| *** miclorb_ has joined #openstack | 00:05 | |
| *** kevnfx has joined #openstack | 00:08 | |
| *** syah has quit IRC | 00:21 | |
| *** jessec has quit IRC | 00:25 | |
| *** kevnfx has left #openstack | 00:26 | |
| *** _anm has joined #openstack | 00:30 | |
| *** _anm is now known as iammartian_ | 00:31 | |
| *** michael_dreamhos has joined #openstack | 00:33 | |
| *** michael_dreamhos is now known as michaeldreamhost | 00:34 | |
| *** maplebed has quit IRC | 00:41 | |
| *** hisaki has joined #openstack | 00:56 | |
| *** pharkmillups has quit IRC | 00:56 | |
| *** joearnol_ has quit IRC | 01:02 | |
| *** aliguori has joined #openstack | 01:03 | |
| *** adjohn has joined #openstack | 01:03 | |
| *** jc_smith has quit IRC | 01:06 | |
| *** ppetraki has quit IRC | 01:07 | |
| *** miclorb_ has quit IRC | 01:33 | |
| *** pvo has joined #openstack | 01:34 | |
| *** ChanServ sets mode: +v pvo | 01:34 | |
| *** miclorb_ has joined #openstack | 01:48 | |
| *** DubLo7 has joined #openstack | 01:50 | |
| *** Cybodog has quit IRC | 01:51 | |
| *** hazmat has quit IRC | 01:51 | |
| *** hazmat has joined #openstack | 01:52 | |
| *** SandCube has joined #openstack | 01:57 | |
| *** SandCube has left #openstack | 01:57 | |
| *** michaeldreamhost has quit IRC | 01:58 | |
| *** hisaki has quit IRC | 01:58 | |
| *** syah has joined #openstack | 02:04 | |
| *** mtaylor has quit IRC | 02:10 | |
| *** burris has quit IRC | 02:14 | |
| *** coredump|br has quit IRC | 02:16 | |
| *** mmalgeri has joined #openstack | 02:20 | |
| *** coredump|br has joined #openstack | 02:26 | |
| *** littleidea has quit IRC | 02:28 | |
| *** hisaki has joined #openstack | 02:38 | |
| *** mtaylor has joined #openstack | 02:45 | |
| *** ChanServ sets mode: +v mtaylor | 02:45 | |
| *** relateable__ has joined #openstack | 03:00 | |
| *** mtaylor has quit IRC | 03:07 | |
| *** mmalgeri has quit IRC | 03:13 | |
| *** kashyapc has joined #openstack | 03:13 | |
| *** littleidea has joined #openstack | 03:13 | |
| *** joearnold has joined #openstack | 03:15 | |
| *** deshantm has quit IRC | 03:17 | |
| *** deshantm_cosi has quit IRC | 03:17 | |
| *** deshantm has joined #openstack | 03:29 | |
| *** deshantm_cosi has joined #openstack | 03:30 | |
| *** zaitcev has quit IRC | 03:32 | |
| *** joearnold has quit IRC | 04:01 | |
| *** holoway has joined #openstack | 04:23 | |
| *** hisaki has joined #openstack | 04:24 | |
| *** adjohn has quit IRC | 04:26 | |
| *** adjohn has joined #openstack | 04:28 | |
| *** silassewell has quit IRC | 04:30 | |
| *** miclorb_ has quit IRC | 04:32 | |
| *** pvo has quit IRC | 04:33 | |
| *** holoway has quit IRC | 04:38 | |
| *** seats has left #openstack | 04:39 | |
| *** joearnold has joined #openstack | 04:40 | |
| *** suchitp has joined #openstack | 04:41 | |
| *** holoway has joined #openstack | 04:41 | |
| *** seats has joined #openstack | 04:41 | |
| *** miclorb has joined #openstack | 04:42 | |
| *** DubLo7 has quit IRC | 05:07 | |
| *** mtaylor has joined #openstack | 05:10 | |
| *** ChanServ sets mode: +v mtaylor | 05:10 | |
| *** suchitp has left #openstack | 05:20 | |
| *** iammartian_ has quit IRC | 05:20 | |
| *** stewart has quit IRC | 05:23 | |
| *** joschi has quit IRC | 05:24 | |
| *** omidhdl has joined #openstack | 05:33 | |
| *** stewart has joined #openstack | 05:36 | |
| *** miclorb has quit IRC | 05:41 | |
| *** joschi has joined #openstack | 05:43 | |
| *** relateable__ has quit IRC | 05:47 | |
| *** joearnold has quit IRC | 05:55 | |
| *** dysinger has quit IRC | 06:01 | |
| *** befreax has joined #openstack | 06:28 | |
| *** kashyapc has quit IRC | 06:46 | |
| *** kashyapc has joined #openstack | 06:50 | |
| *** gaveen has quit IRC | 06:53 | |
| *** brd_from_italy has joined #openstack | 07:05 | |
| *** guigui1 has joined #openstack | 07:12 | |
| *** ibarrera has joined #openstack | 07:14 | |
| *** gaveen has joined #openstack | 07:19 | |
| *** gaveen has quit IRC | 07:50 | |
| *** f4m8_ is now known as f4m8 | 07:52 | |
| *** mikhail has joined #openstack | 08:19 | |
| *** allsystemsarego has joined #openstack | 08:22 | |
| *** irahgel has joined #openstack | 08:25 | |
| *** cloudmeat has quit IRC | 08:32 | |
| *** adjohn has quit IRC | 08:35 | |
| mikhail | hello everyone | 08:36 |
|---|---|---|
| *** guigui4 has joined #openstack | 08:36 | |
| mikhail | is there any document for deploying openstack on centos | 08:37 |
| *** guigui1 has quit IRC | 08:37 | |
| *** omidhdl has quit IRC | 08:37 | |
| mikhail | hello can anyone help me , are there any documentation for deploying openstack on centos | 08:39 |
| *** omidhdl has joined #openstack | 08:39 | |
| *** guigui4 has quit IRC | 08:40 | |
| *** mikhail has quit IRC | 08:41 | |
| *** mikhail has joined #openstack | 08:42 | |
| *** hisaki has quit IRC | 08:43 | |
| *** mikhail has quit IRC | 08:49 | |
| *** mikhail has joined #openstack | 08:50 | |
| soren | mikhail: I don't think so, no. | 08:52 |
| soren | mikhail: If you find any, please let me know. | 08:52 |
| mikhail | soren: thanks | 08:53 |
| mikhail | soren:ok | 08:53 |
| soren | mikhail: There seems to be some half-baked ones at the bottom of http://wiki.openstack.org/InstallFromSource | 08:53 |
| mikhail | soren : yeah i have checked that. | 08:54 |
| *** mikhail has quit IRC | 08:56 | |
| *** mikhail has joined #openstack | 08:57 | |
| *** miclorb has joined #openstack | 09:03 | |
| *** guigui has joined #openstack | 09:03 | |
| *** mikhail has quit IRC | 09:08 | |
| *** mikhail has joined #openstack | 09:08 | |
| *** miclorb has quit IRC | 09:13 | |
| *** mikhail has quit IRC | 09:14 | |
| *** mikhail has joined #openstack | 09:15 | |
| *** mikhail has quit IRC | 09:21 | |
| *** mikhail has joined #openstack | 09:21 | |
| *** perestrelka has quit IRC | 09:23 | |
| *** brd_from_italy has quit IRC | 09:23 | |
| *** mikhail has quit IRC | 09:26 | |
| *** mikhail has joined #openstack | 09:26 | |
| *** mikhail has quit IRC | 09:29 | |
| *** mikhail has joined #openstack | 09:29 | |
| *** jtimberman has quit IRC | 09:36 | |
| *** jtimberman has joined #openstack | 09:39 | |
| *** brd_from_italy has joined #openstack | 09:40 | |
| *** mikhail has quit IRC | 09:45 | |
| *** mikhail has joined #openstack | 09:45 | |
| *** sirp1 has quit IRC | 09:55 | |
| *** mikhail has quit IRC | 09:55 | |
| *** littleidea has quit IRC | 09:57 | |
| *** miclorb has joined #openstack | 10:01 | |
| *** mikhail has joined #openstack | 10:12 | |
| *** guigui has quit IRC | 10:13 | |
| *** krish has joined #openstack | 10:14 | |
| *** mikhail has quit IRC | 10:22 | |
| *** mikhail has joined #openstack | 10:23 | |
| *** miclorb has quit IRC | 10:38 | |
| *** mikhail has quit IRC | 10:38 | |
| *** mikhail has joined #openstack | 10:39 | |
| *** krish has quit IRC | 10:44 | |
| *** kashyapc has left #openstack | 10:45 | |
| *** miclorb_ has joined #openstack | 10:47 | |
| *** mikhail has quit IRC | 10:51 | |
| *** mikhail has joined #openstack | 10:52 | |
| *** omidhdl has quit IRC | 11:02 | |
| *** mikhail has quit IRC | 11:02 | |
| *** mikhail has joined #openstack | 11:03 | |
| *** omidhdl has joined #openstack | 11:07 | |
| *** mikhail has quit IRC | 11:10 | |
| *** mikhail has joined #openstack | 11:11 | |
| soren | eday: Ok, I'm about to give up.. I'm trying desperately to get rpc working in cloud_unittest and I just can't wrap my head around it. | 11:15 |
| soren | eday: So, e.g. run_instances in the cloud controller calls rpc.call(blah). Using the fakerabbit carrot backend, the call message lands in a queue. | 11:16 |
| *** mikhail has quit IRC | 11:16 | |
| *** mikhail has joined #openstack | 11:17 | |
| soren | eday: Immediately after sending the message, rpc.call calls consumer.wait(limit=1). | 11:17 |
| soren | eday: consumer.wait translates into a call that isn't implemented in fakerabbit. It's supposed to be a blocking call, waiting for a new message to arrive. | 11:17 |
| *** miclorb_ has quit IRC | 11:17 | |
