Foreman-installer-staypuft crashes on install

Marek,
Thanks for your note, but die the same applies for CentOS 6? Both links seem to refer to CentOS 7.

IB

··· - Ignacio Bravo LTG Federal Mb: (571) 224-6046

On Aug 22, 2014, at 11:46 AM, Marek Hulan mhulan@redhat.com wrote:

Hello,

since CentOS is not supported, noone noticed that it stopped working on
CentOS. I made a fix on both installer and staypuft side today. You’ll have to
update and rerun the installer (0.3.2 should get to nightly tomorrow) to
precreate staypuft OS v 7. Staypuft fix was not yet merged but you can apply it
manually, see [1] for the patch.

Before creating new deploy you must also fix one issue with CentOS installation
media. For some reason, CentOS 7.0 mirror does not follow standard path so as
a workaround you have to navigate to Hosts -> Installation Media -> CentOS
Mirror and update path to http://mirror.centos.org/centos/$major/os/$arch.
More information at [2].

Fixes should get in and be packaged in the beginning of next week. I’d be
happy if you tested them and shared your results.

[1] https://github.com/theforeman/staypuft/pull/271
[2] Bug #6884: Centos mirror in "Installation Media" doesn't work for versions without $minor - Foreman

On Thursday 21 of August 2014 17:49:52 Marek Hulan wrote:
I was able to reproduce it. I’ll look have a look at it tomorrow and will
let you know.

After downloading images I rerun installer and he told me that everything
success:

Starting configuration…
Starting to seed provisioning data
Use ‘base_CentOS_6’ hostgroup for provisioning

Success!

But UI told me same "missing Operatingsystem"
and in foreman/production.log same error:

Started GET “/deployments/new” for 10.10.10.1 at 2014-08-21 17:14:12 +0300
Processing by Staypuft::DeploymentsController#new as HTML
Operation FAILED: missing Operatingsystem

Rendered common/500.html.erb within staypuft/layouts/staypuft (3.8ms)
Rendered layouts/base.html.erb (0.9ms)
Rendered /opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/

views/staypuft/layouts/application.html.erb (1.2ms)
Completed 500 Internal Server Error in 79ms (Views: 6.7ms | ActiveRecord:
13.4ms)

On Thursday, August 21, 2014 5:01:15 PM UTC+3, Marek Hulan wrote:
There seems to be some ordering issue with discovery images, target
directory
did not exist yet. When there appears an error during installation,
configuration won’t continue. Your staypuft installation resulted in
being
installed but not configured.

Now when you fixed discovery images issue, can you rerun installer once
more?
Until it shows that you can use some hostgroup for provisioning you
can’t
consider staypuft to be ready. But if discovery images was the only
issue
it
should pass now.

On Thursday 21 of August 2014 03:20:15 Nomnom Fuser wrote:
Ok. Yet another fresh install. Centos 6.5 x86_64
repos:

base

                    CentOS-6 - Base

epel

                    Extra Packages for Enterprise Linux 6 -
                    x86_64

extras

                    CentOS-6 - Extras

foreman

                   Foreman nightly

foreman-plugins

                   Foreman plugins nightly

scl

                   CentOS-6 - SCL

updates

                   CentOS-6 - Updates

rpm -qa | grep foreman

foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noarch
foreman-installer-staypuft-0.3.0-1.el6.noarch
foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noarch
rubygem-foreman_api-0.1.11-1.el6.noarch
foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch

First install with "staypuft-installer

–foreman-configure-epel-repo=true

–foreman-plugin-discovery-install-images=true" :

[ INFO 2014-08-21 13:06:14 main] Executing hooks in group post
[DEBUG 2014-08-21 13:06:14 main] Hook

/usr/share/foreman-installer/hooks/

post/10-setup_provisioning.rb returned false
[ INFO 2014-08-21 13:06:14 main] All hooks in group post finished
[DEBUG 2014-08-21 13:06:14 main] Exit with status code: 6 (signal was
6)
[ERROR 2014-08-21 13:06:14 main] Repeating errors encountered during

run:

[ERROR 2014-08-21 13:06:14 main]

/Stage[main]/Foreman::Plugin::Discovery/

Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-late
st
.

el6.iso-vmlinuz]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery-> > > >
image-latest.el6.iso-vmlinuz]/returns: change from notrun to 0 failed:
/usr/

bin/curl -s

http://downloads.theforeman.org/discovery/releases/latest/foreman-discov
er
y

-image-latest.el6.iso-vmlinuz -o
/var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-vmlinuz
returned 23 instead of one of [0] at
/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp:
6
[ERROR 2014-08-21 13:06:14 main]

/Stage[main]/Foreman::Plugin::Discovery/

Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-late
st
.

el6.iso-img]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery-imag
e-> > > latest.el6.iso-img]/returns: change from notrun to 0 failed:
/usr/bin/curl> >

-s

http://downloads.theforeman.org/discovery/releases/latest/foreman-discov
er
y

-image-latest.el6.iso-img -o
/var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-img

returned

23 instead of one of [0] at
/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp:

Ok, I download images manually using same curl command.
Second run with same options. Seems ok. “Success!”

