| *** abecc has quit IRC | 00:00 | |
| *** ali``` has joined #openstack | 00:02 | |
| mtaylor | creiht, gholt: fixed | 00:05 |
|---|---|---|
| *** abecc has joined #openstack | 00:06 | |
| *** ali```_ has joined #openstack | 00:08 | |
| *** silassewell has joined #openstack | 00:10 | |
| *** ali``` has quit IRC | 00:10 | |
| *** ali```_ is now known as ali``` | 00:10 | |
| *** zooko has quit IRC | 00:11 | |
| *** pcrews has quit IRC | 00:12 | |
| *** jdmaturen has quit IRC | 00:15 | |
| *** tobym has joined #openstack | 00:15 | |
| *** ali```_ has joined #openstack | 00:16 | |
| *** maplebed has quit IRC | 00:17 | |
| *** ali``` has quit IRC | 00:19 | |
| *** ali```_ is now known as ali``` | 00:19 | |
| *** adjohn has joined #openstack | 00:25 | |
| *** heckj has quit IRC | 00:38 | |
| *** e1mer has joined #openstack | 00:53 | |
| *** RobertLJ has joined #openstack | 01:01 | |
| *** zooko has joined #openstack | 01:01 | |
| *** pvo has joined #openstack | 01:06 | |
| *** ChanServ sets mode: +v pvo | 01:06 | |
| *** RobertLJ has quit IRC | 01:21 | |
| *** RobertLJ has joined #openstack | 01:23 | |
| *** lmcdowell has joined #openstack | 01:28 | |
| *** pvo has quit IRC | 01:32 | |
| *** binaryWarrior has joined #openstack | 01:35 | |
| *** binaryWarrior has joined #openstack | 01:36 | |
| *** RobertLJ has quit IRC | 01:36 | |
| *** RobertLJ has joined #openstack | 01:38 | |
| *** zooko has quit IRC | 01:41 | |
| *** jamiew has quit IRC | 01:49 | |
| creiht | mtaylor: awesome, thanks! | 01:50 |
| *** jdmaturen has joined #openstack | 01:51 | |
| *** e1mer has quit IRC | 01:56 | |
| gholt | Okay cool, I'm going to try to get the new functional tests in now. | 01:58 |
| *** gustavomzw has quit IRC | 02:09 | |
| *** RobertLJ has quit IRC | 02:10 | |
| *** binaryWarrior has quit IRC | 02:16 | |
| *** RobertLJ has joined #openstack | 02:25 | |
| gholt | Okay, looks like it's all cleaned up now. Sorry for the lotsa-emails. | 02:35 |
| *** e1mer has joined #openstack | 02:38 | |
| mtaylor | gholt: thanks, btw - that problem helped fix up a part of how we were using tarmac | 02:38 |
| gholt | Heh, I guess I'm good at doing the unexpected. | 02:39 |
| *** RobertLJ has quit IRC | 02:41 | |
| *** e1mer has quit IRC | 02:42 | |
| *** ali``` has quit IRC | 02:45 | |
| *** zooko has joined #openstack | 03:06 | |
| *** e1mer has joined #openstack | 03:17 | |
| *** pvo has joined #openstack | 03:21 | |
| *** ChanServ sets mode: +v pvo | 03:21 | |
| *** sirp1 has quit IRC | 03:38 | |
| *** sirp1 has joined #openstack | 03:49 | |
| *** kashyapc has joined #openstack | 03:51 | |
| *** littleidea has quit IRC | 03:59 | |
| *** littleidea has joined #openstack | 04:01 | |
| *** miclorb has quit IRC | 04:01 | |
| *** adjohn has quit IRC | 04:02 | |
| *** jeevan_ullas has quit IRC | 04:09 | |
| *** jeevan_ullas has joined #openstack | 04:09 | |
| *** miclorb has joined #openstack | 04:15 | |
| *** pvo has quit IRC | 04:20 | |
| *** silassewell has quit IRC | 04:22 | |
| *** abecc has quit IRC | 04:25 | |
| *** miclorb has quit IRC | 04:43 | |
| *** miclorb has joined #openstack | 05:12 | |
| *** adjohn has joined #openstack | 05:30 | |
| *** ibarrera has joined #openstack | 05:53 | |
| *** Podilarius has joined #openstack | 05:57 | |
| *** guigui has joined #openstack | 06:02 | |
| *** zooko has quit IRC | 06:05 | |
| *** jdmaturen has quit IRC | 06:09 | |
| *** ChrisAM has quit IRC | 06:09 | |
| *** timrc_ has joined #openstack | 06:09 | |
| *** chromakode has quit IRC | 06:10 | |
| *** comstud has quit IRC | 06:10 | |
| *** stewart_ has joined #openstack | 06:11 | |
| *** jpipes has joined #openstack | 06:11 | |
| *** allsystemsarego has joined #openstack | 06:12 | |
| *** brainproxy has quit IRC | 06:13 | |
| *** stewart has quit IRC | 06:13 | |
| *** jaypipes has quit IRC | 06:13 | |
| *** ChrisAM` has joined #openstack | 06:13 | |
| *** ChrisAM` has joined #openstack | 06:13 | |
| *** ChrisAM` is now known as ChrisAM | 06:13 | |
| *** chromakode has joined #openstack | 06:13 | |
| *** littleidea has quit IRC | 06:14 | |
| *** littleidea has joined #openstack | 06:15 | |
| *** brainproxy has joined #openstack | 06:29 | |
| *** littleidea has quit IRC | 06:34 | |
| *** adjohn has quit IRC | 06:50 | |
| *** adjohn has joined #openstack | 06:54 | |
| *** calavera has joined #openstack | 07:04 | |
| *** sirp1 has quit IRC | 07:05 | |
| *** almaisan-away is now known as al-maisan | 07:12 | |
| *** rnewson has joined #openstack | 07:15 | |
| *** rnewson has joined #openstack | 07:15 | |
| *** justinsheehy has quit IRC | 07:21 | |
| *** adjohn has quit IRC | 07:22 | |
| *** justinsheehy has joined #openstack | 07:27 | |
| *** brd_from_italy has joined #openstack | 07:28 | |
| *** brd_from_italy has quit IRC | 07:35 | |
| *** miclorb has quit IRC | 08:11 | |
| *** zheng_li has joined #openstack | 08:19 | |
| *** rnewson has quit IRC | 08:20 | |
| *** justinsheehy has quit IRC | 08:23 | |
| *** rnewson has joined #openstack | 08:25 | |
| *** rnewson has joined #openstack | 08:25 | |
| *** justinsheehy has joined #openstack | 08:30 | |
| *** fsoyer has joined #openstack | 08:33 | |
| *** fsoyer has left #openstack | 08:35 | |
| *** rajijoom has joined #openstack | 08:36 | |
| *** fsoyer has joined #openstack | 08:41 | |
| *** ptremblett has quit IRC | 08:55 | |
| *** miclorb has joined #openstack | 09:54 | |
| *** miclorb has quit IRC | 09:56 | |
| *** miclorb_ has joined #openstack | 09:58 | |
| *** ptremblett has joined #openstack | 10:01 | |
| *** Guest96517 has joined #openstack | 10:11 | |
| *** rsampaio has joined #openstack | 10:19 | |
| *** gustavomzw has joined #openstack | 10:42 | |
| *** justinsheehy has quit IRC | 10:54 | |
| *** justinsheehy has joined #openstack | 11:00 | |
| fsoyer | Hi all. Sorry il it's a stupid question, but after reading all I found about OpenStack (nice and so interesting project !) I don't understand what is the last layer of the architecture : which virtualization software can be used ?? | 11:05 |
| redbo | fsoyer: I believe it's anything supported by libvirt, which means most hypervisors http://libvirt.org/ | 11:27 |