| soren | eday: There are a couple of things I don't quite get here.. a) As far as I can see, consumer.wait() will return when /any/ message has been received. | 11:19 |
| soren | eday: ...so if it makes two rpc.call()s in rapid succession and the responses come back in an order different than the one in which they were sent, wait_msg will not have its result attribute set yet on the first one and things will fail in spectacular ways. | 11:20 |
| *** mikhail has quit IRC | 11:24 | |
| *** mikhail has joined #openstack | 11:25 | |
| soren | eday: b) I'm having trouble understanding how stuff is supposed to get called in unit tests. Making a call results in stuff getting added to a Queue. In order for that to turn into a call, a ConsumerAdapter must call .fetch(enable_callbacks=True) on itself. This is done by attaching it to a twisted Task.LoopingCall, but.. | 11:26 |
| soren | eday: I can't even finish the sentence :) I've been staring at this for hours and hours now, and I just keep getting more confused. | 11:26 |
| *** mikhail has quit IRC | 11:29 | |
| soren | eday: Maybe the whole problem is because of a). | 11:31 |
| soren | How the heck do I block on stuff with eventlet? | 11:34 |
| *** omidhdl has left #openstack | 11:38 | |
| *** ctennis has quit IRC | 11:41 | |
| *** guigui has joined #openstack | 11:56 | |
| *** ctennis has joined #openstack | 12:08 | |
| *** hisaki has joined #openstack | 12:16 | |
| *** jessec has joined #openstack | 12:18 | |
| *** Podilarius has joined #openstack | 12:21 | |
| soren | \o/ | 12:43 |
| *** pvo has joined #openstack | 12:53 | |
| *** ChanServ sets mode: +v pvo | 12:53 | |
| soren | eday: Ok, I think I've got it to work now. | 12:57 |
| soren | eday: And it only took a whole day. | 12:57 |
| *** mmalgeri has joined #openstack | 13:08 | |
| *** jonwood has joined #openstack | 13:08 | |
| eday | soren: the wait() is bound to a specific consumer queue, which is the unique ID of the msg that was sent, so it will only return when one of those messages comes in | 13:12 |
| jaypipes | eday: hmm, you're up early :) | 13:14 |
| soren | eday: You would think so, wouldn't you? | 13:14 |
| soren | eday: However, wait ends up calling a rather anonymous consume on the backend driver. | 13:14 |
| jonwood | Good afternoon | 13:15 |
| jonwood | Before I jump into this, how hard is it to get OpenStack running at the moment. | 13:15 |
| jaypipes | jonwood: morning :) | 13:15 |
| eday | jaypipes: I'm on the east coast :) (visiting family) | 13:15 |
| jaypipes | jonwood: Nova or Swift? | 13:15 |
| jonwood | I'm curious enough to play with it, but I don't want to end up losing the next 3 days to it ;) | 13:15 |
| jonwood | Nova | 13:15 |
| jaypipes | eday: ah! | 13:15 |
| jaypipes | jonwood: it's a little work, but not 3 days :) | 13:16 |
| jonwood | Cool. Is the installation documentation at nova.openstack.org current? | 13:17 |
| eday | soren: but the connection to the server subscribes to a specific queue/exchange, so the server should only send messages destined for it | 13:17 |
| eday | soren: and two calls will have two different consumer connections, each bound to different unique ID queues | 13:17 |
| soren | eday: Yeah. It's all fakerabbit's fault. | 13:18 |
| soren | eday: I've got it working now, though. I can't really tell have fragile it is, but it does seem to work consistently. | 13:19 |
| soren | eday: Hang on, let me push the magic somewhere.. | 13:19 |
| soren | eday: http://bazaar.launchpad.net/~soren/nova/xen/revision/282 | 13:23 |
| soren | eday: and | 13:24 |
| soren | eday: http://bazaar.launchpad.net/~soren/nova/xen/revision/283 | 13:24 |
| soren | Erk, /me removes debugging code. | 13:24 |
| *** pvo has quit IRC | 13:25 | |
| soren | chmouel: I still cannot reproduce bug 649865 | 13:32 |
| uvirtbot | Launchpad bug 649865 in nova "uploading bundle failure" [Undecided,New] https://launchpad.net/bugs/649865 | 13:32 |
| *** littleidea has joined #openstack | 13:33 | |
| soren | chmouel: As for the patch, you've provided. If one of these flags get declared more than once, that's a symptom of a rather severe problem. Your patch would simply hide this problem. | 13:34 |
| *** sevenseeker has joined #openstack | 13:35 | |
| sevenseeker | morning folks | 13:35 |
| soren | chmouel: The flags get declared at the module level. | 13:35 |
| sevenseeker | creiht, don't see you anymore :) what's up in your world? | 13:35 |
| soren | chmouel: Python is supposed to make sure that importing a module does not reexecute it. | 13:35 |
| soren | chmouel: ...so if it does anyway, it's very likely because it's being referenced in two different ways from different places. | 13:35 |
| soren | chmouel: It's reasonably common if a project uses relative imports. | 13:36 |
| soren | chmouel: ...but we don't. | 13:36 |
| chmouel | soren: yeah that what my comment in the bug | 13:39 |
| chmouel | soren: python is sometime messing up with global variable | 13:39 |
| chmouel | soren: what python version are using ? | 13:39 |
| soren | chmouel: 2.6 | 13:40 |
| soren | 2.6.6, specifically. | 13:40 |
| eday | soren: so, why did you go back to the busy poll? That was what I was trying to remove :) | 13:40 |
| *** kashyapc has joined #openstack | 13:40 | |
| soren | eday: Because I wanted the bloody thing to actually work? :) | 13:41 |
| *** littleidea has quit IRC | 13:41 | |
| chmouel | I am using 2.6.5 if that's change anything it's the default coming up on lucid | 13:41 |
| soren | eday: Removing stuff is easy if it doesn't have to work afterwards :) | 13:41 |
| jaypipes | jonwood: not sure :( we welcome feedback on it though! | 13:41 |
| eday | soren: well, it was working as far as I could tell, what broke exactly? | 13:41 |
| jonwood | jaypipes: Excellent, I'm making notes, so I should have some later ;) | 13:42 |
| soren | eday: rpc.call with eventlet code would just hang. The method on the receiving end would never get called (with the fakerabbit carrot backend, that is). | 13:42 |
| eday | soren: because IMO busy poll is not an acceptable "working" solution | 13:42 |
| *** Cybodog has joined #openstack | 13:43 | |
| soren | eday: I only call .connect_to_eventlet from that one unit test, though. | 13:43 |
| *** mmalgeri has quit IRC | 13:43 | |
| eday | soren: ok, so this is just for fakerabbit? | 13:43 |
| soren | eday: Yes. | 13:43 |
| soren | eday: ...but I couldn't figure out how to fix it from inside fakerabbit. | 13:44 |
| soren | chmouel: Hey, can you try something for me? | 13:46 |
| chmouel | soren: sure | 13:46 |
| chmouel | soren: fire it away | 13:46 |
| eday | soren: so eventlet can yeild for different types of calls, right now it only does so for blocking socket calls in nova. We can use some other type of message passing system for the fake version that will allow for context switching | 13:46 |
| soren | chmouel: in /usr/lib/pymodules/python2.6/nova/utils.py, in import_object... | 13:47 |
| soren | chmouel: change __import__(import_str) to __import__(import_str, globals()) | 13:47 |
| soren | eday: That's what the sleep thing does. | 13:47 |
| soren | Oh, hang on. | 13:48 |
| eday | soren: Yes, it does, but we should probably do this in fake rabbit, not in rpc.py, no? | 13:48 |
| soren | 13:44 <+soren> eday: ...but I couldn't figure out how to fix it from inside fakerabbit. | 13:48 |
| soren | :) | 13:49 |
| eday | ahh | 13:49 |
| soren | I ended up having to yield in both places. | 13:49 |
| soren | I just noticed I left the sleep in rpc.py at a non-zero value. It can just be 0. | 13:49 |
| soren | ...but I think that has different semantics than I thought. | 13:50 |
| soren | Uh, yeah. Ignore my last few comments. Please. :) | 13:50 |
| eday | heh | 13:50 |
| alekibango | dont worry. the channel is logged. | 13:51 |
| alekibango | :) | 13:51 |
| * eday will be back later, need to run for a bit | 13:51 | |
| chmouel | yeah soren did that change but i still need to reboot my install to reproduce give me a bit and i'll let you know how it goes | 13:51 |