Login to UI, go to new deployment of openstack, aaaannnndddd…

Started GET “/deployments/new” for 10.10.10.1 at 2014-08-21 13:14:11

+0300

Processing by Staypuft::DeploymentsController#new as HTML
Operation FAILED: missing Operatingsystem

Rendered common/500.html.erb within staypuft/layouts/staypuft
(7.9ms)
Rendered layouts/base.html.erb (2.0ms)
Rendered
/opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/

views/staypuft/layouts/application.html.erb (3.6ms)
Completed 500 Internal Server Error in 508ms (Views: 16.6ms |

ActiveRecord:

34.1ms)

after installer:

rpm -qa | grep foreman

foreman-installer-staypuft-0.3.0-1.el6.noarch
ruby193-rubygem-foreman-tasks-0.6.4-1.el6.noarch
foreman-postgresql-1.7.0-0.develop.201408201145git87a42cb.el6.noarch
foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noarch
rubygem-foreman_api-0.1.11-1.el6.noarch
foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch
foreman-1.7.0-0.develop.201408201145git87a42cb.el6.noarch
foreman-proxy-1.7.0-0.develop.201408191935git7124635.el6.noarch
foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noarch
ruby193-rubygem-foreman_discovery-1.3.0-1.el6.noarch

Seems like I need install openstack manually from source or using rdo
without provisioning %(

On Thursday, August 21, 2014 10:33:16 AM UTC+3, Marek Hulan wrote:
Hello,

This error would indicate that installer did not finish all steps
successfully.
Could you rerun the installer and try again?


Marek


You received this message because you are subscribed to a topic in the Google Groups “Foreman users” group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/foreman-users/MUM5iAwwVUc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to foreman-users+unsubscribe@googlegroups.com.
To post to this group, send email to foreman-users@googlegroups.com.
Visit this group at http://groups.google.com/group/foreman-users.
For more options, visit https://groups.google.com/d/optout.

Marek,

Thanks for the patch. I updated the file with your details, but the
installation was still failing during the image retrieval phase. I thought
that the issue was related to the files being downloaded before the
directory was created, so I created the directory and it worked! Here is
the script that I was using for the installation:

==Install==
yum install -y
http://yum.theforeman.org/nightly/el6/x86_64/foreman-release.rpm
yum -y localinstall
http://mirror.pnl.gov/epel/6/x86_64/epel-release-6-8.noarch.rpm
yum -y install centos-release-SCL
yum -y install foreman-plugin-staypuft
yum -y install foreman-installer-staypuft

mkdir /var/lib/tftpboot
mkdir /var/lib/tftpboot/boot

Now modify the file:

/opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/models/staypuft/deployment.rb
based on your instructions and then:

staypuft-installer --foreman-configure-epel-repo=true
–foreman-plugin-discovery-install-images=true

== end ====

It appears that the module with problems is this one:
/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp

··· On Friday, August 22, 2014 12:26:55 PM UTC-4, Nomnom Fuser wrote: > > Thx for patch. I'll try it next week and share my results. > > You refer to issue with CentOS 7 and I use Centos 6.5 as host and for > provisioning. > Did I understand correctly that new CentOS versioning (eg. 7.0.1234) > brakes some puppet scripts that runs with staypuft-installer? > > On Friday, August 22, 2014 6:46:44 PM UTC+3, Marek Hulan wrote: >> >> Hello, >> >> since CentOS is not supported, noone noticed that it stopped working on >> CentOS. I made a fix on both installer and staypuft side today. You'll >> have to >> update and rerun the installer (0.3.2 should get to nightly tomorrow) to >> precreate staypuft OS v 7. Staypuft fix was not yet merged but you can >> apply it >> manually, see [1] for the patch. >> >> Before creating new deploy you must also fix one issue with CentOS >> installation >> media. For some reason, CentOS 7.0 mirror does not follow standard path >> so as >> a workaround you have to navigate to Hosts -> Installation Media -> >> CentOS >> Mirror and update path to http://mirror.centos.org/centos/$major/os/$arch. >> >> More information at [2]. >> >> Fixes should get in and be packaged in the beginning of next week. I'd be >> happy if you tested them and shared your results. >> >> [1] https://github.com/theforeman/staypuft/pull/271 >> [2] http://projects.theforeman.org/issues/6884 >> >> On Thursday 21 of August 2014 17:49:52 Marek Hulan wrote: >> > I was able to reproduce it. I'll look have a look at it tomorrow and >> will >> > let you know. >> > >> > > After downloading images I rerun installer and he told me that >> everything >> > > success: >> > > >> > > Starting configuration... >> > > Starting to seed provisioning data >> > > Use 'base_CentOS_6' hostgroup for provisioning >> > > >> > > Success! >> > > * Foreman is running at https://openstack-core.cloud >> > > >> > > Initial credentials are admin / df4Rauujuwxdede6 >> > > >> > > * Foreman Proxy is running at https://openstack-core.cloud:8443 >> > > * Puppetmaster is running at port 8140 >> > > The full log is at /var/log/foreman-installer/foreman-installer.log >> > > >> > > But UI told me same "missing Operatingsystem" >> > > and in foreman/production.log same error: >> > > >> > > Started GET "/deployments/new" for 10.10.10.1 at 2014-08-21 17:14:12 >> +0300 >> > > Processing by Staypuft::DeploymentsController#new as HTML >> > > Operation FAILED: missing Operatingsystem >> > > >> > > Rendered common/500.html.erb within staypuft/layouts/staypuft >> (3.8ms) >> > > Rendered layouts/base.html.erb (0.9ms) >> > > Rendered >> /opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/ >> > > >> > > views/staypuft/layouts/application.html.erb (1.2ms) >> > > Completed 500 Internal Server Error in 79ms (Views: 6.7ms | >> ActiveRecord: >> > > 13.4ms) >> > > >> > > On Thursday, August 21, 2014 5:01:15 PM UTC+3, Marek Hulan wrote: >> > > > There seems to be some ordering issue with discovery images, target >> > > > directory >> > > > did not exist yet. When there appears an error during installation, >> > > > configuration won't continue. Your staypuft installation resulted >> in >> > > > being >> > > > installed but not configured. >> > > > >> > > > Now when you fixed discovery images issue, can you rerun installer >> once >> > > > more? >> > > > Until it shows that you can use some hostgroup for provisioning you >> > > > can't >> > > > consider staypuft to be ready. But if discovery images was the only >> > > > issue >> > > > it >> > > > should pass now. >> > > > >> > > > On Thursday 21 of August 2014 03:20:15 Nomnom Fuser wrote: >> > > > > Ok. Yet another fresh install. Centos 6.5 x86_64 >> > > > > repos: >> > > > > >> > > > > >> > > > > base >> > > > > >> > > > > CentOS-6 - Base >> > > > > >> > > > > epel >> > > > > >> > > > > Extra Packages for Enterprise Linux 6 - >> > > > > x86_64 >> > > > > >> > > > > extras >> > > > > >> > > > > CentOS-6 - Extras >> > > > > >> > > > > foreman >> > > > > >> > > > > Foreman nightly >> > > > > >> > > > > foreman-plugins >> > > > > >> > > > > Foreman plugins nightly >> > > > > >> > > > > scl >> > > > > >> > > > > CentOS-6 - SCL >> > > > > >> > > > > updates >> > > > > >> > > > > CentOS-6 - Updates >> > > > > >> > > > > # rpm -qa | grep foreman >> > > > > foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noarch >> > > > > foreman-installer-staypuft-0.3.0-1.el6.noarch >> > > > > foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noarch >> > > > > rubygem-foreman_api-0.1.11-1.el6.noarch >> > > > > >> foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch >> > > > > >> > > > > First install with "staypuft-installer >> > > > >> > > > --foreman-configure-epel-repo=true >> > > > >> > > > > --foreman-plugin-discovery-install-images=true" : >> > > > > >> > > > > [ INFO 2014-08-21 13:06:14 main] Executing hooks in group post >> > > > > [DEBUG 2014-08-21 13:06:14 main] Hook >> > > > >> > > > /usr/share/foreman-installer/hooks/ >> > > > >> > > > > post/10-setup_provisioning.rb returned false >> > > > > [ INFO 2014-08-21 13:06:14 main] All hooks in group post finished >> > > > > [DEBUG 2014-08-21 13:06:14 main] Exit with status code: 6 (signal >> was >> > > > > 6) >> > > > > [ERROR 2014-08-21 13:06:14 main] Repeating errors encountered >> during >> > > > >> > > > run: >> > > > > [ERROR 2014-08-21 13:06:14 main] >> > > > >> > > > /Stage[main]/Foreman::Plugin::Discovery/ >> > > > >> > > > >> Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-late >> > > > st >> > > > . >> > > > >> > > > >> el6.iso-vmlinuz]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery-> > >> > > >> > > > image-latest.el6.iso-vmlinuz]/returns: change from notrun to 0 >> failed: >> > > > /usr/ >> > > > >> > > > > bin/curl -s >> > > > >> > > > >> http://downloads.theforeman.org/discovery/releases/latest/foreman-discov >> > > > er >> > > > y >> > > > >> > > > > -image-latest.el6.iso-vmlinuz -o >> > > > > >> /var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-vmlinuz >> > > > > returned 23 instead of one of [0] at >> > > > > >> /usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp: >> > > > > 6 >> > > > > [ERROR 2014-08-21 13:06:14 main] >> > > > >> > > > /Stage[main]/Foreman::Plugin::Discovery/ >> > > > >> > > > >> Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-late >> > > > st >> > > > . >> > > > >> > > > >> el6.iso-img]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery-imag >> > > > e-> > > latest.el6.iso-img]/returns: change from notrun to 0 >> failed: >> > > > /usr/bin/curl> > >> > > > > -s >> > > > >> > > > >> http://downloads.theforeman.org/discovery/releases/latest/foreman-discov >> > > > er >> > > > y >> > > > >> > > > > -image-latest.el6.iso-img -o >> > > > > /var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-img >> > > > >> > > > returned >> > > > >> > > > > 23 instead of one of [0] at >> > > > > >> /usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp: >> > > > > >> > > > > Ok, I download images manually using same curl command. >> > > > > Second run with same options. Seems ok. "Success!" >> > > > > >> > > > > Login to UI, go to new deployment of openstack, aaaannnndddd..... >> > > > > >> > > > > Started GET "/deployments/new" for 10.10.10.1 at 2014-08-21 >> 13:14:11 >> > > > >> > > > +0300 >> > > > >> > > > > Processing by Staypuft::DeploymentsController#new as HTML >> > > > > Operation FAILED: missing Operatingsystem >> > > > > >> > > > > Rendered common/500.html.erb within staypuft/layouts/staypuft >> > > > > (7.9ms) >> > > > > Rendered layouts/base.html.erb (2.0ms) >> > > > > Rendered >> > > > > /opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/ >> > > > > >> > > > > views/staypuft/layouts/application.html.erb (3.6ms) >> > > > > Completed 500 Internal Server Error in 508ms (Views: 16.6ms | >> > > > >> > > > ActiveRecord: >> > > > > 34.1ms) >> > > > > >> > > > > after installer: >> > > > > >> > > > > # rpm -qa | grep foreman >> > > > > foreman-installer-staypuft-0.3.0-1.el6.noarch >> > > > > ruby193-rubygem-foreman-tasks-0.6.4-1.el6.noarch >> > > > > >> foreman-postgresql-1.7.0-0.develop.201408201145git87a42cb.el6.noarch >> > > > > foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noarch >> > > > > rubygem-foreman_api-0.1.11-1.el6.noarch >> > > > > >> foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch >> > > > > foreman-1.7.0-0.develop.201408201145git87a42cb.el6.noarch >> > > > > foreman-proxy-1.7.0-0.develop.201408191935git7124635.el6.noarch >> > > > > foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noarch >> > > > > ruby193-rubygem-foreman_discovery-1.3.0-1.el6.noarch >> > > > > >> > > > > Seems like I need install openstack manually from source or using >> rdo >> > > > > without provisioning %( >> > > > > >> > > > > On Thursday, August 21, 2014 10:33:16 AM UTC+3, Marek Hulan >> wrote: >> > > > > > Hello, >> > > > > > >> > > > > > This error would indicate that installer did not finish all >> steps >> > > > > > successfully. >> > > > > > Could you rerun the installer and try again? >> >> -- >> Marek >> >

