- i'm sorry | 16:32 |
| *** dobber has quit IRC | 16:32 | |
| *** mgoldmann has joined #openstack | 16:34 | |
| *** n8 has joined #openstack | 16:36 | |
| *** n8 is now known as Guest23723 | 16:36 | |
| *** Guest23723 is now known as n88 | 16:37 | |
| n88 | can anybody here explain to me the process that occurs when using flatDHCP networking and how the bridge is setup? | 16:37 |
| vishy | n88: http://unchainyourbrain.com/openstack/13-networking-in-nova | 16:38 |
| vishy | has a diagram showing the flags | 16:38 |
| vishy | hopefully that helps | 16:38 |
| n88 | vishy: thanks I've actually modeled my network off that diagram; but I'm trying to understand when the bridge is actually created…when the first VM starts? when nova-compute starts? When nova-network starts? | 16:39 |
| *** masom has joined #openstack | 16:39 | |
| vishy | first vm | 16:39 |
| *** jdurgin has joined #openstack | 16:39 | |
| n88 | vishy: thanks! Same for both nova-compute and nova-volume? | 16:40 |
| vishy | if there are existing vms then it will be recreated when nova-network boots | 16:40 |
| masom | Hello, i am having a weird issue with keystone, adding a user crashes with 'NoneType' object has no attribute 'id'. Anyone know a workaround or which version of keystone will work the the dashboard? | 16:40 |
| isaacfinnegan | I cannot get a remote nova-compute node to start an instnace | 16:40 |
| vishy | in non multi_node it will be created on compute side when nova-compute starts as well | 16:40 |
| isaacfinnegan | it only works on the nova-api node (also running nova-compute) | 16:40 |
| n88 | vishy: So what happens if I'm in non-multi-mode, and my nova-compute starts and doesn't create the bridge? | 16:41 |
| vishy | isaacfinnegan: do you get an error message? | 16:41 |
| vishy | n88: do you have any vms defined? | 16:41 |
| isaacfinnegan | yes | 16:41 |
| isaacfinnegan | 2011-11-03 16:34:27,996 ERROR nova.exception [-] Uncaught exception | 16:41 |
| isaacfinnegan | (nova.exception): TRACE: Traceback (most recent call last): | 16:41 |
| isaacfinnegan | (nova.exception): TRACE: File "/usr/lib/python2.7/dist-packages/nova/exception.py", line 98, in wrapped | 16:41 |
| isaacfinnegan | (nova.exception): TRACE: return f(*args, **kw) | 16:41 |
| isaacfinnegan | (nova.exception): TRACE: File "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 454, in run_instance | 16:41 |
| isaacfinnegan | (nova.exception): TRACE: self._run_instance(context, instance_id, **kwargs) | 16:41 |
| isaacfinnegan | (nova.exception): TRACE: File "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 371, in _run_instance | 16:41 |
| isaacfinnegan | (nova.exception): TRACE: _check_image_size() | 16:41 |
| isaacfinnegan | (nova.exception): TRACE: File "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 327, in _check_image_size | 16:41 |
| isaacfinnegan | (nova.exception): TRACE: image_meta = image_service.show(context, image_id) | 16:41 |
| isaacfinnegan | (nova.exception): TRACE: File "/usr/lib/python2.7/dist-packages/nova/image/glance.py", line 220, in show | 16:41 |
| vishy | isaacfinnegan: use paste.openstack.org | 16:41 |
| n88 | vishy: no VMs yet…I just got all the components installed and setup and configured | 16:41 |
| vishy | isaacfinnegan: looks like your second node is having trouble talking to glance | 16:42 |
| *** cereal_bars has quit IRC | 16:42 | |
| n88 | vishy: And I don't see any bridges on either my nova-network or nova-compute machines so I wanted to make sure things were working right before I tried launching a VM | 16:42 |
| vishy | isaacfinnegan: do you have glance_api_servers set to an ip that is reachable? | 16:42 |
| isaacfinnegan | http://paste.openstack.org/show/3061/ | 16:42 |
| isaacfinnegan | yes | 16:42 |
| vishy | n88: it should all be created on first launch | 16:42 |
| isaacfinnegan | I can telnet to the galnce api / port and it connects | 16:42 |
| isaacfinnegan | from the other nova-compute node | 16:42 |
| n88 | isaac: I was having the same except problem about two days ago…I never resolved it and then my nova-compute stopped working altogether | 16:43 |
| *** jog0 has left #openstack | 16:43 | |
| vishy | isaacfinnegan: do you have the same version of glance on both machines? | 16:43 |
| isaacfinnegan | glance is not on both machine | 16:43 |
| isaacfinnegan | s | 16:43 |
| *** jog0 has joined #openstack | 16:43 | |
| isaacfinnegan | is it required for all nova-compute nodes? | 16:43 |
| vishy | isaacfinnegan: well glance client is required | 16:44 |
| vishy | and it is installed | 16:44 |
| vishy | but it could be a different version | 16:44 |
| isaacfinnegan | ok | 16:44 |
| *** alekibango has quit IRC | 16:44 | |
| vishy | isaacfinnegan: are you installing from packages? | 16:45 |
| isaacfinnegan | ok. I might have an old verison of python-glance | 16:45 |
| isaacfinnegan | udpating now | 16:45 |
| isaacfinnegan | that's probably it | 16:46 |
| uvirtbot | New bug: #885767 in devstack "nova-network get installed when I have ENABLED_SERVCIES="mysql key swift"" [Undecided,New] https://launchpad.net/bugs/885767 | 16:46 |