| *** Guest96517 has quit IRC | 11:37 | |
| *** ctennis has quit IRC | 11:38 | |
| *** JPP has joined #openstack | 11:38 | |
| *** JPP is now known as Guest22623 | 11:38 | |
| *** miclorb_ has quit IRC | 11:42 | |
| *** kashyapc has quit IRC | 11:42 | |
| *** justinsheehy has quit IRC | 11:46 | |
| *** jpopi has joined #openstack | 11:48 | |
| fsoyer | @redbo : thanks. So KVM and Xen, maybe some others ? Is VMware definitely excluded ? (technically or/and phylosophically ?) | 11:55 |
| larissa | fsoyer: Error: "redbo" is not a valid command. | 11:55 |
| fsoyer | ooops.... | 11:56 |
| *** sandywalsh has joined #openstack | 11:56 | |
| *** ctennis has joined #openstack | 11:58 | |
| *** kashyapc has joined #openstack | 11:59 | |
| alekibango | fsoyer: nothing is excluded. feel free to add support, like some people did with virtualbox | 12:00 |
| *** sophiap has quit IRC | 12:00 | |
| alekibango | for now it supports xen, kvm (qemu) and virtualbox | 12:01 |
| *** sophiap has joined #openstack | 12:04 | |
| *** sophiap has quit IRC | 12:14 | |
| fsoyer | alekibango : OK I see. I'm more a technician and (beta) tester, not a developer, so I hope that some geek will find an interest in vmware :) But maybe it's time to re-test kvm or xen. 3 years ago I examine them but was not convinced by the functionalities. | 12:17 |
| *** abecc has joined #openstack | 12:21 | |
| *** jpopi has quit IRC | 12:26 | |
| *** jpopi has joined #openstack | 12:26 | |
| soren | termie: Is this still on your radar? https://code.edge.launchpad.net/~justin-fathomdb/nova/check-subprocess-exit-code/+merge/30707 | 12:32 |
| *** justinsheehy has joined #openstack | 12:32 | |
| soren | fsoyer: It also supports user-mode-linux. | 12:32 |
| fsoyer | soren : OK but my needs include Windows VMs... I see on the libvirt page that VMWare ESX is supported ??? Well, I didn't think it was possible, in my mind libvirt was only attach to linux... And forgot that ESX *is* now linux based ! | 12:39 |
| soren | fsoyer: Mind you, using ESX with Nova isn't just a click of a button just because libvirt supports it. | 12:42 |
| *** DuSTIX has joined #openstack | 12:44 | |
| soren | fsoyer: libvirt is an API for managing (running, stopping, starting, etc) virtual machines (and a bunch of things wrt. to storage, but that's separate to this). *Creating* VM's is a different matter. | 12:45 |
| soren | fsoyer: Different hypervisors require different formats for their VM's and require different information to get them running. Right now, nova's libvirt support handles QEmu style hypervisors and user-mode linux. Nova also has a Xen API backend that talks to Xen server. | 12:46 |
| soren | fsoyer: Anything other than that will require work. | 12:46 |
| soren | fsoyer: Probably not much work, but still work. | 12:46 |
| soren | fsoyer: It's not just a configuration issue. | 12:46 |
| *** pcrews has joined #openstack | 12:47 | |
| fsoyer | soren : I understand , it's why I think that maybe, yes, if I find some time I'll try to play with vmware, after trying OpenStack/Xen or KVM, and if no heavy development is necessary... | 12:49 |
| fsoyer | But let start with a "default" Nova/Switch architecture ! | 12:51 |
| *** zheng_li has quit IRC | 12:55 | |
| *** gdusbabek has left #openstack | 12:56 | |
| *** jdarcy has joined #openstack | 12:57 | |
| *** zheng_li has joined #openstack | 12:59 | |
| *** aliguori has joined #openstack | 13:01 | |
| *** holoway has quit IRC | 13:01 | |
| *** pvo has joined #openstack | 13:02 | |
| *** ChanServ sets mode: +v pvo | 13:02 | |
| *** sophiap has joined #openstack | 13:02 | |
| *** holoway has joined #openstack | 13:03 | |
| *** holoway has quit IRC | 13:06 | |
| *** holoway has joined #openstack | 13:08 | |
| *** pvo has quit IRC | 13:11 | |
| *** zooko has joined #openstack | 13:12 | |
| *** dendro-afk is now known as dendrobates | 13:25 | |
| termie | soren: nope, i'll look at it again | 13:26 |
| *** pvo has joined #openstack | 13:26 | |
| *** ChanServ sets mode: +v pvo | 13:26 | |
| *** DuSTIX has quit IRC | 13:28 | |
| *** fsoyer has quit IRC | 13:35 | |
| *** rajijoom has quit IRC | 13:46 | |
| *** kashyapc has quit IRC | 13:47 | |
| *** zooko has quit IRC | 13:55 | |
| *** kashyapc has joined #openstack | 13:55 | |
| *** mrayzenoss has joined #openstack | 13:56 | |
| *** mrayzenoss is now known as mray | 13:56 | |
| *** mray has joined #openstack | 13:56 | |
| gholt | mtaylor: How should I test debian packaging? The swift package is failing due to a new .png. I was going to fix the problem by doing what it suggests (adding it to include-binaries) but I'd like to be able to test things out before pushing. | 13:57 |
| *** gundlach has joined #openstack | 13:57 | |
| gundlach | morning | 13:58 |
| gundlach | Anyone know how to tell Hudson that there's a new package requirement? He can't merge lp:~gundlach/nova/rsapi because he doesn't have 'routes' installed, but it's in tools/pip-requires. | 13:59 |
| *** pcrews is now known as pcrews_errand | 14:02 | |
| *** dendrobates is now known as dendro-afk | 14:13 | |
| *** dendro-afk is now known as dendrobates | 14:16 | |
| *** holoway has quit IRC | 14:20 | |
| *** holoway has joined #openstack | 14:21 | |
| *** johnpur has joined #openstack | 14:24 | |
| *** ChanServ sets mode: +v johnpur | 14:24 | |
| soren | gundlach: You tell mtaylor, and magic happens. | 14:29 |
| gundlach | soren: is that the way we want it, or should i put in a bug to get that fixed? | 14:29 |
| gundlach | soren: also: thanks :) | 14:29 |
| soren | gundlach: I'm reasonably sure that's how we want it. | 14:33 |
| *** JimCurry has joined #openstack | 14:33 | |
| gundlach | k | 14:33 |
| soren | Hm... I think. | 14:34 |
| *** Glace has quit IRC | 14:37 | |
| *** sirp1 has joined #openstack | 14:40 | |
| *** Glace has joined #openstack | 14:42 | |
| mtaylor | gundlach: adding now - will figure out a better way to do that moving forward | 14:43 |
| gundlach | tyvm! i emailed you the full list (I think there are 3) | 14:43 |
| *** holoway has quit IRC | 14:44 | |
| *** holoway has joined #openstack | 14:45 | |
| *** mray has quit IRC | 14:47 | |
| *** holoway has quit IRC | 14:49 | |
