| smoser | harmw, well, some news. seems busted in my 0.3.2~pre1 also. | 01:46 |
|---|---|---|
| smoser | hm.. and interesting. it seems busted in 0.3.1 also. the partition table does get grown. | 01:50 |
| === ctracey is now known as ctracey|away | ||
| === harlowja is now known as harlowja_away | ||
| harmw | smoser: I just filed https://bugs.launchpad.net/cloud-init/+bug/1261710 | 11:34 |
| harmw | I'll see if I can work on that tonight | 11:34 |
| harmw | annoying | 11:34 |
| harmw | hm, and why didn't I file that against cirros :| | 11:35 |
| harmw | there, changed it | 11:37 |
| smoser | harmw, thanks. i can't figure it out. i think something is killing the resize2fs process after it starts. | 13:39 |
| smoser | i changed 'resize2fs' to 'sleep 5' and it seems to die also. also tried putting the content of the 'sh -c' in its own program. that didnt work either. | 13:40 |
| smoser | the one thing that does work is removing the '&' | 13:40 |
| smoser | ie, not backgrounding. | 13:40 |
| smoser | i'll put this in the bug. | 13:40 |
| harmw | great, nice | 14:03 |
| harmw | (-1 for that silly typo btw) | 14:04 |
| smoser | harmw, it seems that using start-stop-daemon does work. but i couldn't get start-stop-daemon to sanely call a 'sh' -c program | 15:42 |
| smoser | so had to put it in a different program | 15:42 |
| harmw | ah great | 15:43 |
| smoser | harmw, want to take the fix for a spin ? | 16:44 |
| smoser | get cirros trunk (bzr branch lp:cirros) | 16:44 |
| smoser | then | 16:44 |
| smoser | tar -C src -cvf - sbin/resize-filesystem etc/init.d/resizefs lib/cirros/shlib | ssh cirros@$IP 'sudo tar -C / -xvf - && sudo reboot' | 16:45 |
| smoser | seems to work for me, and when its done you get a nice message like: | 16:46 |
| smoser | /dev/root resized successfully [took 43.12s] | 16:46 |
| === harlowja_away is now known as harlowja | ||
| harmw | smoser: ill give it a go in a few minutes | 19:07 |
| smoser | harmw, k. good. i dont really know how i feel about "fixing" it. given the 45 seconds it took to run. | 19:16 |
| smoser | i'm apt to put this in and then turn it off by default. | 19:16 |
| harmw | it kinda goes straight into the whole cirros philosophy | 19:22 |
| harmw | any thoughts on how to enable it if the default where to have it disabled? | 19:23 |
| harmw | === cirros: current=0.3.2~pre1 latest=0.3.1 uptime=100.09 === | 19:23 |
| harmw | +1 for a lightspeed cloud :> | 19:23 |
| harmw | === cirros: current=0.3.2~pre1 latest=0.3.1 uptime=63.64 === | 19:27 |
| harmw | and that was after the reboot + resizefs stuff | 19:27 |
| harmw | # grep resize e1ba8f11-097f-4bc1-bf9b-475710fee5ef/console.log | 19:28 |
| harmw | wheeljack login: /dev/root resized successfully [took 92.04s] | 19:28 |
| === harlowja is now known as harlowja_away | ||
| smoser | one thought on how to enable it if it were disabled is via file injection (through config drive or openstack api) | 19:41 |
| smoser | i think it would work. | 19:41 |
| smoser | if /etc/init.d/resizefs read /etc/default/resizefs.conf | 19:42 |
| harmw | ./sem/instance.i-00000052.userdata | 19:49 |
| harmw | ./sem/instance.i-00000052.cirros-apply-net | 19:49 |
| harmw | ./sem/once.once.resize-rootfs | 19:49 |
| harmw | ./sem/instance.i-00000052.check-version | 19:49 |
| harmw | isn't that missing the tiny touch of consistency? :) | 19:50 |
| harmw | can't it be specified in user-data? a believe there already is a module to allow for this to happen | 19:52 |
| harmw | smoser: perhaps have the resizefs script read /var/run/cirros/datasource/data/user-data to make the decision to run/start | 19:53 |
| smoser | consistency ? | 19:53 |
| smoser | well, 'instance' is that this is 'per-instance', and then the second portion is the instance id. and then the key. | 19:54 |
| smoser | for 'once', there is no useful second portion | 19:54 |
| smoser | so i just reused the sring 'once' | 19:54 |
| smoser | ie, the resize will only run once, ever. | 19:54 |
| smoser | it could be 'per-instance', and probably woudl work. but i really dont think people are "rebundling" cirros :) | 19:54 |