| isaacfinnegan | looks like I had nova-compute @ 2011.3 but python-glance @ 2011.2 | 16:46 |
| *** sgg has quit IRC | 16:46 | |
| *** cloudgeek has quit IRC | 16:46 | |
| *** Barakas has quit IRC | 16:46 | |
| masom | Anyone knows the proper keystone/dashboard combo? keystone master is having trouble with it's db | 16:47 |
| dolphm | masom: what kind of db trouble? | 16:47 |
| isaacfinnegan | that was it | 16:47 |
| isaacfinnegan | thx | 16:47 |
| masom | dolphm: 'NoneType' object has no attribute 'id' when attempting to add users | 16:48 |
| isaacfinnegan | oops | 16:48 |
| isaacfinnegan | spoke too soon. failed later | 16:48 |
| *** jedi4ever has quit IRC | 16:49 | |
| *** zul has quit IRC | 16:52 | |
| dolphm | masom: was your db created from a pre-diablo version of keystone? | 16:52 |
| dolphm | masom: or do you know where 'NoneType' is being referenced? (do you have a back trace?) | 16:54 |
| *** Hakon|mbp has quit IRC | 16:55 | |
| *** Razique has quit IRC | 16:56 | |
| n88 | vishy: thank you! I finally got my first openstack VM up and running!!!!! | 16:56 |
| vishy | hoyoo! | 16:56 |
| n88 | =D | 16:56 |
| *** sgd2 has quit IRC | 16:56 | |
| n88 | vishy: one additional question…when it comes to the floating ips, do I need to use the euca-* commands to allocate a public IP or is there some setting in nova.conf I can use for it to automatically assign a public IP to a running instance on startup from the available floating ip pool? | 16:57 |
| * pixelbeat is away: Away | 16:57 | |
| vishy | there is an --auto_assign_floating_ip | 16:58 |
| vishy | I think it still works | 16:58 |
| n88 | vishy: ok. I will test that out… | 16:58 |
| *** guigui has quit IRC | 16:59 | |
| *** huslage has quit IRC | 17:01 | |
| masom | dolphm: i'm looking at the backtrace. I deleted the previous keystone db just in case. | 17:01 |
| n88 | vishy: I successfully did the euca-allocate-address and see in the DB it's now associated with project1 (my project name) but when I try to use euca-associate-address <instance id> I get this error: These required options are missing: instance | 17:02 |
| masom | ~paste | 17:02 |
| masom | oh no bot... | 17:02 |
| vishy | that is not the command | 17:02 |
| vishy | euca-associate-address -i <instance_id> <address> | 17:02 |
| masom | dolphm: http://paste.openstack.org/show/3063/ | 17:02 |
| n88 | vishy: ah ok…it wasn't giving me a good error message…thanks as always | 17:02 |
| *** reed has quit IRC | 17:05 | |
| *** ashp has joined #openstack | 17:05 | |
| ashp | Hey guys, when I run: | 17:05 |
| ashp | # euca-upload-bundle -b kernel-bucket -m //tmp/ttylinux-uec-amd64-12.1_2.6.35-22_1-vmlinuz.manifest.xml | 17:05 |
| *** mnour has joined #openstack | 17:06 | |
| ashp | I get: ImportError: cannot import name Connection | 17:06 |
| ashp | Anyone have any idea what might be going wrong? | 17:06 |
| *** llang629 has joined #openstack | 17:08 | |
| *** alekibango has joined #openstack | 17:08 | |
| *** Ryan_Lane has joined #openstack | 17:10 | |
| *** n1md4 has joined #openstack | 17:11 | |
| *** rods has quit IRC | 17:12 | |
| *** llang629 has quit IRC | 17:14 | |
| *** dolphm has quit IRC | 17:16 | |
| uvirtbot | New bug: #885778 in nova "nova-manage not executable" [Undecided,Fix committed] https://launchpad.net/bugs/885778 | 17:16 |
| *** naehring has quit IRC | 17:17 | |
| BuZZ-T | does nova-api currently support euca-import-keypair? | 17:18 |
| *** gyee has joined #openstack | 17:19 | |
| *** jdurgin has quit IRC | 17:19 | |
| *** stuntmachine has joined #openstack | 17:20 | |
| *** rods has joined #openstack | 17:20 | |
| *** Gordonz has quit IRC | 17:20 | |
| *** llang629 has joined #openstack | 17:20 | |
| *** katkee has quit IRC | 17:21 | |
| *** gyee has quit IRC | 17:21 | |
| *** Hakon|mbp has joined #openstack | 17:23 | |
| *** llang629 has left #openstack | 17:23 | |
| *** nphase has joined #openstack | 17:24 | |
| *** dirakx1 has quit IRC | 17:24 | |
| *** dirakx1 has joined #openstack | 17:25 | |
| *** dirakx2 has joined #openstack | 17:25 | |
| *** dirakx1 has quit IRC | 17:26 | |
| *** lorin1 has quit IRC | 17:26 | |
| *** shentonfreude has joined #openstack | 17:27 | |
| *** Ryan_Lane has quit IRC | 17:27 | |
| agy | do the messages cast onto the message queue have an expiry? i.e. if a compute node receives a message from 48 hours in the past will the message still be processed or discarded? | 17:27 |
| *** nikolaij has quit IRC | 17:28 | |
| *** mnour has quit IRC | 17:28 | |
| *** shentonfreude has left #openstack | 17:29 | |
| *** Ryan_Lane has joined #openstack | 17:29 | |
| *** dirakx2 has quit IRC | 17:30 | |
| *** tryggvil_ has quit IRC | 17:30 | |
| *** jedi4ever has joined #openstack | 17:30 | |
| *** tryggvil_ has joined #openstack | 17:30 | |
| *** lorin1 has joined #openstack | 17:32 | |
| *** joesavak has joined #openstack | 17:32 | |
| *** Gordonz has joined #openstack | 17:34 | |
| *** jsavak has quit IRC | 17:36 | |
| *** jedi4ever has quit IRC | 17:36 | |