Hello

> You refer to issue with CentOS 7 and I use Centos 6.5 as host and for
> provisioning.

It's a bit confusing but we require CentOS 6.5 for provisioning host (where we
install Foreman) but we provision CetnOS 7 on other machines on which you
deploy OpenStack.

> Did I understand correctly that new CentOS versioning (eg. 7.0.1234) brakes
> some puppet scripts that runs with staypuft-installer?

More precisely, it breaks default foreman installation media which are used
during provisioning. The revision number in CentOS path is not standard,
previously they had only $major.$minor. As a workaround staypuft-installer
could modify it but I'd like to see the fix for upstream foreman first, there
should be some better way. Or CentOS guys will fix their mirrors :slight_smile:

Anyway I built installer 0.3.3 today which contains synxtax error fix, so if
testing, use this (should get into nightly repo shortly). I'll try to get
staypuft fix in today as well.

··· -- Marek

On Friday 22 of August 2014 09:26:55 Nomnom Fuser wrote:

Thx for patch. I’ll try it next week and share my results.

On Friday, August 22, 2014 6:46:44 PM UTC+3, Marek Hulan wrote:

Hello,

since CentOS is not supported, noone noticed that it stopped working on
CentOS. I made a fix on both installer and staypuft side today. You’ll
have to
update and rerun the installer (0.3.2 should get to nightly tomorrow) to
precreate staypuft OS v 7. Staypuft fix was not yet merged but you can
apply it
manually, see [1] for the patch.

Before creating new deploy you must also fix one issue with CentOS
installation
media. For some reason, CentOS 7.0 mirror does not follow standard path so
as
a workaround you have to navigate to Hosts -> Installation Media -> CentOS
Mirror and update path to http://mirror.centos.org/centos/$major/os/$arch.

More information at [2].

Fixes should get in and be packaged in the beginning of next week. I’d be
happy if you tested them and shared your results.

[1] https://github.com/theforeman/staypuft/pull/271
[2] Bug #6884: Centos mirror in "Installation Media" doesn't work for versions without $minor - Foreman

On Thursday 21 of August 2014 17:49:52 Marek Hulan wrote:

I was able to reproduce it. I’ll look have a look at it tomorrow and

will

let you know.

After downloading images I rerun installer and he told me that

everything

success:

Starting configuration…
Starting to seed provisioning data
Use ‘base_CentOS_6’ hostgroup for provisioning

Success!

But UI told me same "missing Operatingsystem"
and in foreman/production.log same error:

Started GET “/deployments/new” for 10.10.10.1 at 2014-08-21 17:14:12

+0300

Processing by Staypuft::DeploymentsController#new as HTML
Operation FAILED: missing Operatingsystem

Rendered common/500.html.erb within staypuft/layouts/staypuft

(3.8ms)

Rendered layouts/base.html.erb (0.9ms)
Rendered

/opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/

views/staypuft/layouts/application.html.erb (1.2ms)
Completed 500 Internal Server Error in 79ms (Views: 6.7ms |

ActiveRecord:

13.4ms)

On Thursday, August 21, 2014 5:01:15 PM UTC+3, Marek Hulan wrote:

There seems to be some ordering issue with discovery images, target
directory
did not exist yet. When there appears an error during installation,
configuration won’t continue. Your staypuft installation resulted in
being
installed but not configured.