| smoser | we could have it specified as user-data, yes. but at the moment cirros doesn't support any sort of "cloud-config" user data. | 19:55 |
| smoser | it only supports '#!' | 19:55 |
| smoser | if i had some cloud-config analog then it'd be perfect. but thats not there yet. | 19:55 |
| smoser | the inject-files in the openstack api "should work" as it is right now i think. | 19:56 |
| smoser | with no other features needed from cirros except having the rc.d script source (if it exists) that file. | 19:56 |
| harmw | true | 19:56 |
| smoser | i agree that cloud-config like function is better long term goal | 19:56 |
| harmw | indeed | 19:56 |
| smoser | but i'd wan to have this enable-able in 0.3.2 and want that "real soon now" | 19:56 |
| harmw | though having it work based on the presence of a certain file is cool enough for now :) | 19:57 |
| smoser | well, i'd have it source the file | 19:57 |
| smoser | if present | 19:57 |
| smoser | ie, so the file could have: | 19:57 |
| smoser | enabled=0 | 19:57 |
| smoser | background=0 | 19:57 |
| smoser | or something to that affect. | 19:57 |
| smoser | or mabye just | 19:57 |
| harmw | isn't it easier to just check if a (0byte) file exists? | 19:58 |
| smoser | mode=(foreground|disabled|background) | 19:58 |
| smoser | existence doesn't allow me to do foreground and background | 19:58 |
| smoser | and i really do thikn the backgrounding is weird | 19:58 |
| harmw | right on that | 19:58 |
| smoser | cause for race conditions | 19:58 |
| smoser | harmw, feel free to give this a try | 20:10 |
| smoser | http://paste.ubuntu.com/6590740/ | 20:10 |
| smoser | bah. bad syntax. | 20:10 |
| harmw | darn, I was just busy on something myself :) | 20:10 |
| smoser | http://paste.ubuntu.com/6590744/ | 20:11 |
| smoser | harmw, sorry to get in your way, i'd love for you to particpate in cirros developement :) | 20:11 |
| harmw | and I should have openstack create the /etc/default/resizefs file | 20:11 |
| harmw | haha np :) | 20:11 |
| smoser | yeah, you should be able to do somethin glike: | 20:12 |
| smoser | nova boot --file /etc/default/resizefs=my.resizefs.txt | 20:13 |
| harmw | yea | 20:14 |
| === harlowja_away is now known as harlowja | ||
| harmw | hm, seems my cloud is not very fond of inserting files | 20:19 |
| harmw | like this: /data/openstack/nova//instances/b0471e49-094c-4eed-b8f0-7a979ed35957/disk: No such file or directory | 20:19 |
| harmw | NovaException: Error mounting /data/openstack/nova//instances/b0471e49-094c-4eed-b8f0-7a979ed35957/disk with libguestfs | 20:21 |
| smoser | harmw, i think you can turn that off. | 20:22 |
| smoser | we dont need the host to insert it | 20:22 |
| smoser | (i would never find that an acceptable solution) | 20:22 |
| smoser | there is a way where the "inserted files" just appear in the config drive | 20:23 |
| smoser | and then cirros reads them and *it* inserts them. | 20:23 |
| smoser | harlowja, do you know how to set that ? | 20:23 |
| harlowja | hmmm | 20:23 |
| * smoser goes to try to remember if cloud-init supports that in config drive. i thin it does | 20:23 | |
| harmw | well, i don't use config drive atm and enabling it causes the other vm's to fail on boot iirc | 20:23 |
| smoser | harmw, well, you dont need config drive. cirros should read it from the MD also | 20:24 |
| smoser | (the files can be inserted there too) | 20:24 |
| harmw | yea, i thought so | 20:24 |
| harmw | was expecting that as well | 20:24 |
| smoser | but clearly if you're trying to insert config for ifconfig eth0' then its probably not going to work :) | 20:24 |
| harmw | hehe | 20:24 |
| smoser | harlowja needs to write a proper cloud-init DataSourceOpenstack | 20:25 |
| harlowja | :) | 20:25 |
| smoser | that reads from the openstck MD. we dont really ahve that now. only could reaad those files from the config drive. | 20:25 |
| harlowja | one sec, let me see what the option is | 20:25 |
| smoser | harlowja, that really is something i'd like for 14.04 | 20:26 |
| harlowja | hmmmm | 20:27 |
| smoser | proper oepnstack metadata service datasource (rather than just riding on ec2) | 20:27 |
| harlowja | agreed | 20:29 |