| *** jsavak has joined #openstack | 17:37 | |
| *** bengrue has quit IRC | 17:37 | |
| *** joesavak has quit IRC | 17:38 | |
| *** jdurgin has joined #openstack | 17:38 | |
| *** bengrue has joined #openstack | 17:38 | |
| ashp | So close, I got the instance to try and launch but now it's just stuck in scheduling forever. | 17:38 |
| *** dendro-afk is now known as dendrobates | 17:38 | |
| n88 | vishy: question about networking…what happens to the fixed_ips when you reach the last one, does the assignment start back at the first IP? I noticed in the database that new instances get the next IP address even if the previous fixed_IP was assigned to a terminated instance | 17:40 |
| *** imsplitbit has joined #openstack | 17:41 | |
| *** neogenix_ has joined #openstack | 17:41 | |
| ashp | I think my issue is it can't connect to libvirtd. Is there some kind of authentication you have to setup between nova-libvirt? | 17:41 |
| *** obino has joined #openstack | 17:42 | |
| ashp | hmm, no auth is required by default with libvirtd :/ | 17:44 |
| ashp | virsh can connect so it's definitely running. :/ | 17:45 |
| *** neogenix has quit IRC | 17:45 | |
| *** neogenix_ is now known as neogenix | 17:45 | |
| *** df1 has joined #openstack | 17:45 | |
| *** bengrue has quit IRC | 17:47 | |
| *** gyee has joined #openstack | 17:47 | |
| *** rnorwood1 has joined #openstack | 17:47 | |
| *** rnorwood has quit IRC | 17:49 | |
| *** dprince has quit IRC | 17:55 | |
| *** reed has joined #openstack | 17:55 | |
| *** jfluhmann has quit IRC | 17:55 | |
| *** TheOsprey has joined #openstack | 17:55 | |
| *** cereal_bars has joined #openstack | 17:56 | |
| *** jdg__ has quit IRC | 17:56 | |
| *** datajerk has quit IRC | 17:58 | |
| *** rnorwood1 has quit IRC | 17:58 | |
| *** dnjaramba_ has quit IRC | 17:59 | |
| *** nacx has quit IRC | 17:59 | |
| *** rnorwood has joined #openstack | 18:00 | |
| *** dendrobates is now known as dendro-afk | 18:00 | |
| *** adjohn has joined #openstack | 18:01 | |
| *** dolphm has joined #openstack | 18:03 | |
| *** datajerk has joined #openstack | 18:04 | |
| *** krow has joined #openstack | 18:04 | |
| *** dolphm_ has joined #openstack | 18:05 | |
| *** tlocalhos has quit IRC | 18:06 | |
| *** lorin1 has quit IRC | 18:06 | |
| *** dolphm has quit IRC | 18:07 | |
| n88 | has anyone seen an issue with iSCSI naming when trying to attach a volume? | 18:07 |
| *** tlocalhos has joined #openstack | 18:08 | |
| *** smeier00 has joined #openstack | 18:08 | |
| n88 | it appears the device.py file is assuming my iscsi mounted disk will have -lun-0 appended to it | 18:08 |
| n88 | but it's actual -lun-1 | 18:09 |
| *** gyee has quit IRC | 18:09 | |
| *** gyee has joined #openstack | 18:09 | |
| *** gyee has quit IRC | 18:09 | |
| vishy | n88: they are recyclied via timeout (the fixed_ips) | 18:09 |
| vishy | if you are on oneiric there is a flag which will release ips immediately | 18:10 |
| vishy | (or if you are using the nova-ppa for dependencies on an older version of ubuntu | 18:10 |
| n88 | visy: I am on oneiric — do you know the flag? | 18:11 |
| vishy | n88: --force_dhcp_release | 18:12 |
| *** huslage has joined #openstack | 18:12 | |
| *** gyee has joined #openstack | 18:12 | |
| vishy | that sends a packet to dnsmasq releasing the ip when you terminate the instance | 18:12 |
| vishy | (otherwise the lease has to timeout, so ips are reclaimed after about 10 min by default) | 18:13 |
| isaacfinnegan | i'm seeing this on a copute node | 18:13 |
| isaacfinnegan | DHCP packet received on br100 which has no address | 18:13 |
| *** masom has quit IRC | 18:13 | |
| n88 | vishy: ah great to know about the 10 min timeout as well as the flag to release; i have that flag now in my nova.conf: --force_dhcp_release=True but perhaps I should try it without setting an actual value (if it's a boolean flag it appears you can just put the name of the configuration parameter without specifiying any boolean value) | 18:15 |
| vishy | correct | 18:15 |
| vishy | no need to put =True | 18:15 |
| isaacfinnegan | dnsmasq-dhcp[4428]: DHCP packet received on br100 which has no address | 18:16 |
| isaacfinnegan | on nova-compute node. anyone see this before? | 18:16 |
| *** lorin1 has joined #openstack | 18:20 | |
| *** oubiwann has joined #openstack | 18:24 | |
| *** clauden_ has quit IRC | 18:25 | |
| *** clauden_ has joined #openstack | 18:25 | |
| *** shentonfreude has joined #openstack | 18:25 | |
| *** vernhart has quit IRC | 18:25 | |
| *** vernhart has joined #openstack | 18:26 | |
| *** shentonfreude has quit IRC | 18:27 | |
| *** huslage has quit IRC | 18:29 | |
| *** huslage has joined #openstack | 18:29 | |
| *** dolphm_ has quit IRC | 18:30 | |
| *** cereal_bars has quit IRC | 18:30 | |
| *** dolphm has joined #openstack | 18:31 | |
| *** jeh has quit IRC | 18:35 | |
| *** GheRivero_ has joined #openstack | 18:36 | |
| *** jeh has joined #openstack | 18:36 | |
| vishy | isaacfinnegan: often that is from dhcprequests from another machine in the network | 18:37 |
