Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel, 2014-06-27

| Channels | #fuel index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
01:19 angdraug jonathan4: are you having https://bugs.launchpad.net/fuel/+bug/1275754 ?
01:21 angdraug jonathan4: or https://bugs.launchpad.net/fuel/+bug/1325483 ?
01:21 angdraug if you find an existing or file a new bug on launchpad, chances that it will be sorted in 5.0.1 or at least 5.1 are much greater
01:27 wrale_ i'd like to see the prepare build env script backported to stable/5.0  and stable/4.1 .. idle wish :)
01:33 xarses joined #fuel
02:03 fandi joined #fuel
02:21 angdraug unfortunately it is :(
02:22 angdraug wish I had an army of minions to delegate such things to :)
02:22 angdraug oh wait, that's exactly why we open sourced Fuel :p
04:00 geekinutah joined #fuel
06:49 guillaume__ joined #fuel
07:29 taj joined #fuel
07:32 ddmitriev joined #fuel
08:18 e0ne joined #fuel
08:22 brain461 joined #fuel
08:27 artem_panchenko joined #fuel
08:34 tnurlygayanov_ joined #fuel
08:46 guillaume__ joined #fuel
09:08 e0ne joined #fuel
09:08 jonathan2 joined #fuel
09:20 tonyco joined #fuel
09:21 tonyco joined #fuel
09:21 igajsin joined #fuel
09:24 tonyco joined #fuel
09:26 igajsin left #fuel
09:58 hyperbaba joined #fuel
10:29 brain461 joined #fuel
11:09 TVR_ what would cause the boot from image, create volume to fail, (image is .raw) and yet after I terminate the failed instance, I can boot from volume of the newly created volume and that works?
11:10 hyperbaba I've just found that heat is not provisioned well on a F5.0 HA deployment
11:11 hyperbaba the heat.conf does not reference to mysql db at all. instead it uses sqlite on localhost thus braking HA concept.
11:33 TVR_ I also have a dhcp network.... how do I disassociate an IP... have no instances on that network, but it issues an IP to a new IP up the pool...
11:34 TVR_ so, as an example... 35 is issued, so 36, 37, and 38 should be next... but instead it issues 39 next
11:34 TVR_ they must be attached to some instances that are now deleted
11:41 omelchek joined #fuel
11:46 TVR_ neutron port-list
11:46 TVR_ pick IP and grab uid
11:46 TVR_ neutron port-delete 425849f2-2aa3-4675-b1d7-5d9c1a302a9e
11:46 TVR_ figured that issue out...
12:18 coreping joined #fuel
12:51 e0ne joined #fuel
13:34 dhblaz joined #fuel
13:50 dhblaz joined #fuel
13:53 mattgriffin joined #fuel
14:24 odyssey4me joined #fuel
14:50 jobewan joined #fuel
14:56 wrale_ is 4.1 any more stable that 5.0?  (if i build the iso from the stable repos for each respective version)?
14:57 wrale_ i bet there are more 4.1 installs in production than 5.0
15:05 MiroslavAnashkin There are more known issues in 5.0 than in 4.1. From the other side - there are issues, existing in 4.1 and fixed in 5.0. Building ISO from stable/4.1 branch currently produces 4.1.2 Beta, with almost all critical 4.1 issues fixed.
15:05 wrale_ will there be an upgrade path to 5.x from 4.x ?  .. Thanks MiroslavAnashkin for the quick and detailed answer
15:07 MiroslavAnashkin Master node upgrade ready functionality is introduced in 5.0. It is completely different with 4.x. Docker in 5.0 is the key component to smooth upgrades in the future.
15:07 wrale_ Thank you.. Good to know
15:07 wrale_ I'll try building 5.0 stable again today :)
15:11 MiroslavAnashkin Building 4.1 is easier. BTW, here are the most recent repositories to build new ISO: http://fuel-repository.mirantis.com/repos/
15:11 wrale_ thanks!