| soren | chmouel: eg? | 13:52 |
| soren | eh? | 13:52 |
| soren | chmouel: Why do you need to reboot? | 13:52 |
| *** abecc has joined #openstack | 13:52 | |
| chmouel | not the server :) just creating admin user and stuff | 13:52 |
| chmouel | ssl certs blah blah | 13:52 |
| chmouel | since I blew it up last night | 13:52 |
| chmouel | should not take me much | 13:52 |
| *** abecc has joined #openstack | 13:52 | |
| *** Cybodog has quit IRC | 13:53 | |
| *** pvo has joined #openstack | 13:54 | |
| *** pvo has joined #openstack | 13:54 | |
| *** ChanServ sets mode: +v pvo | 13:54 | |
| creiht | sevenseeker: work keeps one busy :) | 13:55 |
| sevenseeker | creiht... likely excuse... next you will tell me that folks want their payments and you gotta pay bills, lol | 13:55 |
| chmouel | soren: ok this is weird re-installing with the latest packages from ppa:nova-core does not need to have any hacks | 13:58 |
| *** ar1 has quit IRC | 13:58 | |
| chmouel | soren: I am going to try again with the package generated by myself from trunk | 13:58 |
| jaypipes | jonwood: awesome, much appreciated! | 14:01 |
| chmouel | where is the ldap information are stored by the way ? | 14:03 |
| chmouel | ie: fakeldap stuff | 14:03 |
| chmouel | is that in redis ? | 14:03 |
| notmyname | sevenseeker: ask him how his minecraft fort is coming :-) | 14:03 |
| soren | chmouel: With up-to-date packages, it's in the database. | 14:04 |
| creiht | heh | 14:04 |
| soren | chmouel: Up until yesterday, it was in Redis. | 14:04 |
| chmouel | soren: ah k sweet that was my confusion :) | 14:04 |
| soren | Err... Depending on you config, I guess. | 14:04 |
| * soren ponders for a sec... | 14:04 | |
| soren | No, the default is the DB now. | 14:04 |
| soren | IIRC. | 14:04 |
| * soren takes a (IMO much deserved) break | 14:04 | |
| chmouel | soren: ah ah cool tks man | 14:05 |
| sevenseeker | notmyname: howdy, how have you been? Do you mean 'fork'? | 14:05 |
| chmouel | i'll update the bug report if i can find anything | 14:05 |
| notmyname | I'm good. | 14:06 |
| *** dendro-afk is now known as dendrobates | 14:08 | |
| *** f4m8 is now known as f4m8_ | 14:08 | |
| *** jessec has quit IRC | 14:10 | |
| uvirtbot | New bug: #655152 in nova "Feature Freeze Exception: Xen support" [Undecided,New] https://launchpad.net/bugs/655152 | 14:11 |
| *** ppetraki has joined #openstack | 14:14 | |
| * chmouel wonder if there is a emacs launchpad interface | 14:24 | |
| *** hogstrom has joined #openstack | 14:26 | |
| *** hornbeck has joined #openstack | 14:27 | |
| *** pvo has quit IRC | 14:28 | |
| _0x44 | chmouel: I think launchpad is probably subject to rule #34 and #35. If there isn't one will be created (anything to keep the launchpad devs from actually fixing launchpad. ;) ) | 14:29 |
| chmouel | _0x44: ah ah :) | 14:29 |
| deshantm | soren: we can help test the Xen bits, if that helps for inclusion | 14:30 |
| soren | deshantm: That would be awesome. I was just about to go dust off my Xen box. | 14:35 |
| deshantm | soren: are the instructions to install from source up to date enough or is there a better resource than | 14:37 |
| deshantm | http://wiki.openstack.org/InstallFromSource | 14:37 |
| soren | deshantm: Haven't looked at them for a while, to be honest. | 14:37 |
| soren | deshantm: ...and I never run from source. I build Ubuntu packages and use those. | 14:38 |
| deshantm | soren: we used a install script from one of the devs awhile back and got stuck | 14:38 |
| deshantm | soren: cool, I'll be in touch, I know the code has been moving fast lately | 14:38 |
| deshantm | we'll take another shot at the install | 14:38 |
| soren | gundlach: Any chance you will get around to finishing the redisectomy? I've stared a bit at your id mapping magic and just felt dizzy afterwards :) | 14:39 |
| *** ___mattf has joined #openstack | 14:40 | |
| gundlach | soren: yep, i'm in the middle of making it work | 14:40 |
| * soren hugs gundlach | 14:41 | |
| * gundlach breaks the hug awkwardly | 14:41 | |
| soren | Aw. | 14:41 |
| chmouel | deshantm: I have been trying to keep in touch with the new stuff and when I get something working will post all my notes on the wiki | 14:41 |
| deshantm | chmouel: great! if you can, ping me here too :) | 14:43 |
| chmouel | deshantm: will sure do! | 14:43 |
| soren | I tried twice already to write up install instructions using packages.. Both times, I got caught up in stuff I felt I should polish and started doing that, and then my instructions were outdated. I'm going to try again tomorrow or so. | 14:44 |
| uvirtbot | New bug: #655171 in nova "SQL Errors on a fresh install" [Undecided,New] https://launchpad.net/bugs/655171 | 14:46 |
| soren | Ah, I know what that is. | 14:47 |
| chmouel | soren: the sql errors? | 14:48 |
| soren | Yes. | 14:48 |
| chmouel | there is def some concurrency problem or so it seems | 14:48 |
| soren | Nope. | 14:49 |
| chmouel | any idea about #654584 from the top of your head by any chance ? | 14:49 |
| chmouel | the routes stuff | 14:49 |
| soren | bug 654584 | 14:49 |
| uvirtbot | Launchpad bug 654584 in nova "404 Not found with euca-register " [Undecided,New] https://launchpad.net/bugs/654584 | 14:49 |
| chmouel | i can dig into it if u are busy... | 14:50 |
| soren | I kind of am. | 14:50 |
| soren | upload-bundle should only involve nova-objectstore. | 14:51 |
| soren | Oh, this is euca-register. | 14:51 |
| chmouel | yeah | 14:51 |
| soren | that's both objectstore and api. | 14:51 |
| soren | ...and nothing else. | 14:51 |
| soren | Well, the db, but meh. | 14:51 |
| chmouel | and it's the api server getting back there | 14:51 |
| chmouel | this error i mean | 14:51 |
| soren | Oh, perhaps these are the same problem? | 14:51 |
| soren | bug 654584 and bug 655171, I mean. | 14:51 |
| uvirtbot | Launchpad bug 654584 in nova "404 Not found with euca-register " [Undecided,New] https://launchpad.net/bugs/654584 | 14:51 |
| uvirtbot | Launchpad bug 655171 in nova "SQL Errors on a fresh install" [Undecided,New] https://launchpad.net/bugs/655171 | 14:51 |
| chmouel | yeah | 14:52 |
| chmouel | thought so as well | 14:52 |
| soren | chmouel: Ok, so about that. | 14:52 |
| soren | chmouel: Who owns /var/lib/nova/nova.sqlite? | 14:52 |
| soren | It's meant to be owned by nova. | 14:52 |
| chmouel | nova:nogroups | 14:52 |
| chmouel | 600 | 14:52 |
| soren | That's... well.. ok in this case. | 14:53 |
| chmouel | there is a 'database is locked' error | 14:53 |
| soren | That's normal. | 14:53 |
| chmouel | and already exist table thingy | 14:54 |
| chmouel | does not seem any access / perm error | 14:54 |
| soren | either it's because one of the other workers is using it (not fatal, it'll retry), or it's because the file itself or directory holding the file has permisions problems. | 14:54 |
| soren | chmouel: You probably can't tell. | 14:54 |
| soren | chmouel: It won't look like a file permission error. | 14:55 |
| soren | chmouel: If it can't write to the directory holding the sqlite file, it'll just say it can't lock the db or something like that. | 14:55 |
| chmouel | yep prob... things seems to work as well... | 14:55 |
| soren | does your compute.log really end right there? | 14:56 |
| soren | 2010-10-05 15:33:13+0100 [-] File "/usr/lib/pymodules/python2.6/nova/service.py", line 171, in report_state | 14:56 |
| soren | Looks like an odd last line. | 14:56 |
| *** sevenseeker has quit IRC | 14:56 | |
| chmouel | no finish by 2010-10-05 15:33:48+0100 [-] (root): ERROR Recovered model server connection! | 14:57 |
| chmouel | not sure where that copy and paste went... | 14:57 |
| soren | Heh: "ERROR Recovered model server connection!" | 14:57 |
| soren | It's hardly an error that we now work again :) | 14:57 |
| chmouel | ah i thought it would retry | 14:58 |
| soren | ...but it's convenient that the recovery message is the same level as the error. | 14:58 |
| soren | I does. | 14:58 |