| *** syah has quit IRC | 18:38 | |
| GheRivero_ | morning | 18:38 |
| isaacfinnegan | there are no other machines in the network | 18:38 |
| isaacfinnegan | dnsmasq on the nova-compute node is configured wrong | 18:38 |
| vishy | isaacfinnegan: how many nodes do you have running nova-compute? | 18:39 |
| isaacfinnegan | it fires up listening to addresses that aren't even in the address space i'm using (and don't match what the other node which is also nova-network is running dnsmasq as) | 18:39 |
| *** Hakon|mbp has quit IRC | 18:39 | |
| vishy | isaacfinnegan: you have some flags set wrong then | 18:39 |
| *** Prz_^ has joined #openstack | 18:40 | |
| *** Prz_^ is now known as Barakas | 18:40 | |
| vishy | isaacfinnegan: you also may have another copy of dnsmasq running, for example from libvirt | 18:40 |
| vishy | are you running nova-network on both hosts? | 18:40 |
| isaacfinnegan | where? it works when I run everything on one system. when I try to add an additional nova-compute node, instances on it don't get ip's | 18:40 |
| isaacfinnegan | ok | 18:40 |
| vishy | did you create your network with multi-host=True? | 18:40 |
| isaacfinnegan | I see | 18:40 |
| *** cdub has joined #openstack | 18:40 | |
| isaacfinnegan | so is the nova-compute node supposed to be running dnsmasq at all? | 18:40 |
| vishy | if so, then you need to run nova-network on both hosts | 18:40 |
| isaacfinnegan | I don't know | 18:41 |
| isaacfinnegan | where can I check that | 18:41 |
| vishy | in the database | 18:41 |
| isaacfinnegan | I would prefer to not use multi-host. | 18:41 |
| isaacfinnegan | I also want the instances to be routable. I am plannign on not using floating ips | 18:41 |
| vishy | how did you create your network? | 18:41 |
| vishy | isaacfinnegan: that is a kind of a difficult setup | 18:41 |
| isaacfinnegan | I want to route directly to instance IP's | 18:42 |
| *** darraghb has quit IRC | 18:42 | |
| isaacfinnegan | I have routers on the back end network | 18:42 |
| *** nerdstein has left #openstack | 18:42 | |
| isaacfinnegan | that can route into that address space | 18:42 |
| vishy | ok | 18:42 |
| *** Charz has quit IRC | 18:42 | |
| isaacfinnegan | but yes, I'm noticing that the nova-network node wants to route everything out it's 'front' interface | 18:42 |
| vishy | does it work? | 18:42 |
| isaacfinnegan | it does. | 18:42 |
| isaacfinnegan | kidn of | 18:42 |
| isaacfinnegan | kind of | 18:43 |
| isaacfinnegan | but I have to manually remove the snat iptables rule | 18:43 |
| vishy | you can pass dhcpoption,3 | 18:43 |
| isaacfinnegan | I can't see there's a way to tell openstack to not do nat | 18:43 |
| isaacfinnegan | I did that | 18:43 |
| isaacfinnegan | which works to have the instance use another gateway | 18:43 |
| isaacfinnegan | but. the iptables on thehost still wants to snat everything leaving | 18:43 |
| vishy | you could work around the snat | 18:43 |
| vishy | by setting --dmz_cidr=0.0.0.0/0 | 18:44 |
| *** bengrue has joined #openstack | 18:44 | |
| vishy | it won't snat anything going to the dmz_cidr | 18:44 |
| vishy | multi_host is still probably what you want | 18:44 |
| vishy | so that you don't have one host doing dhcp for all of the hosts | 18:45 |
| vishy | but it is up to you | 18:45 |
| *** mgoldmann has quit IRC | 18:47 | |
| *** bcwaldon has quit IRC | 18:47 | |
| ashp | Despite no help in here I managed to get instances starting anyway :) | 18:48 |
| isaacfinnegan | ah | 18:48 |
| isaacfinnegan | tahnks | 18:48 |
| vishy | ashp: self-service! | 18:49 |
| ashp | it worked ;) | 18:49 |
| ashp | I'm sure I have half the stuff configured wrong but hey, it's a start | 18:49 |
| vishy | ashp: you are aware of devstack.org? | 18:50 |
| ashp | I am but I'm using RHEL so that makes everything a little trickier | 18:50 |
| ashp | now it's just a network thing, I have a running instance that it can't route to, so once I work that part out.. | 18:51 |
| vishy | ashp: ah yes, that is a bit harder :) gl | 18:51 |
| vishy | ashp: checking the console log is helpful there | 18:51 |
| vishy | then you can find out if the instance got an ip via dhcp etc. | 18:51 |
| *** oubiwann has quit IRC | 18:52 | |
| *** bcwaldon has joined #openstack | 18:52 | |
| *** oubiwann has joined #openstack | 18:52 | |
| ashp | aha, something definitely went wrong, i'll look, thanks | 18:52 |
| n88 | ashp: welcome to the club! I spent about 4 days trying different combinations of things in my .conf files until things worked | 18:53 |
| *** Gordonz_ has joined #openstack | 18:54 | |
| n88 | vishy: I've noticed that my nova-compute services don't survive a reboot; is that default behavior? | 18:54 |
| *** Gordonz_ has quit IRC | 18:55 | |
| isaacfinnegan | I'm about to do a tech talk here at work on this | 18:55 |
| vishy | n88: depending on how they were installed, nova-compute should come back up | 18:55 |
| isaacfinnegan | I'll try the network thing again after | 18:55 |
| *** anon_os has left #openstack | 18:55 | |
