| _mup_ | txzookeeper/session-and-conn-fail r53 committed by kapil.foss@gmail.com | 00:04 |
|---|---|---|
| _mup_ | set with transparent retry (using retry_change), correct doc string to note session timeout value is in milliseconds | 00:04 |
| _mup_ | Bug #879731 was filed: Better error handling for session failure and connection loss <juju:New for hazmat> <txzookeeper:New for hazmat> < https://launchpad.net/bugs/879731 > | 00:17 |
| _mup_ | txzookeeper/session-and-conn-fail r54 committed by kapil.foss@gmail.com | 00:25 |
| _mup_ | improve retry delay | 00:25 |
| _mup_ | txzookeeper/session-and-conn-fail r55 committed by kapil.foss@gmail.com | 03:37 |
| _mup_ | simplify retry, add retry delay to watch op retries | 03:37 |
| _mup_ | txzookeeper/session-and-conn-fail r56 committed by kapil.foss@gmail.com | 04:09 |
| _mup_ | test for core retry & retry watch in isolation | 04:09 |
| _mup_ | txzookeeper/session-and-conn-fail r57 committed by kapil.foss@gmail.com | 04:11 |
| _mup_ | version increment, and pep8 | 04:11 |
| _mup_ | txzookeeper/session-and-conn-fail r58 committed by kapil.foss@gmail.com | 04:24 |
| _mup_ | update license headers, add author tag | 04:24 |
| tauren | just discovered juju, looks cool, but have a question | 11:15 |
| tauren | can I only deploy on AWS? | 11:15 |
| tauren | is there a way to use my own servers and virtualization? | 11:16 |
| tauren | nm, found the answer in the FAQ. https://juju.ubuntu.com/docs/faq.html | 11:18 |
| hazmat | tauren, thats outdated | 16:48 |
| hazmat | tauren, you can deploy on openstack or directly onto physical machines, or deploy entire environments on a single machine with lxc | 16:49 |
| tauren | hazmat: sweet! is there a howto somewhere? | 16:49 |
| hazmat | tauren, which would would you like to do? | 16:50 |
| hazmat | tauren, we're having some problems updating our docs atm | 16:50 |
| tauren | probably lxc | 16:50 |
| hazmat | tauren, i've put up some updated ones at http://charms.kapilt.com/docs/provider-configuration-local.html | 16:50 |
| tauren | but when you say directly onto physical machines, is there something that will automatically set them up for virtualization? | 16:50 |
| hazmat | tauren, we deploy openstack with juju onto physical machines | 16:51 |
| hazmat | tauren, juju uses cobbler/orchestra as a machine provider for that scenario | 16:51 |
| tauren | hmmm... here's what I want to do... | 16:51 |
| tauren | I have a single physical server | 16:52 |
| tauren | and I'd like to configure it to be able to run multiple services. | 16:52 |
| tauren | currently it has kvm installed | 16:52 |
| tauren | but i plan to start over. | 16:52 |
| tauren | what solution would you suggest? | 16:52 |
| hazmat | tauren, local provider which uses lxc for isolation | 16:53 |
| hazmat | its lightweight virtualization | 16:53 |
| tauren | i'm running openvz on another server. i assume it is similar? | 16:53 |
| hazmat | well its more properly not virtualization.. more like a solaris zone, freebsd jail, or chroot on steroids. | 16:53 |
| hazmat | tauren, very similiar to openvz | 16:53 |
| tauren | ok | 16:53 |
| hazmat | tauren, except in kernel | 16:53 |
| tauren | nice | 16:53 |
| tauren | so i could deploy charms for wordpress, etc. onto it then. | 16:55 |
| tauren | what if i also need to run some java services? my understanding is kvm is preferred over openvz for that. same for lxc? | 16:55 |
| tauren | hazmat: also, can I scale only locally, or can I expand out to EC2 if my local resources are exhausted? | 16:58 |
| tauren | hazmat: i need to head out. thanks for your help! | 17:03 |
| hazmat | tauren, you can only use one provider for a single environment atm | 17:03 |
| hazmat | tauren, but you can have multiple environments managed by juju | 17:04 |
| hazmat | ie. test local, deploy ec2 | 17:04 |
| tauren | so I couldn't have a wordpress blog hosted locally and add EC2 resources to scale it? | 17:04 |
| hazmat | tauren, cheers | 17:04 |
| hazmat | tauren, no, we don't have public/private bursting, and probably won't for a while | 17:04 |
| tauren | good to know. | 17:05 |
| tauren | is openstack something I run on my servers? | 17:05 |
| hazmat | tauren, yes | 17:05 |
| tauren | or is that an API that rackspace offers? | 17:05 |
| tauren | does openstack use lxc, or something else? | 17:05 |
| hazmat | tauren, openstack is a cloud provider with its own api and a minimal ec2 api, it supports multiple virtualization backends | 17:06 |
| hazmat | from kvm, xen, lxc, etc ... | 17:06 |
| tauren | alright... so the link you sent doesn't show how to get openstack going, just plain local with lxc, right? | 17:06 |
| hazmat | tauren, openstack usage is the same as ec2 | 17:25 |
| hazmat | tauren, google has some answers for juju + openstack + orchestra | 17:25 |
| hazmat | https://wiki.ubuntu.com/ServerTeam/UbuntuCloudOrchestraJuju | 17:25 |
| _mup_ | Bug #880023 was filed: machine agent disconnects from zoopkeeper on heavy loads <juju:New> < https://launchpad.net/bugs/880023 > | 17:43 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!