| zul | is there a way to increase the logging level of nova-compute i have the sneaky suscpion that something is wrong | 14:49 |
| *** holoway has joined #openstack | 14:50 | |
| *** littleidea has joined #openstack | 14:54 | |
| soren | zul: What do you have now? | 15:01 |
| zul | im just using --nodaemon and --verbose | 15:01 |
| zul | but the image isnt loading or anything | 15:02 |
| zul | this is with the latest bzr btw | 15:02 |
| *** zooko has joined #openstack | 15:03 | |
| *** mrayzenoss has joined #openstack | 15:05 | |
| *** mrayzenoss is now known as mray | 15:05 | |
| *** mray has joined #openstack | 15:05 | |
| *** guigui has quit IRC | 15:07 | |
| soren | --nodaemon --verbose should give you everything. | 15:07 |
| *** tobym has quit IRC | 15:08 | |
| *** JimCurry has left #openstack | 15:10 | |
| mtaylor | gholt: pull the lp:swift/debian - make the change, add an entry to the changelog with dch --nmu (you'll need the nmu since you're not listed in the uploaders) , then run bzr bd -S | 15:12 |
| *** JimCurry has joined #openstack | 15:17 | |
| *** ibarrera has quit IRC | 15:23 | |
| *** tobym has joined #openstack | 15:26 | |
| *** pcrews_errand is now known as pcrews | 15:29 | |
| gholt | mtaylor: Okay, I'll work through that today. | 15:29 |
| *** calavera has quit IRC | 15:30 | |
| *** fsoyer has joined #openstack | 15:31 | |
| *** tr3buchet has joined #openstack | 15:31 | |
| mtaylor | gholt: we can also add you to the uploaders list in debian/control if you're going to be hacking on it and stuff | 15:33 |
| *** holoway has quit IRC | 15:37 | |
| *** holoway has joined #openstack | 15:38 | |
| *** hazmat has quit IRC | 15:40 | |
| gholt | mtaylor: Well, I'm very newb when it comes to debian packaging, but yeah I'll try to keep on it. | 15:41 |
| * mtaylor needs to get nova and swift submitted to debian ... | 15:41 | |
| gholt | Also, the bzr bd -S failed on dpkg-buildpackage -rfakeroot -d -us -uc -S with little info about why, but I can run that manually just fine. | 15:41 |
| *** holoway has quit IRC | 15:42 | |
| *** holoway has joined #openstack | 15:43 | |
| *** JimCurry has quit IRC | 15:49 | |
| *** jpopi has left #openstack | 15:52 | |
| gholt | mtaylor: What do you think of: http://bazaar.launchpad.net/~gholt/swift/debiancyberduckhowtopng/revision/54 | 15:54 |
| *** holoway has quit IRC | 15:56 | |
| mtaylor | gholt: two things - now that you added yourself to debian/control, you don't need the --nmu for dch ... and you can go ahead and make the version number 1.0.2-3 ... we use the nmu revs for the automatic hudson merges | 15:56 |
| mtaylor | gholt: also, if you could change the distro series from lucid to unstable - the packaging trunk is targeted at debian and then there's an ubuntu branch we keep for targetting ubuntu | 15:57 |
| mtaylor | gholt: other than that - it looks great! | 15:57 |
| gholt | Ah, heh. Didn't even notice the lucid, okay. | 15:58 |
| *** holoway has joined #openstack | 15:58 | |
| gundlach | anyone know what would cause ReactorAlreadyRunning errors in twisted? I must have touched something related to twisted in my branch, because I'm getting these errors in the unittests. | 16:05 |
| *** RobertLJ has joined #openstack | 16:09 | |
| mtaylor | gholt: yeah - dch on ubuntu is ... 'helpful' | 16:10 |
| *** metoikos has quit IRC | 16:10 | |
| eday | gundlach: eventlet can use twisted under the hood, so if you try to run them both you may get issues | 16:11 |
| *** metoikos has joined #openstack | 16:13 | |
| gundlach | eday: thanks, i just figured that out right when you said that :) the only place we import eventlet is nova.wsgi, but that must get pulled in somewhere else that is called by the unittests. | 16:13 |
| *** avsm has joined #openstack | 16:14 | |
| gundlach | yep, nova.endpoint.__init__ | 16:16 |
| creiht | gundlach: eventlet will detect if twisted is imported, and if so by default will try to use the twisted hub, if that is the problem, you can tell it which hub to use to override it | 16:17 |
| gundlach | creiht: i think the problem is that eventlet starts up the twisted reactor? because then when the unittest tries to start the reactor manually it bombs | 16:18 |
| creiht | that part, I'm not entirely sure, as I have never actually used that part | 16:19 |
| creiht | gundlach: What I'm wondering is if you add something like this right after the imports in the test file, if it fixes the issue: http://paste.openstack.org/show/14/ | 16:20 |
| gundlach | creiht: will try it. though do we want to be hacking the test to avoid this problem? i'm making an existing test (many AWS EC2 API tests) fail, so i don't know that i should just work around it in the test file. | 16:22 |
| *** littleidea has quit IRC | 16:22 | |
| gundlach | i think this is showing me that as the code stands now, eventlet is getting in the way of twisted's normal operation, or something | 16:22 |
| gundlach | (didn't seem to make a difference anyway :) ) | 16:23 |
| gundlach | i'll keep digging | 16:23 |
| creiht | gundlach: That depends if the error is being introduced because of how twisted and eventlet are being used in the tests, or how they are being used in the actual code | 16:24 |
| creiht | k | 16:24 |
| gundlach | tx for the suggestions | 16:24 |
| *** zaitcev has joined #openstack | 16:25 | |
| *** jdmaturen has joined #openstack | 16:26 | |
| *** jdmaturen has joined #openstack | 16:27 | |
| *** littleidea has joined #openstack | 16:28 | |
| gundlach | creiht: what happens when in a nova venv you do this: | 16:36 |
| gundlach | python | 16:36 |
| gundlach | >>> import twisted.internet.reactor | 16:37 |
| gundlach | >>> import eventlet | 16:37 |
| gundlach | ? | 16:37 |
| termie | gundlach: same thing that would happen if you had them both installed | 16:37 |
| creiht | gundlach: I have no idea, I don't have nova set up :) | 16:37 |
| *** RobertLJ has quit IRC | 16:37 | |
| gundlach | creiht: heh, ok | 16:37 |
| gundlach | termie: ImportError: cannot import name getcurrent | 16:37 |
| gundlach | what am i missing? | 16:38 |
| gundlach | [from eventlet/hubs/twistdr.py's line "from eventlet import getcurrent, greenlet" | 16:38 |