| soren | That's what that message means. | 14:58 |
| soren | It tried again, and now it worked. | 14:58 |
| chmouel | ah k sweet | 14:58 |
| chmouel | prob for sake of cosmetics it would be good to don't have that kind of error tho | 14:58 |
| soren | Right. | 14:59 |
| *** sirp1 has joined #openstack | 15:00 | |
| *** brd_from_italy has quit IRC | 15:04 | |
| chmouel | soren: i set thos other bugs i reported as invalid by the way since I can't reproduce | 15:04 |
| chmouel | soren: which still get me stumped | 15:05 |
| *** rnirmal has joined #openstack | 15:07 | |
| chmouel | soren: and by the way the ./genrootca.sh in /var/lib/nova/CA is it supposed to be package doing it or the user ? | 15:07 |
| *** guigui has quit IRC | 15:08 | |
| soren | chmouel: You shouldn't need to use anything other than nova-manage and euca-* | 15:09 |
| chmouel | well if i don't run that i get something like cannot find openssl.cnf from the os.popen("openssl..... | 15:10 |
| alekibango | hmm those things should be really documented (examples), annegentle please note there is no usage docs for nova | 15:10 |
| chmouel | i can generate a proper bug report if you like | 15:10 |
| deshantm | alekibango: no usage docs? is it "documented in the code" currently? | 15:15 |
| soren | chmouel: You've got a weird environment. :) | 15:16 |
| soren | chmouel: It works just fine for me. | 15:17 |
| soren | chmouel: You running nova-manage as root? | 15:17 |
| soren | chmouel: (sudo or whatever) | 15:17 |
| *** benoitc has quit IRC | 15:17 | |
| *** benoitc has joined #openstack | 15:17 | |
| * soren goes to dinner | 15:17 | |
| soren | \o/ My Xen branc still works. | 15:18 |
| * soren really leaves | 15:18 | |
| alekibango | deshantm: i meant usage for admins, endusers -- not for developers | 15:18 |
| alekibango | :) | 15:18 |
| deshantm | alekibango: yeah, me too | 15:18 |
| deshantm | haha | 15:18 |
| deshantm | so we need to work on that then (docs for nova, etc.)? | 15:19 |
| deshantm | i gotta run for now, but I can help out with testing at least | 15:19 |
| *** dagger has quit IRC | 15:21 | |
| *** dagger has joined #openstack | 15:24 | |
| *** dagger has joined #openstack | 15:24 | |
| alekibango | deshantm: i dont yet feel confidend/able to lead novices on nova... as i am still a novice too | 15:25 |
| alekibango | but i can try whining, or (rather) documenting unfilled needs :) | 15:26 |
| *** dpackard has joined #openstack | 15:32 | |
| dpackard | Hi guys, does openstack require that node computers support virtualization? | 15:33 |
| *** hogstrom has quit IRC | 15:35 | |
| alekibango | dpackard: nope, if you use UML or QEMU | 15:35 |
| alekibango | but xen/kvm is much better | 15:36 |
| dpackard | thanks | 15:38 |
| alekibango | hm btw xen can be used without virtualisation too... i somehow forgot that it works :) | 15:39 |
| alekibango | didnt use it for ages now | 15:39 |
| alekibango | dpackard: you can run openstack even in virtualized environment (uml, qemu) | 15:40 |
| *** silassewell has joined #openstack | 15:41 | |
| dpackard | I'm asking because my company is investigating using Nova to control distributed computation in embedded systems (robotics) | 15:45 |
| *** stewart has quit IRC | 15:45 | |
| dpackard | and it's clear that none of these processors will support virtualization (kvm/xen) | 15:45 |
| dpackard | is there a performance penalty using uml and qemu? | 15:46 |
| *** silassewell has quit IRC | 15:46 | |
| *** stewart has joined #openstack | 15:46 | |
| *** kashyapc has quit IRC | 15:49 | |
| jaypipes | phew... | 15:51 |
| * jaypipes finally removed the faulty chirping smoke detector in house...was driving me nuts. | 15:51 | |
| zul | jaypipes: doesnt it have an off button? | 15:57 |
| jaypipes | zul: no, it doesn't. was hard-wired into the electrical system. | 15:57 |
| zul | jaypipes: ah | 15:58 |
| notmyname | jaypipes: for your safety! (but not your sanity) | 15:58 |
| jaypipes | notmyname: indeed :) I have to go get 5 new detectors at the hardware store :( | 15:58 |
| *** dendrobates is now known as dendro-afk | 16:01 | |
| *** kashyapc has joined #openstack | 16:03 | |
| uvirtbot | New bug: #655207 in swift "make swift work with webob 1.0" [Medium,New] https://launchpad.net/bugs/655207 | 16:06 |
| *** pharkmillups has joined #openstack | 16:10 | |
| *** maplebed has joined #openstack | 16:10 | |
| *** burris has joined #openstack | 16:10 | |
| *** dendro-afk is now known as dendrobates | 16:11 | |
| jaypipes | sirp1: ping | 16:14 |
| sirp1 | jaypipes: hey | 16:14 |
| alekibango | dpackard: yes, qemu is deadly slow, but is similar to kvm in usage. UML is faster a bit, but still not like xen/kvm | 16:14 |
| jaypipes | sirp1: how are ya? feel like making a bit more progress in fleshing out the add/update/delete parts of parallax calls today? :) | 16:15 |
| *** johnpur has joined #openstack | 16:15 | |
| *** ChanServ sets mode: +v johnpur | 16:15 | |
| alekibango | qemu has another advantage - can emulate another architecture | 16:15 |
| jaypipes | johnpur: well, hey, howdy sir :) | 16:15 |
| johnpur | hey there! | 16:15 |
| dpackard | Alekibango: Thanks for your input! | 16:15 |
| sirp1 | jaypipes: yeah def up for that | 16:15 |
| jaypipes | johnpur: getting accustomed to your new role yet? :) | 16:15 |
| uvirtbot | New bug: #655217 in nova "euca-run-instances fails to run VM" [Undecided,New] https://launchpad.net/bugs/655217 | 16:16 |
| alekibango | johnpur: note: all mentioned code is not in main source tree yet | 16:16 |
| alekibango | but exists | 16:16 |
| johnpur | jaypipes: definitely getting intot he swing of things... | 16:16 |
| alekibango | ops, sorry, wrong name :) | 16:16 |
| jaypipes | sirp1: cool. I have a little more work left to add a couple more unit tests to my glance work in nova, but if you could do a code review on the glance-image-service branch, that would be super. | 16:16 |
| alekibango | dpackard: ^^ | 16:16 |
| jaypipes | johnpur: good! :) | 16:16 |
| sirp1 | jaypipes: cool, will do | 16:17 |
| jaypipes | sirp1: cheers | 16:18 |
| alekibango | dpackard: what i value on openstack is 1) fully open, scaleable, adaptable 2) nice tools (python,amqp,..) 3) smart people here 4) smart companies behind those smart people 5) good style, mood, spirit | 16:20 |
| chmouel | soren: yeah i run it as root | 16:21 |
| alekibango | documentation is still behind :) but i believe its only matter of time... | 16:21 |
| alekibango | as first release comes in 2 weeks... | 16:22 |
| *** ibarrera has quit IRC | 16:26 | |
| *** befreax has quit IRC | 16:29 | |
| *** dendrobates is now known as dendro-afk | 16:30 | |
| *** dendro-afk is now known as dendrobates | 16:30 | |
| *** _maxim_ has joined #openstack | 16:31 | |
| *** alekibango has quit IRC | 16:34 | |
| *** alekibango has joined #openstack | 16:34 | |
| *** mdomsch has joined #openstack | 16:34 | |
| *** pharkmillups has quit IRC | 16:35 | |
| dpackard | alekibango: good style, mood and spirit are important in any project :) | 16:36 |
| jonwood | Finally, I can get back to playing, now clients have stopped hassling me with this "work" thing. | 16:38 |
| jonwood | I'm getting "nova.exception.NotFound: Class VlanManager cannot be found" when trying to start nova-api - can anyone point me in the right direction? | 16:39 |
| *** alekibango_ has joined #openstack | 16:40 | |
| *** alekibango has quit IRC | 16:40 | |
| *** sophiap has joined #openstack | 16:40 | |
| *** pharkmillups has joined #openstack | 16:41 | |
| *** johnpur is now known as johnpur_afk | 16:43 | |
| alekibango_ | dpackard: yes, i know that :) thats why i listed it - because openstack wins in all those areas i mentioned... | 16:44 |
| *** joearnold has joined #openstack | 16:45 | |
| alekibango_ | at least it feels this way for me now :) | 16:46 |
| *** _maxim_ has quit IRC | 16:55 | |
| *** _maxim_ has joined #openstack | 16:56 | |
| *** aliguori has quit IRC | 17:02 | |