| vishy | n88: on oneiric there are upstart scripts to start them afaik | 18:55 |
| *** Gordonz_ has joined #openstack | 18:55 | |
| n88 | vishy: Ubuntu 11.10 using apt-get … it seems weird to me that they don't start up…maybe I'll check the logs; otherwise I'll just add the startup my self | 18:56 |
| vishy | yes, maybe they crashed because they started too early? | 18:56 |
| vishy | if so file a bug w/ ubuntu :) | 18:56 |
| n88 | vishy: yeah that's what appears to be happening….in my dmesg output I see it terminating; probably waiting for network to come up or something | 18:57 |
| ashp | wget: can't connect to remote host (169.254.169.254): Network is unreachable | 18:57 |
| ashp | That's what I get from the client, I suppose this is the cause of the failure. :/ | 18:58 |
| ashp | (er, from the instance) | 18:58 |
| *** Gordonz has quit IRC | 18:58 | |
| n88 | vishy: maybe it's not the network; the iface links come up at 10.xxx and my error occurs at 14.xxx | 18:59 |
| *** neogenix has quit IRC | 19:00 | |
| ashp | I am pretty sure they aren't getting an IP from dhcp at all. Am I meant to have two dnsmasq's running? | 19:00 |
| *** neogenix has joined #openstack | 19:00 | |
| n88 | vishy: got it…appears that libvirtd is not starting up before nova-compute | 19:01 |
| ashp | Do people generally assign just fake 10/8's to their nodes? I probably made a mistake by trying to use real ip's. | 19:02 |
| *** hggdh has joined #openstack | 19:02 | |
| *** dolphm has quit IRC | 19:02 | |
| n88 | ashp: what do you mean by nodes? instances? or compute machines? | 19:02 |
| *** reed has quit IRC | 19:02 | |
| ashp | instances, sorry, I gotta get my terminology clear. | 19:02 |
| *** syah has joined #openstack | 19:03 | |
| ashp | I was going to use real IPs while getting up with testing but that might be making my setup more complicated as a result. | 19:03 |
| n88 | ashp: what networking mode? | 19:03 |
| *** stuntmachine has quit IRC | 19:03 | |
| ashp | n88: bridge, I think. I've been following about eight different guides. | 19:03 |
| ashp | nova-manage network create private --fixed_range_v4=165.212.105.0/25 --num_networks=1 --network_size=128 --bridge=br100 --bridge_interface=eth0 --multi_host=T | 19:04 |
| ashp | that's how I created the network in the end | 19:04 |
| n88 | ashp: most cloud systems have the concept of private ip and public ips for instances; both are usually assigned by dhcp (although you can manually do the public ips) | 19:04 |
| *** Hakon|mbp has joined #openstack | 19:04 | |
| n88 | ashp: hahaha, I know what you mean by guides…not a single one I found worked completely through and through…by using 3 or 4 I was finally able to get things setup | 19:04 |
| ashp | n88: Yeah, thats how I'm used to things in ec2. I just wasn't sure how I'd have a pool of 10's and then assign public ips to them as required. :/ | 19:05 |
| ashp | I might try and change it all to be 10/8 for now and go from there | 19:05 |
| *** adjohn has quit IRC | 19:05 | |
| n88 | ashp: I switched over to 10/8 for fixed_ips. That seemed to work well for me; as for assigning public…I just asked Vishy…you can either do it manually using euca-allocate and euca-assign | 19:05 |
| *** adjohn has joined #openstack | 19:05 | |
| n88 | ashp: or you can supposedly set a flag in nova.conf (I'm actually about to try this): --auto_assign_floating_ip | 19:06 |
| n88 | ashp: supposedly that flag should have the DHCP automatically assign a public IP to fixed IP once the instance starts up... | 19:06 |
| *** jblesage has joined #openstack | 19:07 | |
| ashp | I can't even delete the network I made with nova-manage network, it's going well :) | 19:07 |
| n88 | ashp: networking is definitely one of the 'trickier' aspects because most setups have 3 or 4 "networks" | 19:08 |
| *** Barakas has left #openstack | 19:08 | |
| n88 | ashp: if it makes you feel any better, I hosed up a volume and now can't get rid of it…will probably need to restart the machine, stop the openstack services, and manually try to purge it =P | 19:08 |
| *** Prz_^ has joined #openstack | 19:09 | |
| *** jerQ has joined #openstack | 19:09 | |
| ashp | 2011-11-03 15:02:44,863 CRITICAL nova [-] Network must be disassociated from project test before delete | 19:10 |
| ashp | Hmm, how do you disassociate them from projects? | 19:10 |
| *** mdomsch has joined #openstack | 19:10 | |
| *** lorin1 has quit IRC | 19:10 | |
| *** develtech has joined #openstack | 19:10 | |
| ashp | Scrub apparently, whatever that does | 19:11 |
| vishy | nova-manage project scrub <xxx> | 19:11 |
| *** bcwaldon_ has joined #openstack | 19:11 | |
| vishy | or set the project_id to null in the db | 19:11 |
| *** bcwaldon has quit IRC | 19:11 | |
| ashp | what does scrub actually do, just so I know for future references | 19:12 |
| ashp | don't want to accidently scrub something important | 19:12 |
| vishy | removes networks and security groups for the project | 19:13 |
| *** df1 has quit IRC | 19:14 | |
| ashp | annoyingly things seem to ignore my network changes, do I have to restart the nova-compute as well? | 19:14 |