| gundlach | ] | 16:38 |
| termie | gundlach, i am guessing that eventlet is trying to override the twisted reactor | 16:38 |
| termie | i don't get that error however | 16:39 |
| gundlach | termie: glad to hear that the error seems weird. guess i'll go looking in eventlet's source. | 16:39 |
| *** aliguori has quit IRC | 16:39 | |
| termie | gundlach: it may be a 0.9.10 error for eventlet | 16:39 |
| termie | my eventlet is 0.9.9 since i haven't updated the venv since bfore eventlet was added i guess | 16:40 |
| gundlach | termie: is 0.9.10 supposed to be flaky? (It's a "stability release" according to their blog) | 16:40 |
| creiht | gundlach: The eventlet/twisted integration was abandoned by the original author, and hasn't really been maintained | 16:40 |
| termie | one sec | 16:40 |
| creiht | so it may very well cause weird issues | 16:40 |
| creiht | gundlach: .9.10 is stable | 16:41 |
| termie | updating mine and checking | 16:41 |
| gundlach | creiht: good to know, thanks. maybe the right thing to do is explicitly keep eventlet from using twisted, in the actual code not just the tests | 16:41 |
| gundlach | termie: ty | 16:41 |
| creiht | gundlach: yeah that may be a good idea, I'm install twisted on my laptop so I can see what happens | 16:41 |
| termie | yup, newer version gets that error | 16:42 |
| creiht | gundlach: yeah I reproduced it as well | 16:43 |
| gundlach | termie: but not 0.9.9? | 16:43 |
| gundlach | there was a "fix for twisted hub" that went into 0.9.9 | 16:43 |
| termie | not that i can tell | 16:43 |
| gundlach | thanks both for your help! | 16:43 |
| gundlach | i'll see if i can work around it, and worst case we'll use 0.9.9 until we've teased apart twisted and eventlet more | 16:43 |
| termie | getcurrent should be in the greenthread library | 16:44 |
| termie | or rather, module | 16:44 |
| gundlach | termie: looks like it's a circular import problem -- getcurrent isn't defined in eventlet/__init__.py before the import loop cycles | 16:45 |
| termie | it doesn't happen in the opposite order, for the record | 16:45 |
| termie | and e comes before t | 16:46 |
| *** pvo has quit IRC | 16:46 | |
| termie | coincidence? I THINK NOT | 16:46 |
| creiht | gundlach: In my python shell if I import eventlet.hubs, and then do eventlet.hubs.use_hub('poll'), import eventlet works for me | 16:46 |
| gundlach | creiht: i'll try it again, i must have put it in the wrong place. thx. | 16:46 |
| gundlach | termie: deep | 16:47 |
| termie | gundlach: it is good sometimes to explore the rich meaning in the worl daround us | 16:47 |
| *** RobertLJ has joined #openstack | 16:47 | |
| *** stewart_ has quit IRC | 16:53 | |
| *** thobbs has joined #openstack | 16:56 | |
| gundlach | creiht: i think your "import eventlet.hubs" thing works because that implicitly imports eventlet, so you're importing eventlet before you import twisted.internet.reactor. | 16:58 |
| *** RobertLJ has quit IRC | 16:59 | |
| *** holoway has quit IRC | 17:01 | |
| *** holoway has joined #openstack | 17:03 | |
| *** zooko has quit IRC | 17:05 | |
| *** holoway has quit IRC | 17:06 | |
| *** stewart_ has joined #openstack | 17:06 | |
| *** holoway has joined #openstack | 17:08 | |
| *** maplebed has joined #openstack | 17:09 | |
| *** jamiew has joined #openstack | 17:10 | |
| *** zheng_li has quit IRC | 17:15 | |
| soren | mtaylor: Were you working on an eventlet package for Ubuntu? | 17:17 |
| soren | mtaylor: Or did I just dream that? | 17:17 |
| *** pvo has joined #openstack | 17:19 | |
| *** ChanServ sets mode: +v pvo | 17:19 | |
| mtaylor | soren: I believe you just dreamt that - although I could certainly do if needed | 17:21 |
| soren | mtaylor: That would be magnificent. | 17:21 |
| soren | mtaylor: We're already past feature freeze in Ubuntu, so getting new stuff like this in is getting harder by the day. | 17:22 |
| soren | mtaylor: ...and I'm working on other packaging bits (as you may have noticed if you're following the ubuntu-packaging branch at all) | 17:22 |
| mtaylor | yes. I noticed | 17:23 |
| soren | Anyways, /me calls it a day | 17:23 |
| mtaylor | soren: I was going to suggest we make a slight change to how we're doing packaging branches - but I'll send you an email about it | 17:23 |
| soren | Who am I kidding? I'll go eat some dinner and be back in a few hours. | 17:23 |
| mtaylor | soren: heh | 17:23 |
| soren | mtaylor: We can chat in a few hours if you prefer. | 17:23 |
| * soren heads dinnerwards | 17:23 | |
| *** zheng_li has joined #openstack | 17:34 | |
| rdw | gundlach: ah, I think perhaps the problem is that twistedr is importing getcurrent from eventlet, and it should be importing from eventlet.greenthread | 17:38 |
| gundlach | rdw: hm, ok. in any case, i filed a bug at eventlet.com, and have moved our code around so eventlet and twisted don't have to touch -- so the tests pass now and my branch merged (finally :D ) | 17:39 |
| *** tobym has quit IRC | 17:39 | |
| rdw | wewt | 17:39 |
| gundlach | indeed | 17:39 |
| rdw | so,wait, you're not even using the twisted stuff? | 17:39 |
| rdw | I didn't realize it got initialized unless you explicitly enabled it | 17:40 |
| creiht | gundlach: In my tests above, I had imported twisted reactor before importing anything from eventlet | 17:40 |
| creiht | rdw: If it detects twisted as imported, it will select it by default | 17:40 |
| rdw | :( | 17:41 |
| rdw | ok yeah so two things then I guess | 17:41 |
| gundlach | creiht: really! weird; if i tried importing twisted.internet.reactor and then "from eventlet import hubs" it barfed | 17:42 |
| gundlach | (in 0.9.10 of course) | 17:42 |
| rdw | oh yep, it sure does | 17:42 |
| creiht | gundlach: interesting as I did a import eventlet.hubs :) | 17:42 |
| rdw | that is some fucked up code | 17:42 |
| gundlach | creiht: shouldn't make a difference, should it? they're equivalent except for what name ends up in yoru namespace | 17:43 |
| creiht | I dunno | 17:43 |
| gundlach | http://paste.openstack.org/show/15/ | 17:43 |
| creiht | Just noting the difference | 17:43 |
| creiht | oh | 17:44 |
| creiht | heh | 17:44 |