Now when you fixed discovery images issue, can you rerun installer

once

more?
Until it shows that you can use some hostgroup for provisioning you
can’t
consider staypuft to be ready. But if discovery images was the only
issue
it
should pass now.

On Thursday 21 of August 2014 03:20:15 Nomnom Fuser wrote:

Ok. Yet another fresh install. Centos 6.5 x86_64
repos:

base

                     CentOS-6 - Base

epel

                     Extra Packages for Enterprise Linux 6 -
                     x86_64

extras

                     CentOS-6 - Extras

foreman

                    Foreman nightly

foreman-plugins

                    Foreman plugins nightly

scl

                    CentOS-6 - SCL

updates

                    CentOS-6 - Updates

rpm -qa | grep foreman

foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noarch
foreman-installer-staypuft-0.3.0-1.el6.noarch
foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noarch
rubygem-foreman_api-0.1.11-1.el6.noarch

foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch

First install with "staypuft-installer

–foreman-configure-epel-repo=true

–foreman-plugin-discovery-install-images=true" :

[ INFO 2014-08-21 13:06:14 main] Executing hooks in group post
[DEBUG 2014-08-21 13:06:14 main] Hook

/usr/share/foreman-installer/hooks/

post/10-setup_provisioning.rb returned false
[ INFO 2014-08-21 13:06:14 main] All hooks in group post finished
[DEBUG 2014-08-21 13:06:14 main] Exit with status code: 6 (signal

was

[ERROR 2014-08-21 13:06:14 main] Repeating errors encountered

during

run:

[ERROR 2014-08-21 13:06:14 main]

/Stage[main]/Foreman::Plugin::Discovery/

Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-late

st
.

el6.iso-vmlinuz]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery->

image-latest.el6.iso-vmlinuz]/returns: change from notrun to 0

failed:

/usr/

bin/curl -s

http://downloads.theforeman.org/discovery/releases/latest/foreman-discov

er
y

-image-latest.el6.iso-vmlinuz -o

/var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-vmlinuz

returned 23 instead of one of [0] at

/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp:

6
[ERROR 2014-08-21 13:06:14 main]

/Stage[main]/Foreman::Plugin::Discovery/

Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-late

st
.

el6.iso-img]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery-imag

e-> > > latest.el6.iso-img]/returns: change from notrun to 0 failed:
/usr/bin/curl> >

-s

http://downloads.theforeman.org/discovery/releases/latest/foreman-discov

er
y

-image-latest.el6.iso-img -o
/var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-img

returned

23 instead of one of [0] at

/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp:

Ok, I download images manually using same curl command.
Second run with same options. Seems ok. “Success!”

Login to UI, go to new deployment of openstack, aaaannnndddd…

Started GET “/deployments/new” for 10.10.10.1 at 2014-08-21

13:14:11

+0300

Processing by Staypuft::DeploymentsController#new as HTML
Operation FAILED: missing Operatingsystem

Rendered common/500.html.erb within staypuft/layouts/staypuft
(7.9ms)
Rendered layouts/base.html.erb (2.0ms)
Rendered
/opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/

views/staypuft/layouts/application.html.erb (3.6ms)
Completed 500 Internal Server Error in 508ms (Views: 16.6ms |

ActiveRecord:

34.1ms)

after installer:

rpm -qa | grep foreman

foreman-installer-staypuft-0.3.0-1.el6.noarch
ruby193-rubygem-foreman-tasks-0.6.4-1.el6.noarch

foreman-postgresql-1.7.0-0.develop.201408201145git87a42cb.el6.noarch

foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noarch
rubygem-foreman_api-0.1.11-1.el6.noarch

foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch

foreman-1.7.0-0.develop.201408201145git87a42cb.el6.noarch
foreman-proxy-1.7.0-0.develop.201408191935git7124635.el6.noarch
foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noarch
ruby193-rubygem-foreman_discovery-1.3.0-1.el6.noarch

Seems like I need install openstack manually from source or using

rdo

without provisioning %(

On Thursday, August 21, 2014 10:33:16 AM UTC+3, Marek Hulan wrote:

Hello,

This error would indicate that installer did not finish all

steps

successfully.
Could you rerun the installer and try again?

Hi,

you're right, it seems like some ordering issue in upstream puppet module.
I'll try to look at that. Meanwhile the workaround is what you say, you can
use shorter variant

mkdir -p /var/lib/tftpboot/boot

Thanks for the report

··· -- Marek

On Friday 22 of August 2014 14:34:04 Ignacio Bravo wrote:

Marek,

Thanks for the patch. I updated the file with your details, but the
installation was still failing during the image retrieval phase. I thought
that the issue was related to the files being downloaded before the
directory was created, so I created the directory and it worked! Here is
the script that I was using for the installation:

==Install==
yum install -y
http://yum.theforeman.org/nightly/el6/x86_64/foreman-release.rpm
yum -y localinstall
http://mirror.pnl.gov/epel/6/x86_64/epel-release-6-8.noarch.rpm
yum -y install centos-release-SCL
yum -y install foreman-plugin-staypuft
yum -y install foreman-installer-staypuft

mkdir /var/lib/tftpboot
mkdir /var/lib/tftpboot/boot

Now modify the file:

/opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/models/staypuft
/deployment.rb based on your instructions and then:

staypuft-installer --foreman-configure-epel-repo=true
–foreman-plugin-discovery-install-images=true

== end ====

It appears that the module with problems is this one:
/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp

On Friday, August 22, 2014 12:26:55 PM UTC-4, Nomnom Fuser wrote:

Thx for patch. I’ll try it next week and share my results.

