=== kiall|AFK is now known as kiall | ||
=== kiall is now known as kiall|AFK | ||
=== kiall|AFK is now known as kiall | ||
=== kiall is now known as kiall|AFK | ||
=== kiall|AFK is now known as kiall | ||
=== kiall is now known as kiall|AFK | ||
=== kiall|AFK is now known as kiall | ||
=== kiall is now known as kiall|AFK | ||
areay | hi all.. i'm using UEC -- is it possible to have more than one VM per cpu core? i understand this is the default but is there a way to override it? | 08:06 |
---|---|---|
kim0 | areay: Hi .. are you saying if your NC has 2 cores .. you can only launch two VMs ? | 08:08 |
TeTeT | areay, kim0 : there's a MAX_CORE or MAX_CORES setting for the nc that overrides the default | 08:09 |
areay | kim0 -- i haven't actually gotten that far yet lol.. but the ubuntu server guide says that "by default, Eucalyptus will only run 1 VM per CPU core on a Node." | 08:10 |
areay | TeTeT, how do i configure that setting? | 08:10 |
TeTeT | areay: I have the following entry: | 08:11 |
TeTeT | /etc/eucalyptus/eucalyptus.conf:MAX_CORES="32" | 08:11 |
TeTeT | on the node controller. So this node can host up to 32 vm's, which is a bit high, but it's only for training and testing | 08:12 |
areay | thanks bro :) sorry i'm new at this -- that should have been the first place i looked | 08:12 |
areay | TeTeT, how do i reload the settings after i edit the file? i better ask or i'll end up rebooting like a noob lol | 08:13 |
TeTeT | areay: sudo restart eucalyptus-nc or sudo restart eucalyptus-nc CLEAN=1 | 08:13 |
areay | tyvm! | 08:13 |
TeTeT | areay: be aware that you loose any already running instances on that node | 08:13 |
areay | np i don't have any yet :) also my node is saying kvm disabled by bios -- is that as big of a problem as it sounds? | 08:14 |
kim0 | I think so .. visit bios and try enabling any virtulization thing | 08:15 |
areay | kk cool i'm on it :) thanks u guys are stars | 08:16 |
areay | ok, i've downloaded and installed an image... once i instantiate it, what happens if it, the NC, or the CC reboots? do i need to save snapshots? i'm kinda lost :/ | 08:53 |
=== kiall|AFK is now known as kiall | ||
=== kiall is now known as kiall|AFK | ||
areay | ok ignore above post -- my instance is running... but i cannot ping or connect via ssh. i instantiated using hybridfox, and when i choose "connect" it's asking me for a private key file? where do i find this key? | 09:20 |
=== kiall|AFK is now known as kiall | ||
areay | ok i'm getting further -- i get output from ssh now but i need that key :/ | 09:24 |
areay | <-- stupid. i didn't create a keypair before i instantiated... disregard above question ;) | 09:34 |
=== kiall is now known as kiall|AFK | ||
=== kiall|AFK is now known as kiall | ||
=== kiall is now known as Kiall | ||
TeTeT | areay: good luck with your discovery of uec :) | 11:25 |
areay | haha thanks tetet :) | 11:47 |
=== dendrobates is now known as dendro-afk | ||
superxgl | echo -at $logdir | 14:19 |
superxgl | oh,sorry | 14:20 |
kim0 | :) | 14:35 |
=== RoAkSoAx is now known as andreserl | ||
=== Kiall is now known as Kiall|AFK | ||
areay | Hi all... I have a couple questions about the basics of UEC/cloud computing in general -- firstly, i understand that instances are considered disposable - if i want specialized "appliances" (mysql, vpn, etc) do i need to create custom images in order to be able to swap them out on the fly with minimal configuration? | 18:01 |
areay | or do i create a "snapshot"? | 18:01 |
areay | also, i understand that sensitive data should be kept on a "volume" -- are these volumes held on the nc or on the clc/cc/sc/walrus box? | 18:03 |
areay | if anyone can answer even one of these questions i'd be extremely appreciative :) | 18:06 |
kim0 | areay: yes instances are disposable, yes important data should be on a volume .. which is stored on SC | 18:34 |
kim0 | areay: for appliances .. I would boot a generic image then customize it using something like cloud-init and/or puppet | 18:34 |
areay | kim0, so customize it as a running instance? or chroot into the image? --also once the image is done what's the best way to save it? | 18:36 |
kim0 | areay: It's not like that .. the "modify instance, then make an image out of it" | 18:36 |
kim0 | is a valid approach .. but not the solution I'd recommend .. instead | 18:37 |
kim0 | I'd boot generinc image, it picks up customization instructions from cloud-init and/or puppet, it becomes "mysql-appliance", then mounts ebs volume which your data, now it's your-appliance | 18:37 |
kim0 | no snapshotting needed | 18:38 |
kim0 | everything can be rebuilt in seconds | 18:38 |
areay | kim0, sounds good -- which is better, cloud-init or puppet? | 18:40 |
kim0 | areay: ehum .. consider them complementary :) | 18:41 |
kim0 | cloud-init is for initial bootstrapping .. you can stop there .. or you can continue with puppet | 18:41 |
kim0 | read more about both .. and see what you want to use | 18:41 |
areay | sorry for the noobish questions -- i'm VERY new to all this and the documentation available is sparse at best.. i was using multi-roled servers before.. is there like a guide or anything? a simple guide to setup a lamp appliance or something would be ideal so i can see all this in practice... | 18:42 |
kim0 | areay: check out my cloud-init stuff at http://foss-boss.blogspot.com/search/label/cloud-init | 18:43 |
areay | awesome :) tyvm kim0 | 18:43 |
kim0 | areay: read bottom to top :) has wordpress example (LAMP stack) and all | 18:43 |
areay | kim0, nice blog btw -- i take it that's yours :) | 18:44 |
kim0 | yep :) | 18:45 |
* kim0 afk for now | 18:45 | |
=== dendro-afk is now known as dendrobates |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!