| creiht | Worked for me due to another sideffect :/ | 17:44 |
| creiht | It was in the same session that it had errored on a previously on import eventlet | 17:45 |
| creiht | if you try to import again, it doesn't give you the error | 17:45 |
| creiht | sorry | 17:45 |
| rdw | ok well I'm disabling the automatic twisted-hub usage | 17:46 |
| gundlach | heh! | 17:46 |
| creiht | rdw: :) | 17:46 |
| gundlach | rdw: how? | 17:46 |
| creiht | gundlach: rdw is the eventlet maintainer | 17:46 |
| gundlach | oh, hey, you're ryan williams | 17:46 |
| creiht | :) | 17:46 |
| * gundlach asks for autograph | 17:46 | |
| * rdw writes "Ryan Williams" illegibly | 17:47 | |
| * gundlach rebuttons his shirt | 17:47 | |
| gundlach | thinks! | 17:47 |
| gundlach | erm, thanks! | 17:47 |
| eday | gundlach: what are you hacking on now? don't want to overlap with your work if I start hacking API stuff too | 17:47 |
| rdw | thanks for finding this weirdness! | 17:47 |
| gundlach | eday: images.py. just merged rsapi into trunk | 17:47 |
| gundlach | rdw: no prob | 17:47 |
| gundlach | and thanks for fixing it :) | 17:48 |
| eday | gundlach: ok, cool | 17:50 |
| gundlach | eday: lemme know if you have any more specific questions | 17:50 |
| *** kashyapc has quit IRC | 17:51 | |
| eday | gundlach: will do. I've been getting a real nova+kvm setup going on a machine so I can test things end-to-end... I think it's all there now :) | 17:51 |
| gundlach | eday: cool, put your instructions on the wiki somewhere so i can follow them :) | 17:51 |
| *** comstud has joined #openstack | 17:52 | |
| eday | gundlach: hah, well there are 4 sets of instructions on the wiki already... I'm thinking we need to consolidate (I didn't follow any of them) | 17:52 |
| gundlach | eday: yesyesyes | 17:52 |
| gundlach | eday: or remove all links to them, put yours up as the official one, and at the bottom in a footnote link to the old instructions | 17:53 |
| gundlach | until a few people have followed yours and been successful | 17:53 |
| eday | I need to polish up the method I'm using too | 17:54 |
| *** stewart_ is now known as stewart | 18:00 | |
| *** mray has quit IRC | 18:03 | |
| eday | zul: I am just running firectly out of bzr source so I can hack/restart easily. Just got libvirtd installed and setup network bridges, and am using boto to control it | 18:07 |
| eday | err, directly | 18:08 |
| *** tobym has joined #openstack | 18:11 | |
| *** holoway has quit IRC | 18:15 | |
| *** holoway has joined #openstack | 18:17 | |
| *** avsm has quit IRC | 18:19 | |
| *** brd_from_italy has joined #openstack | 18:20 | |
| *** sirp1 has quit IRC | 18:21 | |
| *** aliguori has joined #openstack | 18:22 | |
| *** sirp1 has joined #openstack | 18:23 | |
| *** sirp1 has quit IRC | 18:26 | |
| *** tobym has quit IRC | 18:30 | |
| *** tobym has joined #openstack | 18:31 | |
| *** al-maisan is now known as almaisan-away | 18:39 | |
| comstud | i followed the 'install from source' to run directly from a bzr checkout | 18:45 |
| comstud | (install from source wiki page) | 18:46 |
| soren | zul: You expressed an interest in the user-mode-linux stuff. Are you still running lucid anywhere? | 18:46 |
| zul | smoser: nope | 18:46 |
| zul | doh...i mean soren | 18:47 |
| soren | zul: alright, never mind. | 18:47 |
| zaitcev | Has anyone succeeded in running a smaller Swift cluster than the 4x that SAIO provides? Say, if I have an old laptop and I want to run an instance to test clients... Is it doable? | 18:50 |
| notmyname | zaitcev: you need at least 3 nodes (even if they are all on th same box, as in the SAIO) to get swift to run | 18:51 |
| zaitcev | notmyname: Thanks. | 18:51 |
| *** rajijoom has joined #openstack | 18:51 | |
| notmyname | zaitcev: the 4th node in the SAIO allows for handoffs in failure scenarios | 18:51 |
| *** thobbs has left #openstack | 18:53 | |
| jero | is there documentation about how high availability is handled in OpenStack ? | 18:57 |
| *** ptremblett has quit IRC | 18:58 | |
| notmyname | jero: compute or storage? | 18:58 |
| jero | notmyname: for all the critical components | 18:59 |
| jero | i'll take any link you have :-) | 18:59 |
| notmyname | I can only speak for swift (the storage piece). http://swift.openstack.org/overview_architecture.html is a good place to start for knowing how swift works | 19:00 |
| notmyname | http://swift.openstack.org/ is the swift doc site | 19:00 |
| creiht | jero: high available is quite a broad term, is there anything specifically you were wondering or about, or just generally? | 19:00 |
| jero | creiht: what I'm wondering, is if openstack is desingned from the bottom up to be highly available | 19:01 |
| creiht | yes | 19:02 |
| creiht | next question :) | 19:02 |
| notmyname | hehe | 19:02 |
| jero | and horizontally scalable ? | 19:02 |
| creiht | yes | 19:02 |
| creiht | :) | 19:02 |
| creiht | at least I can speak for swift, but I'm pretty sure those are the goals of nova as well | 19:02 |
| notmyname | jero: on the storage side, we have often said that a storage system is easy. what we have made is a storage system that handles failures | 19:03 |
| jero | ok | 19:03 |
| creiht | jero: A little more specifically, swift stores 3 copys of every object, and ensures on a write at least 2 copies have been written to return a success | 19:03 |
| jero | does swift also store the persistent disk volumes ? or just system images/stuff like that | 19:03 |
| notmyname | however, understand that error handling in the code is only part of the full picture. the other part is how your deployment is built out and how good your ops guys are :-) | 19:04 |
| creiht | jero: swift is "just" an object store, similar in scope to S3 | 19:04 |
| jero | okay | 19:04 |
| *** rsampaio has quit IRC | 19:04 | |
| jero | then how are the persistant disk volumes managed ? | 19:05 |
| * creiht passes off to the nova guys | 19:05 | |
| jero | :) | 19:05 |
| jero | ok I found a documentation pointer | 19:07 |
| creiht | nova.openstack.org and wiki.openstack.org are good starting places | 19:07 |
| jero | yeah | 19:07 |
| jero | anyone here on the choice of AoE for storage ? | 19:08 |