| *** reed has joined #openstack | 19:14 | |
| ashp | it starts them up in the non-10/8 range | 19:14 |
| vishy | ashp: how did you change the network? | 19:14 |
| *** stuntmachine has joined #openstack | 19:14 | |
| vishy | so i believe there is a bug in network delete where it doesn't delete the associated fixed ips | 19:15 |
| vishy | someone mentioned that in irc | 19:15 |
| ashp | vishy: I deleted the old network and added a new one (so I only have one in nova-manage network list) - I then killed dnsmasq and started up nova-network again | 19:15 |
| ashp | ah hmm, that might be what I hit then | 19:15 |
| vishy | so you might have to go into the db and delete the other fixed_ips | 19:15 |
| vishy | yeah :( | 19:15 |
| ashp | just in networks? | 19:16 |
| ashp | because in there is just the 10 range | 19:16 |
| ashp | oh no, I see, the fixed_ip table, like you mentioned | 19:16 |
| *** reed has quit IRC | 19:17 | |
| *** bcwaldon_ has quit IRC | 19:17 | |
| *** bcwaldon has joined #openstack | 19:17 | |
| ashp | mysql> delete from fixed_ips where address LIKE "165.212%"; | 19:17 |
| ashp | Query OK, 128 rows affected (0.01 sec) | 19:17 |
| ashp | Ok, I can confirm THAT bug | 19:17 |
| *** lorin1 has joined #openstack | 19:17 | |
| ashp | INSTANCEi-00000008ami-0000000310.0.0.310.0.0.3 <-- looks better. | 19:18 |
| *** lorin1 has joined #openstack | 19:18 | |
| *** lorin1 has quit IRC | 19:18 | |
| *** lorin1 has joined #openstack | 19:19 | |
| *** javiF has quit IRC | 19:19 | |
| ashp | Linux server-8 2.6.35-22-virtual #35-Ubuntu SMP Sat Oct 16 23:19:29 UTC 2010 x86_64 GNU/Linux | 19:22 |
| ashp | It's a miracle! I made an instance! | 19:22 |
| ashp | Now to start the actual hard work :) | 19:22 |
| *** wittwerch has quit IRC | 19:24 | |
| *** clauden__ has joined #openstack | 19:26 | |
| *** clauden_ has quit IRC | 19:26 | |
| *** dendro-afk is now known as dendrobates | 19:26 | |
| n88 | vishy: that was me regarding the associated fixed ips; it sets the deleted flag, but I wasn't using scrub, just network delete | 19:26 |
| n88 | ashp: congrats! I was in the same boat as you this morning (no instances, bad IPs, etc.) then switched over to 10/8 like you and also went through my config for the 1,000 time and this afternoon got things working too | 19:27 |
| *** Ryan_Lane has quit IRC | 19:27 | |
| *** Ryan_Lane has joined #openstack | 19:28 | |
| vishy | it sets the deleted flag for fixed_ips? | 19:28 |
| n88 | ashp: you might want to invest in a graphical MySQL viewer…I've been using one and it's been extremely helpful | 19:28 |
| vishy | or just for network? | 19:28 |
| vishy | n88: wuss! | 19:28 |
| vishy | cli ftw | 19:28 |
| vishy | :) | 19:28 |
| ashp | n88: congrats on reaching the working point :) | 19:29 |
| ashp | my next job is to roll all the crap I did by hand into puppet and then rebuild this box from scratch again | 19:29 |
| ashp | and then the next job is "figure out how to build a redhat ami" | 19:29 |
| ashp | sucks that I can't just steal the one from ec2 but you can't win them all | 19:29 |
| vishy | oh puppet | 19:30 |
| n88 | vishy: actually that does look like a bug. I can confirm…I just did this: sudo nova-manage network delete --network=10.0.0.0/24 and it deleted the network from the networks table, but left fixed_ip table untouched…with those IPs not even receiving the deleted flag; not sure about the scrub command | 19:31 |
| n88 | vishy: haha yeah I'm a slacker…what can I say. Sequel Pro on a mac is a lifesaver =P | 19:32 |
| *** mrevell has joined #openstack | 19:32 | |
| *** hggdh has quit IRC | 19:33 | |
| n88 | vishy: what's annoying about that is then when I re-create my old network, it dumps another 256 addresses with the same IP back into the fixed_ip table so now I have 512 IP entries with duplicate IPs (but different network_ids so I'm sure it will work fine…just makes db messy) | 19:34 |
| vishy | ashp: you might get a little headstart by checking out http://github.com/cloudbuilders/openstack-puppet | 19:36 |
| ashp | next I need to look at the web frontend to see if I can skip a lot of the script type stuff on the backend | 19:36 |
| vishy | can't promise it will be maintained, but it does work... | 19:37 |
| ashp | yeah, I'll take a look, see what I can translate to rhel | 19:37 |
| *** jblesage has quit IRC | 19:38 | |
| *** tryggvil_ has quit IRC | 19:38 | |
| vishy | n88: are the old ips marked deleted=1? | 19:39 |
| *** jedi4ever has joined #openstack | 19:39 | |
| vishy | if so than it should be fine | 19:39 |
| n88 | vishy: no, they left deleted=0 | 19:39 |
| vishy | then it won't work | 19:39 |
| vishy | :) | 19:39 |
| vishy | they should be marked deleted=1 | 19:39 |
| vishy | nova is careful not to "actually" delete anything | 19:39 |
| n88 | vishy: nope…haha. Yeah I figured as much. The database architecture seems pretty solid actually | 19:40 |
| *** smeier00 has left #openstack | 19:40 | |
| n88 | vishy: yeah I saw that everything just gets a deleted tag, which is actually great for auditing purposes | 19:40 |