15:20 blahRus joined #fuel
15:23 omelchek joined #fuel
15:36 taj joined #fuel
16:03 dhblaz joined #fuel
16:43 wrale_ it's really too bad that an archive of old isos isn't available, ala centos mirrors
16:43 wrale_ building takes HOURS
16:44 wrale_ and they fail 9/10 times
17:05 wrale_ trying to build 5.0.1 (5.0 stable repo from github), by editing config.mk of the master branch of fuel-main to stable/5.0 fails.. here are two yum package calls causing this:No Match for argument openstack-neutron-vmware; No Match for argument python-daemonize
17:06 wrale_ i would open a bug, but i'm off the beaten path, i guess
17:13 wrale_ these errors seem related to: the two most recent changes to requirements-rpm.txt: https://github.com/stackforge/fuel-main/commits/master/requirements-rpm.txt
17:13 angdraug joined #fuel
17:16 wrale_ see: http://paste.openstack.org/show/85085/
17:23 MiroslavAnashkin wrale_: Please make deep_clean at least before switching branches to build from. Your errors are all about non-synced local repository.
17:26 wrale_ i'm trying the make deep_clean.. thanks
17:28 xarses joined #fuel
17:38 e0ne joined #fuel
17:40 e0ne joined #fuel
17:47 geekinutah joined #fuel
18:08 e0ne joined #fuel
18:15 wrale_ 4.1.2 beta built :)
19:04 wrale_ will slave nodes that were provisioned in a previous install of the master node be re-discovered even if ubuntu was installed on the slave?  it appears that the pxe loader boots the local disk blindly without knowing that the local disk contains an install of ubuntu not registered with the new master node?.. convoluted question, i know.. do i need blank disks on my slaves from the start of a new fuel node install?
19:15 e0ne joined #fuel
19:41 Kupo24z1 joined #fuel
19:43 mattymo joined #fuel
19:43 mutex joined #fuel
19:43 alex_didenko joined #fuel
19:44 bogdando joined #fuel
20:30 e0ne joined #fuel
20:44 wrale_ i temporarily edited the pxe bootloader cfg to go directly to bootstrap.. win!
20:47 sanek_ joined #fuel
20:50 xarses wrale_: if cobbler server doesn't know about the nodes, they should boot to the bootstrap image
20:51 xarses wrale_: nodes will not re 'descover' unless they are in the bootstrap image
20:57 alex_didenko joined #fuel
21:02 alex_didenko joined #fuel
21:04 wrale_ xarses: thank you...all was for naught.. upon install of ubuntu i met an error message: unable to install busybox-initframfs... syslog says a couple of packages don't exist... :(  libc6-udeb and libnewt0.52... this was 4.1.2 beta, i guess.. :(
21:04 wrale_ (controller auto install of ubuntu)
21:04 wrale_ fuel hates me
21:04 xarses wrale_: you should file a bug for that
21:05 wrale_ xarses: will do
21:06 wrale_ surely someone has an image that works and is posted for download outside the portal
21:26 gleam_ joined #fuel
21:36 BradLsys joined #fuel
21:37 agordeev joined #fuel
21:58 angdraug joined #fuel
22:06 geekinutah joined #fuel
22:06 Kupo24z1 xarses: is it possible to install/use puppet on deployed nodes connecting to a specified puppet master?
22:08 xarses Kupo24z1: it should be, we run puppet manually. However I'd avoid putting openstack configs under your puppet master unless you want a big mess
22:08 xarses if you want to run the fuel manifests from puppet master, we used to
22:08 xarses but stripped it out because it was a serious performance bottleneck
22:08 xarses also it didn't scale well
22:09 xarses 3.2.x was the last version with puppet master
22:10 wrale_ when building the latest 5.0.1 pre-release from github repos, should i use the stable 5.0 branch of fuel-main or should i use the master branch of fuel-main ?
22:10 wrale_ (editing for 5.0.1)
22:11 wrale_ ^config.mk
22:11 xarses stable/5.0 would become 5.0.1
22:11 xarses master will become 5.1
22:11 wrale_ that sounds good, so i should change branches before doing the make iso?
22:12 wrale_ and because there is no prep script in stable 5.0, i should prep manually per docs?
22:12 wrale_ perhaps with make deep_clean before to be sure
22:12 xarses I'm not sure, I don't do the ISO builds
22:12 wrale_ thanks though
22:12 wrale_ i'll try again
22:13 wrale_ bbl
22:13 xarses your probably best off if you see what the prep script does and see if it makes sense to do any of it
22:13 wrale_ will do
22:14 wrale_ i think i'll try cherrypicking the prep script, pulling it into stable 5.0 fuel-main
22:14 wrale_ looks about the same as manual process
22:14 wrale_ (but better)
23:56 geekinutah joined #fuel

| Channels | #fuel index | Today | | Search | Google Search | Plain-Text | summary