| soren | mtaylor: So... packaging. | 19:09 |
| mtaylor | soren: yes. | 19:10 |
| mtaylor | soren: so - I set up something for the swift guys that I wound up liking - namely that hudson merges trunk into the packaging branch on every trunk push so that there is always a branch that one can pull and build packages from | 19:11 |
| mtaylor | soren: I was thinking that if we did that for nova as a sort of trunk-packaging branch, and then had the debian and ubuntu branches be forks of that - we can roll up all of the extra changelog entries when it's time to release to deb or ubuntu | 19:12 |
| mtaylor | but I wanted to run it by you first before just doing it | 19:13 |
| redbo | age of empires for storage? | 19:16 |
| _0x44 | redbo: ATA over Ethernet? | 19:16 |
| jero | yes | 19:17 |
| redbo | OH! | 19:17 |
| _0x44 | redbo: But I like the way you're thinking. | 19:17 |
| creiht | :) | 19:17 |
| _0x44 | With four nodes, we could probably get swift to play a pretty good game of Risk. ;) | 19:17 |
| jero | this should be implemented as an easter egg. | 19:18 |
| creiht | It would be easy with wsgi middleware :) | 19:18 |
| redbo | it's like farmville, but with attack mode | 19:18 |
| *** rajijoom has quit IRC | 19:19 | |
| jero | so AoE is the current stuff, iSCSI is on the way, | 19:21 |
| creiht | jero: From what I hear that seems correct | 19:21 |
| creiht | mtaylor: can you take a look at: http://etherpad.openstack.org/FPJmxV34yP | 19:21 |
| creiht | and see if that seems reasonable | 19:21 |
| mtaylor | creiht: looking | 19:22 |
| soren | mtaylor: Sure, that sounds like fun. | 19:22 |
| mtaylor | soren: cool. I'll put it on my todo list | 19:22 |
| eday | vishy: you around? how are things with the sql work going? | 19:23 |
| mtaylor | creiht: the ~/.bazaar/bazaar.conf part can be replaced just with the bzr whoami command | 19:23 |
| soren | termie: You've been on top of the gflags business before. Would you mind taking a peek at https://code.edge.launchpad.net/~soren/nova/derootification/+merge/32888? I need it to move forward on a fresh upload to Ubuntu. | 19:24 |
| zaitcev | swift/README says "To build documentation run `python setup.py build_sphinx`", but this ends in "error: invalid command 'build_sphinx'". Any ideas? | 19:26 |
| soren | zaitcev: Do you have sphinx installed? | 19:26 |
| creiht | zaitcev: Yeah you have to have sphinx installed | 19:26 |
| *** rdw has quit IRC | 19:28 | |
| zaitcev | Aww, how silly of me. But of course. | 19:30 |
| creiht | zaitcev: and btw, docs are also available online at swift.openstack.org | 19:30 |
| *** Steve0 has joined #openstack | 19:37 | |
| *** Robi_ has joined #openstack | 19:39 | |
| *** theron has joined #openstack | 19:40 | |
| * Robi_ waves at theron | 19:40 | |
| theron | me waves from the brownbag podcast | 19:41 |
| Robi_ | nod, good questions | 19:41 |
| theron | Hi Robi_ | 19:41 |
| *** sirp1 has joined #openstack | 19:41 | |
| theron | already lovin' swift, the idea of having nodes strung across multiple providers with openvpn tying nodes together = awesome. | 19:42 |
| Robi_ | nice.. any migrations possible yet? | 19:42 |
| Robi_ | as in workload | 19:45 |
| Robi_ | vmotion type | 19:46 |
| creiht | What brownbag podcast is this? | 19:47 |
| theron | Robi_ still evolving, I'm pining the VMs to hosts... scaling out, really just leveraging vmware for templates, quick deployments, the idea of migrations or failover isn't required obviously. | 19:47 |
| creiht | theron: and thanks about the comment with swift | 19:47 |
| theron | creiht: someone needs to say "welcome to the community, just make sure to RTFM" | 19:47 |
| creiht | theron: A small note about the openvpn thing, while that will probably work, I'm not sure that would give the desired performance for large data sets | 19:48 |
| theron | creiht: certainly not. | 19:48 |
| theron | creight: thinking more along the lines of "private cloud storage" files email blah blah in public cloud. | 19:49 |
| creiht | ahh | 19:49 |
| gholt | Currently, swift has no concept of 'closest node'. But that'd be a cool feature if someone ever gets to it. :) | 19:49 |
| theron | gholt as long as I can tweak dns, closest node can be done externally, but yea, that would make life easier :) | 19:50 |
| Robi_ | heh, i had to do some dns hacking recently migrating a live app | 19:51 |
| *** sirp1 has quit IRC | 19:51 | |
| Robi_ | hosts file manipulation ftw | 19:51 |
| creiht | theron: You might be interested in http://www.nasuni.com/ | 19:51 |
| creiht | They can't put files accross different providers yet, but it sounded like something they were interested in | 19:52 |
| Robi_ | seems you get a choice of your provider though | 19:52 |
| gholt | theron: Well, I mean on the backend. All the backend process talk to each other with IPs (DNS would lag). For instance, only the proxy servers need to be publicly accessible. When a proxy is asked for an object it can retrieve that from any one of the backend replica nodes. If it knew which was closet to it.... | 19:52 |
| *** ptremblett has joined #openstack | 19:53 | |
| Robi_ | hmm this sounds a lot like the distributed storage model LaCie has been working on for some time with Wuala | 19:53 |
| theron | creiht nasuni looks interesting. | 19:53 |
| Robi_ | it's a dropbox like free service, but has a very neat distribution model | 19:54 |
| Robi_ | i remember talks at google years ago from the wuala team | 19:54 |
| Robi_ | invite code if wanted: http://is.gd/em5gk | 19:55 |
| creiht | I saw a presentation from Nasuni just a couple weeks ago and was impressed | 19:55 |
| theron | creiht a swift backed node where management is retained would be more palitable for certain customers, at least making an easy bridge to a public service down the road. :) | 19:56 |
| uvirtbot | New bug: #607235 in swift "Add cloudfiles developer docs to the swift docs" [Undecided,New] https://launchpad.net/bugs/607235 | 19:56 |
| theron | gholt ah. understood. | 19:56 |
| creiht | theron: indeed | 19:57 |
| gholt | uvirtbot: That's only a new bug to you. :P | 19:57 |
| uvirtbot | gholt: Error: "That's" is not a valid command. | 19:57 |