| ashp | So how much stuff can the web interface do in terms of stuff like, creating keypairs, all the kind of stuff you do in ec2 today? | 19:43 |
| *** reed has joined #openstack | 19:43 | |
| *** mrevell has quit IRC | 19:44 | |
| *** mrevell has joined #openstack | 19:44 | |
| gnu111 | ashp: are you using RHEL with Diablo? | 19:45 |
| *** mrevell has quit IRC | 19:45 | |
| ashp | yes | 19:45 |
| gnu111 | ashp: did you even you run a ubuntu instance? how many nodes do you have? I am having some issues. | 19:46 |
| ashp | gnu111: I ran an ubuntu instance on RHEL6.1, I only have 1 node right now | 19:46 |
| ashp | gnu111: that's as far as I managed to take it so far, what's the problem you ran into? | 19:46 |
| *** rsampaio has quit IRC | 19:46 | |
| gnu111 | ashp: ok. that worked for me too. multiple nodes, I had issues: http://paste.openstack.org/show/3056/ | 19:46 |
| ashp | Ah I was going to work out to multiple nodes once I had this single one building in a repeatable fashion :/ | 19:47 |
| gnu111 | ashp: please try it. If it doesn't work then it is not just me. :) | 19:47 |
| ashp | I will do, it might be a day or so I'm afraid, i carved out a few hours to play but I'll be back to the grind soon :) | 19:48 |
| *** gmih has quit IRC | 19:48 | |
| ashp | argh I will have to install keystone to get the web interface going | 19:50 |
| ashp | openstack, y u so difficult!! *rageface* | 19:50 |
| *** adjohn has quit IRC | 19:51 | |
| *** cdub has quit IRC | 19:53 | |
| *** jeh has quit IRC | 19:53 | |
| *** rsampaio has joined #openstack | 19:54 | |
| *** GheRivero_ has quit IRC | 19:55 | |
| *** dolphm has joined #openstack | 19:55 | |
| *** huslage has quit IRC | 19:56 | |
| *** jblesage has joined #openstack | 19:56 | |
| *** clopez has quit IRC | 20:00 | |
| *** MarkAtwood has quit IRC | 20:00 | |
| *** jedi4ever has quit IRC | 20:01 | |
| *** mrevell has joined #openstack | 20:01 | |
| *** jakedahn has quit IRC | 20:03 | |
| *** mrevell_ has joined #openstack | 20:03 | |
| *** mrevell has quit IRC | 20:03 | |
| *** jakedahn has joined #openstack | 20:03 | |
| *** mrevell_ has quit IRC | 20:04 | |
| *** mrevell has joined #openstack | 20:04 | |
| *** cereal_bars has joined #openstack | 20:08 | |
| *** mrevell has quit IRC | 20:10 | |
| *** dendrobates is now known as dendro-afk | 20:11 | |
| *** gyee has quit IRC | 20:13 | |
| soren | ashp: "It was hard to write, so it should be hard to use." | 20:16 |
| ashp | ;) | 20:16 |
| *** dolphm has quit IRC | 20:17 | |
| *** isaacfinnegan has quit IRC | 20:18 | |
| *** pixelbeat has quit IRC | 20:22 | |
| *** msivanes has quit IRC | 20:22 | |
| *** bcwaldon has quit IRC | 20:23 | |
| *** jdg has joined #openstack | 20:24 | |
| *** dtroyer has quit IRC | 20:24 | |
| *** vidd-away is now known as vidd | 20:27 | |
| vidd | ashp, if it was easy, everyone woul use it =] | 20:27 |
| *** bengrue has quit IRC | 20:29 | |
| *** krow has quit IRC | 20:31 | |
| *** jakedahn_ has joined #openstack | 20:31 | |
| jdg | Anybody here running devstack in virtualbox on? | 20:31 |
| *** llang629_ has joined #openstack | 20:33 | |
| *** stevegjacobs has joined #openstack | 20:33 | |
| *** jakedahn has quit IRC | 20:34 | |
| *** jakedahn_ is now known as jakedahn | 20:34 | |
| *** clopez has joined #openstack | 20:36 | |
| *** dgags has joined #openstack | 20:40 | |
| *** jeh has joined #openstack | 20:40 | |
| *** jeh has quit IRC | 20:41 | |
| *** bengrue has joined #openstack | 20:43 | |
| *** Gordonz_ has quit IRC | 20:43 | |
| jdg | password | 20:44 |
| *** neogenix has quit IRC | 20:45 | |
| *** krow has joined #openstack | 20:45 | |
| tjoy | jdg: horror | 20:45 |
| jdg | tjoy: devstack or 'password' :) | 20:46 |
| tjoy | mostly 'password' | 20:47 |
| jdg | :) That was the old, oops... wrong terminal | 20:47 |
| tjoy | i'd remind you to change it but it's pretty clear you don't care :P | 20:48 |
| jdg | LOL | 20:48 |
| *** mdomsch has quit IRC | 20:49 | |
| *** stevegjacobs has quit IRC | 20:56 | |
| *** jsavak has quit IRC | 20:56 | |
| *** TheOsprey has quit IRC | 20:57 | |
| *** cp16net has quit IRC | 20:59 | |
| *** hezekiah_ has joined #openstack | 20:59 | |
| *** stuntmachine has quit IRC | 21:01 | |
| *** stuntmachine has joined #openstack | 21:01 | |
| *** gyee has joined #openstack | 21:02 | |
| *** lorin1 has quit IRC | 21:03 | |
| *** lorin1 has joined #openstack | 21:03 | |
| *** adjohn has joined #openstack | 21:04 | |
| *** hezekiah_ is now known as isaacfinnegan | 21:06 | |
| *** lorin1 has left #openstack | 21:06 | |
| n88 | does anybody know for images, the attributes min_ram and min_disk what type of values they should be? Bytes? Megabytes? Gigabytes? | 21:10 |
| n88 | this is using glance and the columns in the glance.image table | 21:12 |
| *** stuntmachine has quit IRC | 21:13 | |
| *** develtech has quit IRC | 21:13 | |
| *** kieron has quit IRC | 21:15 | |
| *** df1 has joined #openstack | 21:17 | |
| *** llang629_ is now known as llang629 | 21:17 | |
