#theforeman

hi all, is there any way to search for hosts with duplicate interfaces that isn't `GET /api/hosts/:host_id/interfaces` for each host?

searching by domain/subnet only returns the one valid interface for each host

Any Katello devs in the room? I have a question about Foreman hostgoups and node classification.

mhcurlee: more of a foreman question, but feel free to ask

Thanks.... So here is my question. After the host is provisioned, is it possible to change the hostgroup just to re-classify the node from a Puppet perspective. I have tested this and it works. I only completed the needed fields (i.e content view, puppet env, and classes).

mhcurlee: hosts can move between host groups at will. what happens to a host after the change is made greatly depends on how your host groups are configured; in this case with puppet classes being assigned and possibly even parameters

cool. My main concern is all of the fields I did not utilize (i.e. build-centric fields)

mhcurlee: what are some examples? I am assuming you filled in fields like architecture, domain, etc

discr33t_: that is correct. I only completed the fields required from a CV and Puppet perspective.

mhcurlee: gotcha. you shouldn't see any issues. the main things to watch i believe are what puppet classes are applied and what the parameters are.

also remember anything that is overridden for a host should remain overridden for the host even with moving between host groups if there are the same options available

My use-case is supporting servers/applications that drift from the hostgroup they were built from. For example, application centric changes such as additional iptable rules, packages, etc..

mhcurlee: is there a reason the hosts aren't originally built in the proper host group?

i ask because this might be something that can be solved by looking at the host group structure you are using. . .though i don't know your use cases and constraints

discr33t_: That's a good point. I might go that route of trying to create hostgroups for each new application/server. I can't predict how or if a server will drift from the group (some will, some will not). I am starting to think I should play it safe and plan for specific hostgorups at build-time.

mhcurlee: you can curtail some of that if you are using hiera and loosely structure your host group structure based on your hiera hierarchy

I am trying to find out when I click "build" foreman will copy these files from where Ubuntu-16.04-x86_64-linux Ubuntu-16.04-x86_64-initrd.gz

from the repo you setup in "installation media"

that is what I thought thanks...

ubuntu 16.04 keeps getting kernel panic

discr33t_: Thanks for the info. We are running Sat6 in production. I need to find out if hiera is supported (even with a simple yaml backend) on the Sat server. The package is installed, so I would think so.

mhcurlee: are you running some type of all in one? or is your puppet infrastructure separated from your foreman/sat6 infrastructure?

discr33t_: We are running a single satellite 6 server (no external Puppet master/CA)

hi all. I'm using foreman and katello 3.0-rc4 here. due to an issue with katello 3.0rc3, I had some froze tasks/jobs which i had to force unlock. now, when I try to remove a couple of test hosts, I get the following error:

is there any way to protect host from deletion?

what can I do to remove these entries?

kireevco: there's a feature request for that. I think it's not done , yet.

http://projects.theforeman.org/issues/3909 ?

no.. it's a different one.. forget it:)

ther're a couple more recent ones. trying to find the number

mhcurlee: i don't believe sat6 has any issues with using hiera with puppet. i do know that you should be moving smart-class parameters and smart-variables out of foreman and into hiera if you plan on using it

eh, I can't find it now, but I know there are two or three recent'ish feature requests for something like that, kireevco

salman: what was the error with the hosts?

guys - would you know if there will be any differences between Satellite 6.2 and Katello 3.0?

hello. I am recovering my environment from a puppet master cert expiration. All agents are fixed but something is still broke on foreman. I am not getting any reports. I already genreated new ssldir for the puppet agent on foreman. Any thoughts or places to start looking?

Hi all, I'm upgrading Foreman from 1.8.4 to 1.9.x (and eventually to 1.10, 1.11...). I'm at http://theforeman.org/manuals/1.9/index.html#3.... , "Step 3 (A) - Database migration and cleanup", and `foreman-rake db:migrate` fails with the error `Gem loading error: cannot load such file -- audited/activerecord`. Am I missing a gem? http://pastebin.com/8YuxS5hY

I added `foreman-rake db:migrate --trace` here: http://pastebin.com/bd64i5Kt

george__: satellite 6.2 will end up containing more bug fixes than 3.0

Hi! I'm having a bit of a hard time with the provisioning side of things, when I provision a new server with 1 interface everything works perfect but when I provision another host with 2 interfaces the server seems to get dhcp for the correct interface (2nd) but it doesn't do a tftp boot, keeps telling me tftp open timeout. Any ideas?

other than that, very similar

The host with 2 interfaces doesn't even try to fetch the boot files.

jsherrill: it was related to #14759

jsherrill: #14759 is http://theforeman.org/issues/14759 "Bug #14759: Katello 3.0 RC3 depends on Foreman 1.12 - Packaging - Foreman"

salman: ah was this an upgrade from RC3 to RC4 ?

they were stuck in a bad state in rc3, I had to force unlock the tasks. did an upgrade to rc4, but now attempting to re-register, or removing the hosts from within the foreman ui thorws the candlepin error

all new hosts register fine, no problems since rc4

salman: okay good to hear, so its just data cleanup from RC3

salman: i may need to write a few lines of ruby code to clean those up, give me a few

Looks like my issue may be related to using Scientific Linux, with SCL. Looks like I might have run into the same issue as sagarC_ here: https://botbot.me/freenode/theforeman/2016-04-0...

jsherrill: thanks

This fixed my problem. `yum reinstall ruby193-rubygem-activesupport`

salman: thank you in any case!

salman: https://gist.github.com/jlsherrill/44a53c561d1a...

I have a general question about upgrading Foreman. The manual at http://theforeman.org/manuals/1.11/index.html#s... says that commands like `foreman-rake db:migrate` should produce no errors or output. Is that true for you? I've updated Foreman from 1.4 to 1.9, and that command has always produced some sort of output.

trying this now, jsherrill

jsherrill: I get a nomethoderror: NoMethodError: undefined method `destroy!' for #<Host::Managed::ActiveRecord_Relation:0x000000099c47b8> ( http://pastebin.com/6RMMj6fH )

gigglesworth_: did you jump from 1.4 to 1.9 at once? could you pastebin the actual error?

salman: ahh shoot one sec

gigglesworth_: oh, or is it about the output as such?

salman: refresh (changed it to destroy)

tried that as well. it needs an argument passed

I wasn't sure what that should be

thinking about it, I'd even _expect_ output also on a successful run for the respective statuses

http://pastebin.com/GriUJDkc -- that's what I get back with host.destroy

hey all, i was wondering if anyone had any information on how to find the 'content-source-id' used for creating hostgroups via hammer?

i know the main katello server itself is 1 and obviously the order created would make it somewhat simple to decipher which one is 2 or 3 or 4, but I see no way to view this parameter for final verification other than going to the GUI.

jsherrill_afk: thanks for pointing me in the right direction! I think it's fixed, now. passing host.destroy() the host id turned up the table in foreman db where the record was stuck. deleting from that table, and then running destroy again removed the host

Foreman have support for Puppet 4.4 ?

The documentation says the installer does not

but I want to be sure packages/source do

1.12 will

and I'm able to re-register the host with katello

it currently does not

Ah, okay. Thanks

Do you know when Foreman is tentatively looking to release 1.12 ?

there is a mailing list post somewhere... one sec

mmoll : Right, I'm just asking about the output in general. There's no specific error. I'm just confused about the documentation.