You refer to issue with CentOS 7 and I use Centos 6.5 as host and for
provisioning.
Did I understand correctly that new CentOS versioning (eg. 7.0.1234)
brakes some puppet scripts that runs with staypuft-installer?

On Friday, August 22, 2014 6:46:44 PM UTC+3, Marek Hulan wrote:

Hello,

since CentOS is not supported, noone noticed that it stopped working on
CentOS. I made a fix on both installer and staypuft side today. You’ll
have to
update and rerun the installer (0.3.2 should get to nightly tomorrow) to
precreate staypuft OS v 7. Staypuft fix was not yet merged but you can
apply it
manually, see [1] for the patch.

Before creating new deploy you must also fix one issue with CentOS
installation
media. For some reason, CentOS 7.0 mirror does not follow standard path
so as
a workaround you have to navigate to Hosts -> Installation Media ->
CentOS
Mirror and update path to
http://mirror.centos.org/centos/$major/os/$arch.

More information at [2].

Fixes should get in and be packaged in the beginning of next week. I’d be
happy if you tested them and shared your results.

[1] https://github.com/theforeman/staypuft/pull/271
[2] Bug #6884: Centos mirror in "Installation Media" doesn't work for versions without $minor - Foreman

On Thursday 21 of August 2014 17:49:52 Marek Hulan wrote:

I was able to reproduce it. I’ll look have a look at it tomorrow and

will

let you know.

After downloading images I rerun installer and he told me that

everything

success:

Starting configuration…
Starting to seed provisioning data
Use ‘base_CentOS_6’ hostgroup for provisioning

Success!

But UI told me same "missing Operatingsystem"
and in foreman/production.log same error:

Started GET “/deployments/new” for 10.10.10.1 at 2014-08-21 17:14:12

+0300

Processing by Staypuft::DeploymentsController#new as HTML
Operation FAILED: missing Operatingsystem

Rendered common/500.html.erb within staypuft/layouts/staypuft

(3.8ms)

Rendered layouts/base.html.erb (0.9ms)
Rendered

/opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/

views/staypuft/layouts/application.html.erb (1.2ms)
Completed 500 Internal Server Error in 79ms (Views: 6.7ms |

ActiveRecord:

13.4ms)

On Thursday, August 21, 2014 5:01:15 PM UTC+3, Marek Hulan wrote:

There seems to be some ordering issue with discovery images, target
directory
did not exist yet. When there appears an error during installation,
configuration won’t continue. Your staypuft installation resulted

in

being
installed but not configured.

Now when you fixed discovery images issue, can you rerun installer

once

more?
Until it shows that you can use some hostgroup for provisioning you
can’t
consider staypuft to be ready. But if discovery images was the only
issue
it
should pass now.

On Thursday 21 of August 2014 03:20:15 Nomnom Fuser wrote:

Ok. Yet another fresh install. Centos 6.5 x86_64
repos:

base

                     CentOS-6 - Base

epel

                     Extra Packages for Enterprise Linux 6 -
                     x86_64

extras

                     CentOS-6 - Extras

foreman

                    Foreman nightly

foreman-plugins

                    Foreman plugins nightly

scl

                    CentOS-6 - SCL

updates

                    CentOS-6 - Updates

rpm -qa | grep foreman

foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noarch
foreman-installer-staypuft-0.3.0-1.el6.noarch
foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noarch
rubygem-foreman_api-0.1.11-1.el6.noarch

foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch

First install with "staypuft-installer

–foreman-configure-epel-repo=true

–foreman-plugin-discovery-install-images=true" :

[ INFO 2014-08-21 13:06:14 main] Executing hooks in group post
[DEBUG 2014-08-21 13:06:14 main] Hook

/usr/share/foreman-installer/hooks/

post/10-setup_provisioning.rb returned false
[ INFO 2014-08-21 13:06:14 main] All hooks in group post finished
[DEBUG 2014-08-21 13:06:14 main] Exit with status code: 6 (signal

was

[ERROR 2014-08-21 13:06:14 main] Repeating errors encountered

during

run:

[ERROR 2014-08-21 13:06:14 main]

/Stage[main]/Foreman::Plugin::Discovery/

Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-late

st
.

el6.iso-vmlinuz]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery->

image-latest.el6.iso-vmlinuz]/returns: change from notrun to 0

failed:

/usr/

bin/curl -s

http://downloads.theforeman.org/discovery/releases/latest/foreman-discov

er
y

-image-latest.el6.iso-vmlinuz -o

/var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-vmlinuz

returned 23 instead of one of [0] at

/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp:

6
[ERROR 2014-08-21 13:06:14 main]

/Stage[main]/Foreman::Plugin::Discovery/

Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-late

st
.

el6.iso-img]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery-imag

e-> > > latest.el6.iso-img]/returns: change from notrun to 0

failed:

/usr/bin/curl> >

-s

http://downloads.theforeman.org/discovery/releases/latest/foreman-discov

er
y

-image-latest.el6.iso-img -o
/var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-img

returned

23 instead of one of [0] at

/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp:

Ok, I download images manually using same curl command.
Second run with same options. Seems ok. “Success!”