| *** Ruetobas has quit IRC | 21:17 | |
| *** benrodrigue has quit IRC | 21:18 | |
| isaacfinnegan | ok. so i've set dmz_cidr | 21:25 |
| isaacfinnegan | and I have made sure no extra dnsmasq processes are running | 21:25 |
| isaacfinnegan | and have a nova-compute node trying to bringup an instance but it cannot get an IP | 21:26 |
| *** Ruetobas has joined #openstack | 21:26 | |
| isaacfinnegan | nova list shows the instances having IP's | 21:26 |
| isaacfinnegan | but the instance times out | 21:26 |
| *** npmapn has quit IRC | 21:28 | |
| vidd | Kiall, you here? | 21:28 |
| *** clauden__ has quit IRC | 21:29 | |
| *** clauden_ has joined #openstack | 21:30 | |
| vidd | anyone here have keystone and euca2ools working together? | 21:30 |
| *** adjohn has quit IRC | 21:31 | |
| *** llang629 has quit IRC | 21:31 | |
| *** llang629 has joined #openstack | 21:32 | |
| *** MarkAtwood has joined #openstack | 21:34 | |
| *** neogenix has joined #openstack | 21:34 | |
| *** jwalcik has quit IRC | 21:44 | |
| *** sdake has quit IRC | 21:45 | |
| *** mrevell has joined #openstack | 21:46 | |
| *** mrevell has joined #openstack | 21:46 | |
| *** bsza-I has joined #openstack | 21:50 | |
| *** bsza has quit IRC | 21:51 | |
| *** imsplitbit has quit IRC | 21:53 | |
| *** bsza-I has quit IRC | 21:55 | |
| vidd | is it just me or is keystone evil? | 21:55 |
| *** miclorb_ has joined #openstack | 21:59 | |
| *** llang629_ has joined #openstack | 22:00 | |
| *** syah has quit IRC | 22:00 | |
| *** llang629 has quit IRC | 22:01 | |
| *** neogenix has quit IRC | 22:01 | |
| *** reed has quit IRC | 22:01 | |
| *** lborda has quit IRC | 22:04 | |
| errr | its evil :) | 22:05 |
| *** mattray has quit IRC | 22:05 | |
| *** markvoelker has quit IRC | 22:08 | |
| *** llang629_ has left #openstack | 22:08 | |
| *** alekibango has quit IRC | 22:09 | |
| *** dubsquared has quit IRC | 22:10 | |
| *** siwos has joined #openstack | 22:11 | |
| *** mrevell has quit IRC | 22:12 | |
| uvirtbot | New bug: #885891 in openstack-manuals "Need to replace all the euca- commands with nova client ones" [Undecided,New] https://launchpad.net/bugs/885891 | 22:13 |
| *** clauden___ has joined #openstack | 22:15 | |
| *** clauden_ has quit IRC | 22:15 | |
| *** rsampaio has quit IRC | 22:16 | |
| *** clauden___ has quit IRC | 22:17 | |
| *** clauden_ has joined #openstack | 22:17 | |
| *** reed has joined #openstack | 22:19 | |
| *** freeflying has quit IRC | 22:19 | |
| *** Hakon|mbp has quit IRC | 22:26 | |
| *** freeflying has joined #openstack | 22:26 | |
| *** catintheroof has quit IRC | 22:27 | |
| *** jblesage has quit IRC | 22:27 | |
| *** ppradhan has left #openstack | 22:32 | |
| *** dtroyer has joined #openstack | 22:37 | |
| *** siwos has quit IRC | 22:39 | |
| *** Prz_^ has quit IRC | 22:43 | |
| *** rnorwood has quit IRC | 22:49 | |
| *** PeteDaGuru has left #openstack | 22:50 | |
| *** adjohn has joined #openstack | 22:55 | |
| *** alekibango has joined #openstack | 22:56 | |
| *** cereal_bars has quit IRC | 23:00 | |
| *** cdub has joined #openstack | 23:00 | |
| *** cereal_bars has joined #openstack | 23:00 | |
| *** takarider has quit IRC | 23:04 | |
| *** andreaf has joined #openstack | 23:04 | |
| *** cereal_bars has quit IRC | 23:04 | |
| *** cereal_bars has joined #openstack | 23:04 | |
| *** doorlock has joined #openstack | 23:04 | |
| *** andreaf has quit IRC | 23:07 | |
| *** alekibango has quit IRC | 23:07 | |
| *** alekibango has joined #openstack | 23:08 | |
| *** andreaf has joined #openstack | 23:08 | |
| *** bhall has quit IRC | 23:09 | |
| *** sdake has joined #openstack | 23:09 | |
| *** bhall has joined #openstack | 23:10 | |
| *** alekibango has quit IRC | 23:14 | |
| *** alekibango_ has joined #openstack | 23:14 | |
| *** pixelbeat has joined #openstack | 23:14 | |
| *** cdub has quit IRC | 23:17 | |
| *** adjohn has quit IRC | 23:19 | |
| *** adjohn has joined #openstack | 23:21 | |
| *** isaacfinnegan has quit IRC | 23:23 | |
| *** ahasenack has quit IRC | 23:24 | |
| *** ahasenack has joined #openstack | 23:24 | |
| *** rnirmal has quit IRC | 23:29 | |
| *** misheska has joined #openstack | 23:32 | |
| *** stevegjacobs has joined #openstack | 23:34 | |
| *** neogenix has joined #openstack | 23:36 | |
| *** jdg has quit IRC | 23:38 | |
| *** df1 has quit IRC | 23:40 | |
| *** cdub has joined #openstack | 23:44 | |
| *** aliguori has quit IRC | 23:46 | |
| *** aliguori has joined #openstack | 23:48 | |
| *** andreaf has quit IRC | 23:49 | |
| *** andreaf has joined #openstack | 23:50 | |
| *** cdub has quit IRC | 23:51 | |
| *** code_franco has quit IRC | 23:51 | |
| *** dtroyer_ has joined #openstack | 23:53 | |
| *** dendro-afk is now known as dendrobates | 23:55 | |
| *** andreaf has quit IRC | 23:55 | |
| *** dtroyer has quit IRC | 23:56 | |
| *** dtroyer_ is now known as dtroyer | 23:56 | |
| *** andreaf has joined #openstack | 23:56 | |
| *** oubiwann has quit IRC | 23:59 | |
| *** andreaf has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!