| harmw | smoser: does cirros read metadata? since then we could use a metadata key+value instead of inserting a file | 20:37 |
| smoser | harmw, i think it does read it in some sense. | 20:38 |
| smoser | in that it would write keys to files in the datadir | 20:38 |
| smoser | so it could pretty easiliy use that | 20:38 |
| smoser | the thing i dont like about that is that i dont know that i want to support it going forward | 20:39 |
| smoser | i can trivially keep support for /etc/default/.... injection | 20:39 |
| harlowja | harlowja i think if u make sure all 'inject_partition=0', 'inject_key=0' nova.conf values are off that will be step 1 | 20:41 |
| harlowja | apparently this is on by default | 20:42 |
| harlowja | inject_key=true | 20:42 |
| harlowja | https://github.com/openstack/nova/blob/master/etc/nova/nova.conf.sample#L2510 | 20:42 |
| harlowja | but u are hitting this code | 20:43 |
| harlowja | https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L2570 | 20:43 |
| harlowja | "If we're not using config_drive, inject into root fs" | 20:43 |
| harlowja | :( | 20:43 |
| harlowja | harmw i think u can also use the following | 20:43 |
| harlowja | https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L2544 | 20:43 |
| harlowja | inject_files and CONF.libvirt.inject_partition != -2 | 20:43 |
| harlowja | so make inject_partition=-2 | 20:44 |
| harlowja | weirddd | 20:44 |
| harlowja | ha | 20:44 |
| harlowja | # The partition to inject to : -2 => disable, -1 => inspect | 20:44 |
| harlowja | # (libguestfs only), 0 => not partitioned, >0 => partition | 20:44 |
| harlowja | # number (integer value) | 20:44 |
| harlowja | #inject_partition=1 | 20:44 |
| harlowja | by default 1 | 20:44 |
| harlowja | crazy | 20:44 |
| harmw | hehe | 20:45 |
| smoser | some day the default will be -2 | 20:45 |
| harlowja | ya, i hope so | 20:46 |
| harmw | so change to -2 and then nova boot --file should work just fine | 20:46 |
| harlowja | harmw using config drive or nothing at all? | 20:47 |
| harmw | nope, just metadata | 20:47 |
| harmw | no configdrive here | 20:48 |
| harlowja | so sure, then if smoser is correct and cirros understands special openstack metadata then u should be good to go | 20:48 |
| harlowja | never know with that smoser guy, ha | 20:49 |
| harmw | time to find out :) | 20:49 |
| harlowja | :) | 20:50 |
| harlowja | but ya, i should (or should get someone from y!) to write a best-impl of an openstack metadata/config-drive thing | 20:50 |
| harlowja | *datasource thing | 20:50 |
| harlowja | as smoser suggested | 20:50 |
| smoser | uh oh. | 20:53 |
| smoser | it imight only do that from config-drive (not MD). hmm. | 20:53 |
| smoser | suck. i don tthink cirros has proper openstack metadata source either. | 20:53 |
| smoser | so harmw that isn't going to work :-( | 20:54 |
| harlowja | hmmm | 20:56 |
| harmw | indeed, it didn't inject the file | 20:58 |
| harmw | uhm, why doesn't it read the stuff from user-data again? | 20:59 |
| smoser | it should execute user-data | 21:01 |
| smoser | if it starts with '#!' | 21:01 |
| smoser | but thats its only consumption of user-data at the moment. | 21:01 |
| smoser | and that runs after resize would have run | 21:01 |
| smoser | but if you want, from that '#!' you could just *run* '/sbin/resize-filesystem /dev/root' | 21:01 |
| smoser | at the beginning. | 21:01 |
| harmw | nah, thats far to easy :) | 21:04 |
| harlowja | smoser a start at this, http://paste.openstack.org/show/55206/ | 21:18 |
| smoser | harmw, you think thats a acceptable work around for now ? | 21:19 |
| smoser | and then when we get either a true file insertion or "cirros-config" format in user-data we'll allow enabling it that way. | 21:19 |
| smoser | (in the mean time the user can enable it by actually modifying the image, or by config-drive still should work) | 21:20 |
| smoser | harlowja, woohoo | 21:20 |
| harmw | smoser: yes, probably good enough for now | 21:21 |
| harmw | though ultimately reading from user-data would be the best choice | 21:22 |
| smoser | yeah. i agree. | 21:23 |
| smoser | https://bugs.launchpad.net/cirros/+bug/1232239 | 21:23 |