Login to UI, go to new deployment of openstack, aaaannnndddd…

Started GET “/deployments/new” for 10.10.10.1 at 2014-08-21

13:14:11

+0300

Processing by Staypuft::DeploymentsController#new as HTML
Operation FAILED: missing Operatingsystem

Rendered common/500.html.erb within staypuft/layouts/staypuft
(7.9ms)
Rendered layouts/base.html.erb (2.0ms)
Rendered
/opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/

views/staypuft/layouts/application.html.erb (3.6ms)
Completed 500 Internal Server Error in 508ms (Views: 16.6ms |

ActiveRecord:

34.1ms)

after installer:

rpm -qa | grep foreman

foreman-installer-staypuft-0.3.0-1.el6.noarch
ruby193-rubygem-foreman-tasks-0.6.4-1.el6.noarch

foreman-postgresql-1.7.0-0.develop.201408201145git87a42cb.el6.noarch

foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noarch
rubygem-foreman_api-0.1.11-1.el6.noarch

foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch

foreman-1.7.0-0.develop.201408201145git87a42cb.el6.noarch
foreman-proxy-1.7.0-0.develop.201408191935git7124635.el6.noarch
foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noarch
ruby193-rubygem-foreman_discovery-1.3.0-1.el6.noarch

Seems like I need install openstack manually from source or using

rdo

without provisioning %(

On Thursday, August 21, 2014 10:33:16 AM UTC+3, Marek Hulan > >> > >> wrote:

Hello,

This error would indicate that installer did not finish all

steps

successfully.
Could you rerun the installer and try again?

>
> Hello
>
> > You refer to issue with CentOS 7 and I use Centos 6.5 as host and for
> > provisioning.
>
> It's a bit confusing but we require CentOS 6.5 for provisioning host
> (where we
> install Foreman) but we provision CetnOS 7 on other machines on which you
> deploy OpenStack.
>

Ah. I see. Is it possible to use CentOS 6.5 for host node AND openstack
deployment?
CentOS 7 is so new that I'm afraid install it on prod right now + I don't
know is Ceph storage already compatible with CentOS 7

>
> > Did I understand correctly that new CentOS versioning (eg. 7.0.1234)
> brakes
> > some puppet scripts that runs with staypuft-installer?
>
> More precisely, it breaks default foreman installation media which are
> used
> during provisioning. The revision number in CentOS path is not standard,
> previously they had only $major.$minor. As a workaround staypuft-installer
> could modify it but I'd like to see the fix for upstream foreman first,
> there
> should be some better way. Or CentOS guys will fix their mirrors :slight_smile:
>

I think that CenOS guys won't fix that :slight_smile: . Seems that they decided to use
new version numbering.
For now I use dirty hack with mirror url
like http://mirror.centos.org/centos/7/os/$arch

··· On Monday, August 25, 2014 10:46:48 AM UTC+3, Marek Hulan wrote:

Anyway I built installer 0.3.3 today which contains synxtax error fix, so
if
testing, use this (should get into nightly repo shortly). I’ll try to get
staypuft fix in today as well.


Marek

On Friday 22 of August 2014 09:26:55 Nomnom Fuser wrote:

Thx for patch. I’ll try it next week and share my results.

On Friday, August 22, 2014 6:46:44 PM UTC+3, Marek Hulan wrote:

Hello,

since CentOS is not supported, noone noticed that it stopped working
on

CentOS. I made a fix on both installer and staypuft side today. You’ll
have to
update and rerun the installer (0.3.2 should get to nightly tomorrow)
to

precreate staypuft OS v 7. Staypuft fix was not yet merged but you can
apply it
manually, see [1] for the patch.

Before creating new deploy you must also fix one issue with CentOS
installation
media. For some reason, CentOS 7.0 mirror does not follow standard
path so

as
a workaround you have to navigate to Hosts -> Installation Media ->
CentOS

Mirror and update path to
http://mirror.centos.org/centos/$major/os/$arch.

More information at [2].

Fixes should get in and be packaged in the beginning of next week. I’d
be

happy if you tested them and shared your results.

[1] https://github.com/theforeman/staypuft/pull/271
[2] Bug #6884: Centos mirror in "Installation Media" doesn't work for versions without $minor - Foreman

On Thursday 21 of August 2014 17:49:52 Marek Hulan wrote:

I was able to reproduce it. I’ll look have a look at it tomorrow and

will

let you know.

After downloading images I rerun installer and he told me that

everything

success:

Starting configuration…
Starting to seed provisioning data
Use ‘base_CentOS_6’ hostgroup for provisioning

Success!