| *** sevenseeker has joined #openstack | 17:04 | |
| *** jc_smith has joined #openstack | 17:11 | |
| uvirtbot | New bug: #655253 in nova "Deprecation warnings about no request context should be addressed" [Low,Confirmed] https://launchpad.net/bugs/655253 | 17:11 |
| *** littleidea has joined #openstack | 17:13 | |
| *** joearnol_ has joined #openstack | 17:14 | |
| *** joearnold has quit IRC | 17:17 | |
| *** joearnold has joined #openstack | 17:18 | |
| *** joearnol_ has quit IRC | 17:21 | |
| *** pharkmillups has joined #openstack | 17:26 | |
| dpackard | alekibango: i've seen performance benchmarks in favor of using openVZ, would this be compatible with openstack? | 17:29 |
| *** zul has quit IRC | 17:32 | |
| vishy | jaypipes: here? | 17:34 |
| jaypipes | vishy: yes sir! | 17:34 |
| vishy | jaypipes: I have a branch with all the deprecation warnings fixed | 17:35 |
| vishy | it still needs a bit of cleanup | 17:35 |
| jaypipes | vishy: ah, ok. I'll re-assign that bug to you then? | 17:35 |
| vishy | please | 17:35 |
| jaypipes | vishy: no prob. | 17:35 |
| *** rlucio has joined #openstack | 17:36 | |
| jaypipes | vishy: done. | 17:36 |
| *** xfaf has joined #openstack | 17:36 | |
| gundlach | devcamca-: ping? | 17:37 |
| *** jdarcy has joined #openstack | 17:40 | |
| *** joearnold has quit IRC | 17:50 | |
| *** joearnold has joined #openstack | 17:50 | |
| *** xfaf is now known as zul | 17:53 | |
| *** zul has joined #openstack | 17:53 | |
| *** mdomsch has quit IRC | 18:01 | |
| *** joearnold has quit IRC | 18:03 | |
| *** hisaki has quit IRC | 18:05 | |
| *** holoway has quit IRC | 18:08 | |
| *** holoway has joined #openstack | 18:09 | |
| *** ptremblett has joined #openstack | 18:10 | |
| *** mdomsch has joined #openstack | 18:10 | |
| *** zheng_li has joined #openstack | 18:18 | |
| *** zheng_li has quit IRC | 18:18 | |
| *** joearnold has joined #openstack | 18:29 | |
| *** johnbergoon has joined #openstack | 18:37 | |
| *** aliguori has joined #openstack | 18:40 | |
| gundlach | mtaylor: ping? | 18:41 |
| mtaylor | gundlach: pong | 18:41 |
| gundlach | or in his stead, does anyone know how to get bugfixes merged to trunk post-FF? | 18:41 |
| gundlach | oops :) ^^ | 18:41 |
| gundlach | https://code.launchpad.net/~gundlach/nova/ec2_to_internal_id/+merge/37520 has just been marked as "Merged" by LP, but it's not in trunk yet | 18:42 |
| mtaylor | gundlach: same way as before ... it's all about process/policy ... | 18:42 |
| mtaylor | gundlach: looking | 18:42 |
| gundlach | ty | 18:42 |
| jaypipes | soren, gundlach: if you have a moment, pls review lp:~jaypipes/nova/glance-image-service. thx in advance! | 18:45 |
| jaypipes | gundlach: same way as before FF... | 18:45 |
| gundlach | jaypipes: sure thing, will do it now. | 18:46 |
| jaypipes | gundlach: cheers mate | 18:46 |
| gundlach | jaypipes: yeah, i thought it was same as pre-FF but lp:nova shows the last checkin as 11 hours ago, while my branch claims to be Merged. | 18:46 |
| gundlach | mtaylor: maybe I accidentally chose "Merged" instead of "Approved" after I got an approval review? How would I detect that? | 18:46 |
| jaypipes | gundlach: which branch? | 18:47 |
| gundlach | https://code.launchpad.net/~gundlach/nova/ec2_to_internal_id/+merge/37520 | 18:47 |
| *** johnpur_afk is now known as johnpur | 18:49 | |
| mtaylor | gundlach: yeah. I think that may be the case. I can't see anything weird with the branch that would indicate that it got half merged or something | 18:51 |
| jaypipes | gundlach: yeah, looks like it was set to Merged and not Approved, since the merge revision is marked "not available" | 18:51 |
| jaypipes | mtaylor: the merge revision is marked "not available" | 18:51 |
| jaypipes | gundlach: set it back to Approved... | 18:52 |
| mtaylor | jaypipes, gundlach: this will be better soon when the launchpad merge queues are released, since then approval status and merge status will be independent of one another | 18:52 |
| gundlach | ok, pardon me for my clumsy clicking :) | 18:52 |
| jaypipes | mtaylor: nice :) | 18:52 |
| jaypipes | gundlach: no worries, man. done it many times myself ;) | 18:52 |
| *** silassewell has joined #openstack | 18:58 | |
| *** deshantm_cosi has quit IRC | 19:01 | |
| *** irahgel has left #openstack | 19:06 | |
| jaypipes | vishy: ping | 19:14 |
| jaypipes | vishy: you still working on this one? https://bugs.launchpad.net/nova/+bug/649188 | 19:14 |
| uvirtbot | Launchpad bug 649188 in nova "RS API is still on Redis" [High,New] | 19:14 |
| vishy | jaypipes: I'm not really, no time at the moment | 19:14 |
| jaypipes | vishy: mind if I grab that one? | 19:15 |
| vishy | not at all, but talk to gundlach | 19:15 |
| jaypipes | gundlach: ^^ pls advise... | 19:15 |
| vishy | i think the only thing he uses it for is to convert ec2 ids into rs ids | 19:15 |
| gundlach | jaypipes: i think dietz was waiting for my branch to merge and then was pulling out the translation layer | 19:15 |
| vishy | and there is discussion about it | 19:15 |
| gundlach | he did some work on it last thursday at the SFO meetup but i'm not sure the status. | 19:15 |
| jaypipes | _cerberus_: pls advicse :) | 19:15 |
| gundlach | _cerberus_: ? | 19:15 |
| _cerberus_ | I was waiting for the branch to merge, yes | 19:16 |
| jaypipes | gundlach: I'm going through all the bugs and trying to get them all properly assigned and the statuses updated... | 19:16 |
| jaypipes | _cerberus_: it's all merged now. | 19:16 |
| jaypipes | _cerberus_: I will mark that bug in progress, then, and assigned to you... | 19:17 |
| gundlach | jaypipes: thx for the administration | 19:17 |
| gundlach | an unpleasant chore :) | 19:17 |
| jaypipes | gundlach: heh, no worries | 19:17 |
| jaypipes | gundlach, eday: are the 2 concurrency-related bugs fixed or still outstanding? if outstanding, can I assign them to someone? | 19:19 |
| gundlach | link? | 19:20 |
| gundlach | s | 19:20 |
| jaypipes | gundlach: https://bugs.launchpad.net/nova/+bug/649806 | 19:20 |
| uvirtbot | Launchpad bug 649806 in nova "AMQP concurrencty problem in API server" [High,New] | 19:20 |
| gundlach | oh, those | 19:20 |
| jaypipes | gundlach: https://bugs.launchpad.net/nova/+bug/649807 | 19:20 |
| uvirtbot | Launchpad bug 649807 in nova "Redis concurrency problems in API server" [High,New] | 19:20 |
| gundlach | i don't have state on those -- soren should | 19:20 |
| jaypipes | gundlach: k | 19:20 |
| _cerberus_ | gundlach: bunch of conflicts for me in sqlalchemy/api, so it's going to take me a bit to figure out what's going on | 19:24 |
| jaypipes | _cerberus_: likely because of the authorization hooks that devcamca- added... | 19:25 |
| *** pvo has joined #openstack | 19:34 | |
| *** ChanServ sets mode: +v pvo | 19:34 | |
| *** ctennis has quit IRC | 19:35 | |
| gundlach | _cerberus_: yeah, i had conflicts earlier there -- if it helps, i already figured out which code to keep (my renames and his code modifications and decorator additions.) | 19:40 |
| _cerberus_ | gundlach: I think I have it figured out. I'm getting a ton of deprecation warnings about the None contexts used all over the place when running the unit tests. Is that expected now? | 19:41 |
| gundlach | yes | 19:43 |
| gundlach | _cerberus_: somebody said something yesterday/today about that in the mailing lists -- maybe jaypipes? | 19:44 |
| _cerberus_ | Fair enough | 19:44 |
| gundlach | bug link: https://bugs.launchpad.net/bugs/655253 | 19:45 |
| uvirtbot | Launchpad bug 655253 in nova "Deprecation warnings about no request context should be addressed" [Low,Confirmed] | 19:45 |
| gundlach | jaypipes: text conflicts in https://code.launchpad.net/~jaypipes/nova/glance-image-service/+merge/37523 | 19:45 |
| jaypipes | _cerberus_: vishy is working on those warnings | 19:47 |
| uvirtbot | New bug: #607351 in nova "test_max_processes is not idempotent" [Undecided,Won't fix] https://launchpad.net/bugs/607351 | 19:48 |
| jaypipes | gundlach: cool, thx for the heads up. I'll get those fixed up. | 19:48 |