| creiht | hah | 19:58 |
| *** sirp1 has joined #openstack | 20:01 | |
| *** Xenith has joined #openstack | 20:03 | |
| *** theron_ has joined #openstack | 20:03 | |
| *** theron has quit IRC | 20:05 | |
| eday | _cerberus_: hey! hows API work going? Have you had much time for it? | 20:07 |
| *** ppetraki has joined #openstack | 20:07 | |
| *** theron__ has joined #openstack | 20:08 | |
| _cerberus_ | eday: slowly :-) I just picked it up today for the first time in a few days. | 20:09 |
| *** theron__ has quit IRC | 20:09 | |
| *** theron_ has quit IRC | 20:11 | |
| eday | _cerberus_: ahh | 20:14 |
| *** ThisDB has quit IRC | 20:14 | |
| *** gustavomzw has quit IRC | 20:15 | |
| _cerberus_ | eday: I plan to look at testing in a little bit. Looking through all the stuff that gundlach merged in over the last couple days first to see if I can get a clear picture in my head | 20:15 |
| gundlach | _cerberus_: feel free to ask me any q's you got | 20:15 |
| _cerberus_ | gundlach: I never hesitate :-) | 20:15 |
| *** gustavomzw has joined #openstack | 20:16 | |
| *** littleidea has quit IRC | 20:26 | |
| *** rdw has joined #openstack | 20:28 | |
| *** gustavomzw has quit IRC | 20:40 | |
| *** e1mer has quit IRC | 20:40 | |
| *** brd_from_italy has quit IRC | 20:44 | |
| *** jbd has joined #openstack | 20:44 | |
| *** ppetraki has quit IRC | 20:46 | |
| *** sophiap has quit IRC | 20:47 | |
| *** sophiap has joined #openstack | 20:48 | |
| *** ctennis has quit IRC | 20:51 | |
| *** ctennis has joined #openstack | 21:03 | |
| *** ctennis has joined #openstack | 21:03 | |
| gundlach | out, bye | 21:06 |
| *** gundlach has quit IRC | 21:07 | |
| termie | soren: done | 21:09 |
| *** e1mer has joined #openstack | 21:09 | |
| termie | soren: just style fixes, otherwise looks great | 21:09 |
| cw | creiht: poke | 21:18 |
| *** allsystemsarego has quit IRC | 21:19 | |
| *** jdarcy has quit IRC | 21:20 | |
| gholt | He's out for a little while, probably on again in a few hours. | 21:29 |
| soren | termie: Lovely, thanks. Will fix it up tomorrow. thanks for the review. | 21:37 |
| * soren heads bedwards | 21:37 | |
| termie | soren: gnight | 21:38 |
| comstud | hm | 21:38 |
| *** aliguori has quit IRC | 21:48 | |
| *** justinsheehy has quit IRC | 21:48 | |
| *** zaitcev has quit IRC | 21:50 | |
| *** ppetraki has joined #openstack | 21:50 | |
| *** avsm has joined #openstack | 21:57 | |
| creiht | cw: howdy | 22:04 |
| creiht | mtaylor: https://launchpad.net/~swift-core/+archive/ppa | 22:11 |
| creiht | swift ppa fail | 22:11 |
| mtaylor | creiht: oh wow. yeah. that's pretty mcuh fail | 22:12 |
| *** aliguori has joined #openstack | 22:15 | |
| *** zaitcev has joined #openstack | 22:20 | |
| cw | creiht: wrt to hashing a path to get a ring location ... surely i can game that? | 22:26 |
| cw | ie. generate paths to always git the same spot in the ring and flood that with large objects | 22:26 |
| creiht | cw: you can surely try :) | 22:26 |
| cw | since there is no salt in the process | 22:26 |
| creiht | There is indeed a salt | 22:26 |
| *** sophiap has quit IRC | 22:26 | |
| cw | ok good | 22:26 |
| cw | i must have missed that | 22:27 |
| cw | where is that? | 22:27 |
| * creiht looks | 22:27 | |
| creiht | At least kind of a salt | 22:27 |
| cw | salt makes things tasty | 22:27 |
| creiht | A configurable string of chars is added to the path | 22:27 |
| cw | that suffices | 22:27 |
| creiht | We worked through it abit, and even without the salt it would be very difficult, be we added it just to make everyone happy :) | 22:28 |
| cw | w/o it my math said it would be trivial ... 100ms or so effort for a 24-bit ring | 22:28 |
| cw | md5 on modern cpus is fast ... and everyone has lotsa cores now | 22:28 |
| creiht | cw: plus we figured that we could catch onto it pretty fast, and stop the user | 22:29 |
| cw | upload speed will limit you | 22:29 |
| creiht | indeed | 22:29 |
| creiht | and we would notice a node being off balance before it became an issue | 22:29 |
| cw | creiht: so you just use stochastic spreading to avoid filling up partitions? | 22:29 |
| creiht | cw: swift/common/utils.py hash_path() | 22:30 |
| creiht | not the HASH_PATH_SUFFIX | 22:30 |
| creiht | /not/note | 22:30 |
| cw | ok, awesome | 22:31 |
| cw | so when adding new physical space ... you rebalance the ring ... this means most things end up with 2 copies? | 22:31 |
| cw | and you let it replicate out? | 22:31 |
| creiht | cw: So we recommend that you configure the number of partitions so that there are at least 100 partitions per disk (estimating the max number of disks that you would use), which gives on average a 1% variance | 22:32 |
| cw | i saw that .... 100 per disk seemed to stem from some old rsync nasties | 22:32 |
| cw | i wondered if it might be relaxed | 22:32 |
| redbo | it's to keep things balanced with a larger sample | 22:33 |
| creiht | It can be relaxed if you don't mind more variance | 22:33 |
| creiht | (and less even distribution) | 22:33 |
| cw | the rign sized is fixed presently surely? | 22:33 |
| cw | so as you add more hw the # partitions per spindle will decrease | 22:34 |
| creiht | When adding new physical space, the ring rebalances moving at most 1 replica of a partition, and everything then replicates out | 22:34 |
| creiht | cw: correct | 22:34 |
| cw | (btw, partitions is an over-used and much confusing mataphore) | 22:34 |
| cw | right, so you have to plan for some target | 22:34 |
| creiht | Once you set the partition count, you can't change it | 22:34 |
| creiht | correct | 22:34 |
| cw | and if you're at 10% of that to start with ... you have 1000 / spindle | 22:34 |
| creiht | correct | 22:34 |
| * cw waves hands with vague numbers | 22:34 | |
| creiht | hehe | 22:34 |
| cw | do you have any idea on the access patters for the data? | 22:35 |
| cw | this is tangential | 22:35 |
| cw | ie. what objects are accessed hot-cache? | 22:35 |
| cw | latency would tell you | 22:35 |
| cw | this goes back to earlier talks about limited RAM and slab vs page-cache value | 22:35 |