| smoser | is a start at that. | 21:23 |
| smoser | ie, there is code there to read mime-multipart | 21:23 |
| smoser | and since i have a json reader, we can pretty easily support a "cloud-config-archive" like thing in a similar fashion | 21:23 |
| harmw | nice | 21:25 |
| harmw | how does this behave when I've put yaml in user-data? | 21:26 |
| smoser | probably barf. | 21:27 |
| smoser | i only have a json parser | 21:27 |
| smoser | and i dont plan on a proper yaml one. | 21:28 |
| smoser | its easy enough to convert yaml to json before passing to cirros | 21:28 |
| smoser | and hard (without C programming) in cirros to write a yaml parser. | 21:28 |
| harmw | you mean I should feed json user-data to nova on instance boot, or have cirros convert user-data to json upon loading/booting? | 21:29 |
| smoser | 2 things here (kind of unrelated) | 21:31 |
| smoser | a.) want to have cirros support "multi-part" user-data as input. | 21:31 |
| smoser | a.1) mime-multipart (as seen in that bug with the awk parser) | 21:32 |
| smoser | a.2) cloud-config-archive like multipart (cloud-config-archive is a cloud-init thing that basically just allows you to make an "archive" of parts in yaml rather than mime) | 21:32 |
| smoser | for a.2, cirros would not support that being yaml, but instead just json. but same basic format. | 21:33 |
| smoser | b.) some of those parts could be "cirros-config" which would be a json config format similar to "cloud-config" | 21:33 |
| smoser | with the same caveat of not supporting yaml, but only json | 21:33 |
| smoser | in the future if somehow i got a small enough yaml parser, we could just support yaml in both of those places. | 21:34 |
| harmw | ok | 21:34 |
| harmw | would've been easier to just put cloud-init in cirros :p | 21:35 |
| smoser | $ apt-cache show python2.7-minimal | grep -i size | 21:42 |
| smoser | Installed-Size: 3399 | 21:42 |
| smoser | Size: 1187530 | 21:42 |
| harmw | ye yea, bit to large | 21:43 |
| smoser | so 3.XMB for python-minimal. | 21:43 |
| smoser | i'm sure you could get stuff smaller | 21:43 |
| smoser | but all of cirros is ~ 8M (the disk images are bigger because of filesystem overhead and the fact that the disk1.img file actually has *2* copies of cirros in it) | 21:44 |
| smoser | oh. wait. no it doesnt. | 21:45 |
| harlowja | smoser i'll try to work on improving that | 21:52 |
| smoser | python ? | 21:53 |
| harlowja | *openstackdatasource, ha | 21:53 |
| harlowja | not python, ha | 21:53 |
| harlowja | smoser u fix python, lol | 21:53 |
| smoser | i have to admit, i kind of like programming in awk and busybox sh | 21:55 |
| smoser | having python in cirros would just be no fun. | 21:55 |
| harmw | hehe | 21:55 |
| harmw | smoser: is multipart realy required straight away? or would a simpler implementation be fine as well, for starters. To just have some json user-data beeing applied. Allowing more input in a second phase, where you implement multipart. | 21:59 |
| harmw | having it read user-data and apply any json found inside would be a nice start I'd guess | 21:59 |
| harmw | however, given there already is a multipart parser... | 22:00 |
| harmw | shoot, I'll just go back to do some work on the fbsd module - tomorrow | 22:01 |
| smoser | harmw, i dont understand what yoiure asking really. | 22:02 |
| smoser | i dont wanto to just overload userdata. like grepping through it to see if there is a key in it. | 22:02 |
| harmw | agreed | 22:04 |
| smoser | harmw, thanks for all your input. i will try to get a re-build of cirros and a proper "0.3.2~pre2" out. | 22:09 |
| smoser | (proper meaning actually available in simplestreams data and without 'local | 22:10 |
| smoser | and tagged and such) | 22:10 |
| smoser | before i leave for holiday (friday). | 22:10 |
| harmw | ah excellent | 22:10 |
| smoser | i just pushed the configuration change. | 22:10 |
| harmw | ill take a look tomorrow | 22:10 |
| === harlowja is now known as harlowja_away | ||
| === harlowja_away is now known as harlowja | ||
| === shardy is now known as shardy_afk | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!