| *** pvo has quit IRC | 19:49 | |
| *** littleidea has quit IRC | 19:50 | |
| *** joearnold has quit IRC | 19:50 | |
| jaypipes | gundlach: ooops, the latest merge broke /nova/tests/api/rackspace/test_servers.py ... forgot to change stubs that call instance_get_by_ec2_id... | 19:51 |
| jaypipes | gundlach: I'll fix them up and push to the glance-image-service branch along with merge conflict fixups | 19:52 |
| gundlach | jaypipes: whoops, thanks! i thought you had integrated nosetests into run_tests so that should have been caught? | 19:52 |
| jaypipes | gundlach: nope, not yet | 19:53 |
| gundlach | wait a sec... | 19:56 |
| gundlach | _cerberus_: aren't you in the process of converting the RS API from ec2 to internal_id? since my branch didn't merge till just now you couldn't have pushed that code last thursday | 19:56 |
| _cerberus_ | Right, I just pushed it 10 seconds ago | 19:57 |
| gundlach | jaypipes: so you shouldn't touch the ec2_id that you found. | 19:57 |
| jaypipes | gundlach: ? why not? | 19:57 |
| gundlach | _cerberus_: cool, gimme a link to review when it's ready, looking forward to getting this done | 19:57 |
| gundlach | jaypipes: because the RS API hadn't been converted off of get_by_ec2_id yet | 19:57 |
| gundlach | and cerberus just pushed that branch | 19:57 |
| gundlach | so whoever merges second will have merge conflicts | 19:58 |
| jaypipes | gundlach: :( ok, I guess I'm not understanding... | 19:58 |
| jaypipes | I'll uncommit | 19:58 |
| gundlach | jaypipes: probably my fault: lemme try again | 19:58 |
| gundlach | jaypipes: your branch probably shouldn't mix your glance work, and an unrelated bug fix regarding ec2_ids, just for the sake of each branch being about one thing. | 19:58 |
| gundlach | jaypipes: and in this case, the ec2_ids are actually being cleaned up as we speak by cerberus, who just pushed the changes. | 19:58 |
| gundlach | i had converted the EC2 API from get_inst_by_ec2_id to get_inst_by_internal_id | 19:59 |
| *** pvo has joined #openstack | 19:59 | |
| *** ChanServ sets mode: +v pvo | 19:59 | |
| jaypipes | gundlach: true. but I can't merge trunk without breaking 100% of the tests :( | 19:59 |
| gundlach | cerberus is converting the RS API in the same way | 19:59 |
| jaypipes | gundlach: k | 19:59 |
| gundlach | jaypipes: really? it's not just that one test which isn't run by run_tests.py? now *i'm* not understanding :) | 19:59 |
| jaypipes | gundlach: ok, I understand now... | 19:59 |
| gundlach | k -- sorry for the confusion! | 20:00 |
| jaypipes | gundlach: cerberus stuff merge first, then I'll bzr pull and merge trunk, then push to glance-image-service :) | 20:00 |
| jaypipes | gundlach: no worries... | 20:00 |
| _cerberus_ | gundlach: https://code.launchpad.net/~cerberus/nova/api_cleanup/+merge/37657 | 20:00 |
| jaypipes | _cerberus_: doing it now.. | 20:00 |
| jonwood | Is there anywhere which provides a high level overview of how everything works together in Nova? | 20:00 |
| jaypipes | jonwood: not too much right now :( annegentle is working on doc improvements | 20:01 |
| jaypipes | _cerberus_: approve from me. gundlach, approve? if so, let's push it rhough... | 20:01 |
| gundlach | _cerberus_: aren't the ids passed to delete() etc strings, while instance_get_by_internal_id takes an int? | 20:01 |
| jonwood | Fair enough. I'll keep trying to work it out then. | 20:02 |
| _cerberus_ | gundlach: ahh, good call | 20:02 |
| jonwood | btw, the getting started docs seem to be quite considerably out of date. | 20:02 |
| gundlach | similarly, line 109 inst['hostname'] = ref.internal_id -- hostname is probably supposed to be a string. | 20:02 |
| gundlach | (not sure about that one) | 20:03 |
| gundlach | otherwise, looks good and i'll approve | 20:04 |
| _cerberus_ | You're right about the above comments. Let me fix those first | 20:04 |
| * jaypipes waits with baited breath. | 20:05 | |
| gundlach | jaypipes: Sally, having swallowed cheese / directs down holes the scented breeze / enticing thus with baited breath / nice mice to an untimely death. | 20:06 |
| _cerberus_ | pushed | 20:06 |
| *** alekibango_ is now known as alekibango | 20:07 | |
| gundlach | _cerberus_: if you addressed both things in my Approve comment, then lgtm | 20:07 |
| _cerberus_ | marked as approved | 20:08 |
| jaypipes | gundlach, _cerberus_ :) | 20:09 |
| * jaypipes slaps _cerberus_ for having an awkward IRC nick to type.... | 20:10 | |
| gundlach | jaypipes: oh, that's calling the kettle black -- you had to use *every single letter* in your first and last name, when 'jpipes' or the like would have saved us keystrokes! :) | 20:10 |
| alekibango | openstack meeting agenda page http://wiki.openstack.org/Meetings seems to be whole week old - what is planned for today? | 20:10 |
| openstack | alekibango: Error: "meeting" is not a valid command. | 20:10 |
| alekibango | ouch :) | 20:11 |
| *** miclorb_ has joined #openstack | 20:11 | |
| gundlach | alekibango: oh, hey, there's a meeting today -- thanks for the reminder :) | 20:11 |
| alekibango | in 50 minutes | 20:11 |
| _cerberus_ | jaypipes: use a client that was made after 1990 and you'll get tab completion :-P | 20:11 |
| gundlach | openstack gundlach:_Error:_"_:rorrE_:hcaldnug | 20:11 |
| openstack | gundlach: Error: "gundlach:_Error:_"_:rorrE_:hcaldnug" is not a valid command. | 20:11 |
| jaypipes | gundlach: lol | 20:12 |
| gundlach | darn, screwed up the palindrome | 20:12 |
| jaypipes | _cerberus_: I have tab completion...but typing _ is annoying :P | 20:12 |
| _cerberus_ | haha, fair enough | 20:12 |
| gundlach | jaypipes: then you should switch to ruby? | 20:12 |
| _cerberus_ | If I can get cerberus freed up on freenode, I'd gladly take it :-P | 20:12 |
| *** jaypipes is now known as ___________jaypi | 20:12 | |
| gundlach | lol | 20:12 |
| *** ___________jaypi is now known as jpipes | 20:12 | |
| jpipes | how's that? :) | 20:12 |
| gundlach | jaypipes: _cerberus_'s branch merged -- merge away | 20:13 |
| jpipes | dendrobates: yo. you running the meeting today? :) | 20:13 |
| jpipes | gundlach: rock on. | 20:13 |
| * gundlach missed the nick change, oops | 20:13 | |
| gundlach | jpipes: can you approve https://code.launchpad.net/~gundlach/nova/ec2_id_typo/+merge/37661 real quick | 20:17 |
| gundlach | one-word typo | 20:17 |
| alekibango | jonwood: look here -> http://wiki.openstack.org/ArchitecturalOverview + http://wiki.openstack.org/Overview | 20:18 |
| jpipes | gundlach: done | 20:19 |
| gundlach | ty | 20:19 |
| *** aliguori has quit IRC | 20:21 | |
| *** metcalfc has joined #openstack | 20:21 | |
| jpipes | gundlach, _cerberus_: http://paste.openstack.org/show/47/ Shows up now running nosetests. Likely due to internal id now returning an int? | 20:21 |
| _cerberus_ | Bah, ues | 20:22 |
| _cerberus_ | Yes even | 20:22 |
| jpipes | _cerberus_: :) | 20:23 |
| *** ttx has quit IRC | 20:23 | |
| jpipes | _cerberus_, gundlach: pls review (and approve if possible) the glance-image-service merge req. should have no more conflicts and all tests except above noted one pass | 20:23 |
| gundlach | jpipes: approved | 20:27 |
| jpipes | gundlach: cheers | 20:28 |
| gundlach | _cerberus_: i assume you're handling the paste above, or would you like me to? | 20:30 |
| _cerberus_ | I'm handling them, yes | 20:31 |
| *** allsystemsarego has quit IRC | 20:34 | |
| *** aliguori has joined #openstack | 20:36 | |
| *** ptremblett has quit IRC | 20:38 | |
| jonwood | alekibango: Cheers :) | 20:43 |
| *** pvo has quit IRC | 20:44 | |
| *** JimCurry has joined #openstack | 20:49 | |
| *** ddumitriu has joined #openstack | 20:49 | |
| *** befreax has joined #openstack | 20:50 | |
| gundlach | _cerberus_: left a comment w/ qs on your mergeprop | 20:51 |
| soren | chmouel: Ok. Then I'm pulling a blank. | 20:52 |
| *** adjohn has joined #openstack | 20:52 | |