| creiht | cw: In our use case, I don't think we get many cache hits, as that type of traffic is pushed to the CDN | 22:35 |
| cw | is that true for metadata too? | 22:36 |
| creiht | I don't know that for sure, but my guess would be that it is true as well | 22:36 |
| cw | again, latency would tell you | 22:37 |
| redbo | Right now we don't use the page cache for file data, because it's not good for our use case. I expect other people will wanna change that eventually. | 22:37 |
| cw | stat from cache is low usecs (faster maybe) ... hitting the disk is 5ms+ | 22:37 |
| *** justinsheehy has joined #openstack | 22:37 | |
| cw | redbo: you use regular IO | 22:37 |
| cw | so it does use the page-cache | 22:37 |
| cw | which tends to push the slab down | 22:37 |
| redbo | okay, we use the page cache, then throw the page cache away | 22:37 |
| creiht | hehe | 22:38 |
| cw | how do you throw it away? | 22:38 |
| redbo | posix_fadvise | 22:38 |
| cw | there is no syscall to purge it unless you unlink the inode | 22:38 |
| redbo | sure there is | 22:38 |
| cw | that's advise :-) | 22:38 |
| cw | i'm not 100% sure that works that well ... at least it never did in the past | 22:38 |
| cw | cat /proc/meminfo ... and see how much is in the page-cache | 22:38 |
| redbo | if you advise it that you don't need the page cache, it throws it away | 22:38 |
| redbo | it'll be full of inodes, I hope | 22:39 |
| redbo | and directories | 22:39 |
| cw | ok, you win :-) | 22:40 |
| * cw owes more beer | 22:40 | |
| creiht | hah | 22:40 |
| redbo | I'm thinking we might remove that fadvise call, because it's bad for other peoples use cases. And use the vfs cache pressure settings on our install so it'll tend to cache inodes but not file data on its own. | 22:40 |
| cw | there is code that will invalidate on POSIX_FADV_DONTNEED | 22:40 |
| *** jc_smith has joined #openstack | 22:40 | |
| creiht | or at least make it configurable | 22:41 |
| cw | vfs cache pressure from page cache means your inodes will get cached very poorly | 22:41 |
| cw | i did tweaks in the past to try avoid this | 22:41 |
| cw | on other stuff i use DIO | 22:41 |
| cw | and avoid the page cache completely | 22:41 |
| cw | s/on/in/ | 22:41 |
| redbo | I think it depends on which way you tune it? | 22:42 |
| cw | dio or page-cache vs inodes? | 22:42 |
| cw | the latter is does a tiny bit | 22:42 |
| cw | but not as much as you think ... there is a vfs pressure knob from 100 to 0 ... but it doesn't work very well | 22:43 |
| redbo | "Decreasing vfs_cache_pressure causes the kernel to prefer to retain dentry and inode caches. Increasing vfs_cache_pressure beyond 100 causes the kernel to prefer to reclaim dentries and inodes." | 22:43 |
| cw | _prefer_ | 22:43 |
| cw | it's not linear | 22:43 |
| cw | non-zero still means page-cache puts a lot of pressure on slab | 22:43 |
| cw | it's basically a count of loops into the reclaim code in some cases | 22:43 |
| redbo | I'd still like to test it. | 22:44 |
| redbo | but yeah, maybe leaving it as an option is better | 22:44 |
| redbo | there will probably be other people who don't have repeated reads to a single file very often | 22:45 |
| redbo | and that might be changing for us soon anyway, if we do public access containers | 22:45 |
| cw | redbo: ok, so i am showing my age ... i checked, the 'fadvise problem' was fixed april 2004 :) | 22:46 |
| redbo | haha | 22:46 |
| creiht | hah | 22:46 |
| redbo | I know it makes re-reading files slower, so I assume it's dumping the cache. | 22:46 |
| eday | comstud: any word on your scheduler branch? I know jesse was hacking on it, but probably good to get that merged sooner than later :) | 22:48 |
| cw | redbo: it invalidates the pages yes | 22:48 |
| *** rnewson has quit IRC | 22:49 | |
| comstud | eday, had to work on these couple of tickets | 22:49 |
| cw | i have code that uses mincore to measure how much of a file is cached, i'll try it out later and verify it does as it claims | 22:49 |
| comstud | jesse got caught up with other things, so I'm doing some additional work | 22:49 |
| comstud | i'll try to propose a merge tomorrow | 22:49 |
| eday | cool | 22:49 |
| *** justinsheehy has quit IRC | 22:49 | |
| *** johnpur has quit IRC | 22:50 | |
| *** tobym has quit IRC | 22:55 | |
| *** justinsheehy has joined #openstack | 22:56 | |
| cw | redbo: removing the fadvise would be ugly actually | 22:56 |
| cw | redbo: w/o that things will initially be _really_ fast | 22:57 |
| cw | once your working set gets larger though ... you'll be like 10x to 100x slower | 22:57 |
| cw | and people will pelt you with angry kittens | 22:57 |
| *** avsm has quit IRC | 23:05 | |
| *** sophiap has joined #openstack | 23:06 | |
| redbo | I expect our working set to be a tiny percentage of our total storage. But if we have people serving images directly from cloudfiles, it could get important. | 23:07 |
| *** jamiew is now known as jamiew_moviez | 23:07 | |
| *** pcrews has left #openstack | 23:08 | |
| redbo | why can't we just have 128gb of ram on the storage machines like I wanted? | 23:09 |
| *** ppetraki has quit IRC | 23:09 | |
| cw | 16gb dimms are like unicorm hair | 23:09 |
| cw | right now | 23:09 |
| redbo | They also didn't like my networking idea. "Everything needs to talk to everything else really fast." | 23:09 |
| *** pvo has left #openstack | 23:10 | |
| _0x44 | redbo: TCP over New Jersey nannies isn't as fast as you'd think. | 23:10 |
| cw | high bi-section bandwidth is getting affordable now | 23:10 |
| creiht | hah | 23:11 |
| *** anm_ has quit IRC | 23:11 | |
| *** justinsb has joined #openstack | 23:12 | |
| *** rdw has joined #openstack | 23:13 | |
| *** miclorb_ has joined #openstack | 23:13 | |
| cw | redbo: it seems you should be able to increase the ring size actually | 23:14 |
| cw | double it on demand | 23:14 |
| *** aliguori has quit IRC | 23:16 | |
| *** sophiap has quit IRC | 23:30 | |
| *** sandywalsh has quit IRC | 23:33 | |
| *** sophiap has joined #openstack | 23:39 | |
| *** jero- has joined #openstack | 23:45 | |
| *** tr3buchet has quit IRC | 23:52 | |
| *** silassewell has joined #openstack | 23:54 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!