| soren | gundlach: In response to "maybe I accidentally chose "Merged" instead of "Approved" after I got an approval review? How would I detect that?": E-mail. Launchpad sends an e-mail when the status is changed and you should be able to tell who did it based on that e-mail. | 20:53 |
| gundlach | soren: oh, good point, thanks | 20:53 |
| gundlach | yep, the email said it went from needs review => merged :) | 20:54 |
| soren | gundlach: There you go :) | 20:54 |
| gundlach | soren: do you know if we're meeting in 5? | 20:54 |
| soren | gundlach: We are. | 20:55 |
| gundlach | k | 20:55 |
| soren | Meeting in #openstack-meeting in 5 minutes, people! | 20:55 |
| *** ewanmellor has quit IRC | 20:57 | |
| *** dizz has joined #openstack | 20:58 | |
| *** befreax has joined #openstack | 20:59 | |
| soren | That means now. #openstack-meeting. Be there or be rectangular. | 21:01 |
| *** michaeldreamhost has joined #openstack | 21:05 | |
| *** pvo has joined #openstack | 21:07 | |
| *** ChanServ sets mode: +v pvo | 21:07 | |
| *** zooko has joined #openstack | 21:09 | |
| creiht | alekibango: btw, anyone should be able to comment on a merge proposal | 21:10 |
| alekibango | ic. ty | 21:14 |
| *** jdarcy has quit IRC | 21:22 | |
| *** befreax has quit IRC | 21:24 | |
| *** littleidea has joined #openstack | 21:27 | |
| alekibango | jc_smith: here for arch diagrams: http://wiki.openstack.org/ArchitecturalOverview + http://wiki.openstack.org/Overview | 21:33 |
| alekibango | there were also another, ascii diagram somewhere but i cant find it | 21:33 |
| alekibango | annegentle: ping? | 21:34 |
| jc_smith | yeah, its very minimal. There should be something that shows all the flags, what the options are, how they relate to one another | 21:34 |
| creiht | alekibango: she is in orientation training this week, so she may be in and out | 21:35 |
| alekibango | ty creiht | 21:35 |
| alekibango | jc_smith: +1, we need to WRITE this. volunteers? :) | 21:35 |
| jc_smith | I think the hesitation has been because so many things are in flux | 21:36 |
| creiht | if you are interested in documentation, please coordinate with annegentle, as we don't want to duplicate efforts | 21:37 |
| jc_smith | but we should be at a stable release checkpoint now | 21:37 |
| *** joearnold has joined #openstack | 21:39 | |
| *** ptremblett has joined #openstack | 21:39 | |
| *** sevenseeker has quit IRC | 21:39 | |
| *** sevenseeker has joined #openstack | 21:40 | |
| jonwood | Can anyone clarify what a "project" is in Nova terms? I'm assuming it's a logical grouping of VMs with communication to each other, but that's largely based on experience with EC2. | 21:41 |
| jc_smith | it has some networking implications as well, like if you use vlan networking, you have a vlan dedicated to a project | 21:42 |
| jonwood | That was what I meant by communication with each other. | 21:43 |
| jc_smith | if you use flat networking all the VMs are in the same networking, even if in different projects | 21:44 |
| jonwood | OK... it appears my install has defaulted to creating VPNs for things. Is taht normal? | 21:44 |
| jc_smith | yeah, I guess the use case at NASA is that you projects get some private address space that you normally need to VPN into in order to use your VMs | 21:46 |
| jc_smith | where at hosting companies probably just setup a gateway and VMs are publicly routable | 21:46 |
| jonwood | That makes sense. I'm still trying to get up and running, so havn't played with different networking options at all yet. | 21:47 |
| jc_smith | I'd start with flat networking, it will insert the IP into the network scripts of the guest. No DHCP or VLAN stuff needed | 21:49 |
| jc_smith | for starting out | 21:50 |
| jonwood | That may be what's happening anyway... I'm running through http://wiki.openstack.org/NovaInstallFestInstructions right now, but it's not entirely clear what half the options are actually doing. | 21:51 |
| vishy | jonwood: i have a branch in for flat networking + dhcp | 21:51 |
| vishy | which hopefully will land | 21:51 |
| *** ppetraki has quit IRC | 21:51 | |
| *** sevenseeker has quit IRC | 21:52 | |
| jc_smith | jonwood: That link is a bit outdated. I'd look at this as well http://github.com/vishvananda/novascript/blob/master/nova.sh | 21:54 |
| jc_smith | and see what it's doing, it more up to date | 21:54 |
| jonwood | Argh! | 21:55 |
| jonwood | (But thanks) ;) | 21:55 |
| jc_smith | In reality I had to look at three different docs to get it up, and look at the code, and repeat, then 2 weeks later the code changed significantly and had to do a bunch of things different, you aren't alone | 21:58 |
| *** mdomsch has quit IRC | 21:58 | |
| soren | vishy: So.. I've sent you a couple of messages in here over the last... 48 hours. | 21:58 |
| soren | vishy: Did you see them or should I sum up? | 21:58 |
| vishy | soren: apologies, We're releasing 1.0 80 hour week last week | 21:58 |
| vishy | soren: sum up | 21:59 |
| soren | vishy: No problem. | 21:59 |
| soren | vishy: So, I saw you made a bunch of changes. good stuff, by the way! I took the liberty of extracting them and putting them in my branch. | 21:59 |
| vishy | oh :) | 22:00 |
| vishy | you've been looking at the deploy branch? | 22:00 |
| soren | vishy: You may have made changes since I did that, so I'll take another look, but it should be mostly in order. | 22:00 |
| soren | vishy: Yes. I hope that's the right one. | 22:00 |
| dendrobates | sorry I've been MIA guy's. | 22:00 |
| dendrobates | I was in Paris at the open world forum and then at the OSGR in Barcelona | 22:00 |
| vishy | soren: that is where we're doing our last minute hacks. We've unfortunately diverged a bit, but i'm hoping we can submit all the stuff as bugfixes soon | 22:00 |
| soren | vishy: Yup, that's what I was expecting you'd do. | 22:01 |
| soren | vishy: So I figured I'd a) get a sneak preview and b) pitch in by taking at least that chunk off your back. | 22:01 |
| soren | The stuff I sent you yesterday: | 22:01 |
| soren | 20:03 <+soren> vishy (or whoever cares): I've attempted to extract all your security group related changes from ~anso/nasa/deploy (r379). It's all in my ec2-security-gro | 22:01 |
| soren | ups branch. | 22:01 |
| soren | 20:03 <+soren> vishy: So when you decide to go through your changes and push them, this part should be pretty much done. | 22:01 |
| soren | Just for reference. | 22:01 |
| soren | 20:04 <+soren> vishy: Notable exception: The vpn related changes. I was hoping you could do those separately. | 22:02 |
| vishy | awesome | 22:02 |
| vishy | yeah no problem | 22:02 |
| soren | My secret trick to deal with the diversion, by the way: | 22:02 |
| soren | bzr merge --weave. | 22:02 |
| soren | Took care of a LOT of conflicts. | 22:02 |
| vishy | i think those should go in to fix-cloudpipe | 22:02 |
| soren | vishy: Haven't looked at that yet, but it sounds plausible. | 22:03 |
| vishy | the vpn related stuff is pretty broken in trunk anyway | 22:03 |
| alekibango | btw vishy your novascript si not working well for me when i USE_VENV=1 -- did you try that recently? | 22:03 |
| vishy | so that is my branch while i fix everthing | 22:03 |
| vishy | alekibango: no i haven't tested with USE_VENV for some time | 22:04 |
| *** _maxim_ has left #openstack | 22:04 | |
| vishy | are you using MYSQL? | 22:04 |
| *** Cybodog has joined #openstack | 22:04 | |
| alekibango | vishy: no, sqlite | 22:04 |
| alekibango | paths are messed up somehow somwhere | 22:04 |
| alekibango | vishy: and later psql | 22:04 |
| vishy | i'll go through and clean it up at some point when i actually have some time | 22:04 |
| alekibango | vishy: i might have few small improvements laterz | 22:04 |
| vishy | soren: i didn't know about weave | 22:04 |
| vishy | soren: i do know that i have been really really annoyed with bzr's ability to merge | 22:05 |
| alekibango | soren: +1 for git :) | 22:05 |
| vishy | soren: duplications, re-fixing the same merge errors multiple times, etc. | 22:05 |
| soren | vishy: I think we're seeing the absolute worst of it, because everyone is working in all the same files at all the same places. | 22:05 |
| vishy | soren: agreed | 22:05 |
| soren | vishy: See "bzr help criss-cross" for an explanation about --weave. | 22:06 |
| vishy | k | 22:06 |
| vishy | soren: devin put in a fix for catching duplicated rules | 22:07 |
| soren | vishy: I saw. | 22:07 |
| soren | vishy: ..and merged. :) | 22:07 |
| vishy | soren: we are having libvirt hang though | 22:08 |
| soren | vishy: Oh, right, that was the other thing. | 22:08 |
| soren | vishy: I've seen that too. | 22:08 |
| soren | vishy: ..and have a vague idea what's going on. | 22:08 |
| soren | vishy: I'll try to attack it tomorrow. | 22:08 |
| vishy | is it related to security groups? | 22:09 |
| soren | Yes. | 22:09 |
| soren | It's a locking issue in the nwfilter code :( | 22:09 |
| vishy | doh | 22:09 |
| vishy | is there a workaround? | 22:09 |
| vishy | can we lock in our code somewhere? | 22:09 |
| soren | Not that I know of. | 22:09 |
| vishy | damn | 22:10 |
| soren | vishy: I expect to nail it tomorrow, though. | 22:10 |
| soren | ..but of course, now that I've said that, I won't. | 22:10 |
| soren | Meh. | 22:10 |
| soren | vishy: It's very near the top of my list for tomorrow. | 22:10 |
| vishy | i guess we have to disable them for the moment then | 22:10 |
| vishy | :( | 22:10 |
| soren | vishy: How often do you see it? | 22:10 |
| vishy | under load, it is happening every 10 minutes or so | 22:11 |
| vishy | if we start launching and terminating tons of instances | 22:11 |
| soren | I could fire up and kill around 40 VM's before it started happening, but once it got into it, it was pretty consistent. | 22:11 |
| vishy | does it happen even if you aren't firing them concurrently? | 22:12 |
| soren | vishy: Not sure. | 22:12 |
| vishy | i'm wondering if we could just lower the threadpool to 1 | 22:12 |
| soren | vishy: I didn't really care, to be honest. I was more interested in making it happen more often rather than less. | 22:12 |
| soren | Once I start working on debugging it, having a sure fire way to reproduce it is more important. | 22:13 |
| vishy | well if it is a locking/concurrencly issue, only running one libvirt command at a time may fix it | 22:13 |
| vishy | which is annoying for one box, but considering our deployment, we could handle until you can fix the underlying problem | 22:14 |
| soren | vishy: Sounds like a hypothesis worth testing. | 22:14 |
| soren | vishy: for you. I'll focus on actually fixing it. | 22:15 |
| *** metcalfc has quit IRC | 22:15 | |
| soren | vishy: Let me know how it goes. It might turn out to be a useful data point. | 22:15 |
| vishy | ok | 22:15 |
| *** rnirmal has quit IRC | 22:15 | |
| *** pvo has quit IRC | 22:16 | |
| * soren heads bedwards | 22:19 | |
| soren | G'night guys! | 22:19 |
| *** aliguori has quit IRC | 22:25 | |
| *** kevnfx has joined #openstack | 22:26 | |
| *** whack has joined #openstack | 22:28 | |
| *** ctennis has joined #openstack | 22:30 | |
| zooko | gdusbabek: great! Thanks. | 22:30 |
| zooko | wrong channel. | 22:31 |
| *** hazmat has quit IRC | 22:35 | |
| *** whack has left #openstack | 22:36 | |
| *** hazmat has joined #openstack | 22:36 | |
| *** sevenseeker has joined #openstack | 22:37 | |
| *** Cybodog has quit IRC | 22:40 | |
| *** hazmat has quit IRC | 22:40 | |
| kevnfx | Is there support/documentation for dynamically adding/removing physical node resources from a nova configuration? | 22:42 |
| jonwood | I'm getting a 404 error when trying to register an image with Nova - any hints? | 22:42 |
| jonwood | DEBUG:routes.middleware:No route matched for GET /services/Cloud/ | 22:42 |
| *** dizz has quit IRC | 22:43 | |
| *** dizz has joined #openstack | 22:43 | |
| *** dizz_ has joined #openstack | 22:46 | |
| uvirtbot | New bug: #655401 in swift "swift-bench doesn't clean up containers" [Undecided,New] https://launchpad.net/bugs/655401 | 22:48 |
| *** dizz has quit IRC | 22:49 | |
| kevnfx | jonwood: I had a similar problem a few days back that Vishy helped me with. Check the irc log from October 1st, there were several things he walked me through. In particular, I needed to set my HOST_IP environment variable before running nova.sh run ... | 22:49 |
| vishy | yeah actually there is a bug in that | 22:49 |
| vishy | HOST_IP logic is backwards | 22:49 |
| jonwood | kevnfx: Thanks. I think I'm going to leave it for tonight, and give it another try tomorrow. | 22:51 |
| *** jonwood has quit IRC | 22:51 | |
| vishy | pushing the fix for that | 22:52 |
| vishy | jonwood: people have been complaining about register not working | 22:52 |
| vishy | make sure you have the newest version of eventlet installed | 22:52 |
| vishy | and make sure your conf has --FAKE_subdomain=ec2 | 22:52 |
| *** johnbergoon has quit IRC | 22:53 | |
| kevnfx | vishy: while I see you're here .... is there an "easy" way to dynamically add a node resource to a nova configuration? | 22:55 |
| *** aliguori has joined #openstack | 22:55 | |
| *** ddumitriu has quit IRC | 22:56 | |
| vishy | node resource? | 22:56 |
| vishy | not sure what you mean? | 22:56 |
| kevnfx | I have a single node configured, and want to add a second physical node in order to startup more instances ... | 22:57 |
| kevnfx | That is, what daemons do I need to start on the new node, and can I simply point them to the "parent" node in order to get them engaged ... | 22:59 |
| kevnfx | If it's not an easy thing to do that's fine ... just curious if it was currently supported ... | 22:59 |
| kevnfx | Alternately, is there any info on starting/configuring multiple physical nodes (as opposed to the single node that nova.sh currently supports)? | 23:04 |
| *** dendrobates is now known as dendro-afk | 23:05 | |
| *** dizz_ has quit IRC | 23:06 | |
| *** dizz has joined #openstack | 23:07 | |
| kevnfx | vishy: is that still unclear? | 23:07 |
| vishy | ah gotcha | 23:08 |
| vishy | you only need compute on the other node | 23:08 |
| vishy | and if it has the right settings for s3_host, redis_host, rabbit_host, ldap_url, sql_connection | 23:09 |
| vishy | it will work | 23:09 |
| vishy | but you can't use sqlite | 23:09 |
| vishy | has to be a real db | 23:09 |
| kevnfx | vishy: I tried that, and vectored it off to the other node in its cli arguments, but it didn't seem to be utilized when resources were getting tight on the initial node. Where do I set the settings you describe? | 23:09 |
| vishy | well you can specify them on the command line | 23:10 |
| kevnfx | vishy: nova-manage.conf? | 23:10 |
| vishy | or you can put them in a flagfile and --flagfile=<path to flagfile> | 23:10 |
| kevnfx | cool ... let me take a look at that .... thanks! | 23:11 |
| vishy | the scheduler should pick the least loaded compute when you run an instance | 23:11 |
| kevnfx | right, that's what I expected ... I'll add the additional flags and see what happens ... | 23:12 |
| *** Cybodog has joined #openstack | 23:12 | |
| *** sevenseeker has quit IRC | 23:13 | |
| *** johnpur has quit IRC | 23:13 | |
| *** sevenseeker has joined #openstack | 23:17 | |
| *** rlucio has quit IRC | 23:18 | |
| *** kevnfx has quit IRC | 23:19 | |
| *** dpackard has quit IRC | 23:23 | |
| *** dendro-afk is now known as dendrobates | 23:24 | |
| *** gundlach has quit IRC | 23:26 | |
| *** mtaylor has quit IRC | 23:27 | |
| *** miclorb_ has quit IRC | 23:28 | |
| *** miclorb has joined #openstack | 23:32 | |
| *** dizz is now known as dizz|away | 23:33 | |
| *** hazmat has joined #openstack | 23:34 | |
| *** aliguori has quit IRC | 23:35 | |
| *** dizz has joined #openstack | 23:39 | |
| *** dizz|away has quit IRC | 23:42 | |
| *** miclorb has quit IRC | 23:46 | |
| *** JimCurry has quit IRC | 23:46 | |
| *** dendrobates is now known as dendro-afk | 23:46 | |
| *** sevenseeker has quit IRC | 23:47 | |
| *** abecc has quit IRC | 23:52 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!