But UI told me same "missing Operatingsystem"
and in foreman/production.log same error:

Started GET “/deployments/new” for 10.10.10.1 at 2014-08-21
17:14:12

+0300

Processing by Staypuft::DeploymentsController#new as HTML
Operation FAILED: missing Operatingsystem

Rendered common/500.html.erb within staypuft/layouts/staypuft

(3.8ms)

Rendered layouts/base.html.erb (0.9ms)
Rendered

/opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/

views/staypuft/layouts/application.html.erb (1.2ms)
Completed 500 Internal Server Error in 79ms (Views: 6.7ms |

ActiveRecord:

13.4ms)

On Thursday, August 21, 2014 5:01:15 PM UTC+3, Marek Hulan wrote:

There seems to be some ordering issue with discovery images,
target

directory
did not exist yet. When there appears an error during
installation,

configuration won’t continue. Your staypuft installation
resulted in

being
installed but not configured.

Now when you fixed discovery images issue, can you rerun
installer

once

more?
Until it shows that you can use some hostgroup for provisioning
you

can’t
consider staypuft to be ready. But if discovery images was the
only

issue
it
should pass now.

On Thursday 21 of August 2014 03:20:15 Nomnom Fuser wrote:

Ok. Yet another fresh install. Centos 6.5 x86_64
repos:

base

                     CentOS-6 - Base 

epel

                     Extra Packages for Enterprise Linux 6 
                     x86_64 

extras

                     CentOS-6 - Extras 

foreman

                    Foreman nightly 

foreman-plugins

                    Foreman plugins nightly 

scl

                    CentOS-6 - SCL 

updates

                    CentOS-6 - Updates 

rpm -qa | grep foreman

foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noarch

foreman-installer-staypuft-0.3.0-1.el6.noarch

foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noarch

rubygem-foreman_api-0.1.11-1.el6.noarch

foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch

First install with "staypuft-installer

–foreman-configure-epel-repo=true

–foreman-plugin-discovery-install-images=true" :

[ INFO 2014-08-21 13:06:14 main] Executing hooks in group post
[DEBUG 2014-08-21 13:06:14 main] Hook

/usr/share/foreman-installer/hooks/

post/10-setup_provisioning.rb returned false
[ INFO 2014-08-21 13:06:14 main] All hooks in group post
finished

[DEBUG 2014-08-21 13:06:14 main] Exit with status code: 6
(signal

was

[ERROR 2014-08-21 13:06:14 main] Repeating errors encountered

during

run:

[ERROR 2014-08-21 13:06:14 main]

/Stage[main]/Foreman::Plugin::Discovery/

Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-late

st
.

el6.iso-vmlinuz]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery->

image-latest.el6.iso-vmlinuz]/returns: change from notrun to 0

failed:

/usr/

bin/curl -s

http://downloads.theforeman.org/discovery/releases/latest/foreman-discov

er
y

-image-latest.el6.iso-vmlinuz -o

/var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-vmlinuz

returned 23 instead of one of [0] at

/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp:

6
[ERROR 2014-08-21 13:06:14 main]

/Stage[main]/Foreman::Plugin::Discovery/

Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-late

st
.

el6.iso-img]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery-imag

e-> > > latest.el6.iso-img]/returns: change from notrun to 0
failed:

/usr/bin/curl> >

-s

http://downloads.theforeman.org/discovery/releases/latest/foreman-discov

er
y

-image-latest.el6.iso-img -o

/var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-img

returned

23 instead of one of [0] at

/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp:

Ok, I download images manually using same curl command.
Second run with same options. Seems ok. “Success!”

Login to UI, go to new deployment of openstack,
aaaannnndddd…

Started GET “/deployments/new” for 10.10.10.1 at 2014-08-21

13:14:11

+0300

Processing by Staypuft::DeploymentsController#new as HTML
Operation FAILED: missing Operatingsystem

Rendered common/500.html.erb within
staypuft/layouts/staypuft

(7.9ms)
Rendered layouts/base.html.erb (2.0ms)
Rendered

/opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/

views/staypuft/layouts/application.html.erb (3.6ms)
Completed 500 Internal Server Error in 508ms (Views: 16.6ms |

ActiveRecord:

34.1ms)

after installer:

rpm -qa | grep foreman

foreman-installer-staypuft-0.3.0-1.el6.noarch
ruby193-rubygem-foreman-tasks-0.6.4-1.el6.noarch

foreman-postgresql-1.7.0-0.develop.201408201145git87a42cb.el6.noarch

foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noarch

rubygem-foreman_api-0.1.11-1.el6.noarch

foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch

foreman-1.7.0-0.develop.201408201145git87a42cb.el6.noarch

foreman-proxy-1.7.0-0.develop.201408191935git7124635.el6.noarch

foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noarch

ruby193-rubygem-foreman_discovery-1.3.0-1.el6.noarch

Seems like I need install openstack manually from source or
using

rdo

without provisioning %(

On Thursday, August 21, 2014 10:33:16 AM UTC+3, Marek Hulan > wrote:

Hello,

This error would indicate that installer did not finish all

steps

successfully.
Could you rerun the installer and try again?

Hello guys,

staypuft 0.3.0 and foreman-installer-staypuft 0.3.3 were uploaded to nightly
repos containing all CentOS repos up to date. There's still workaround needed
regarding installation media. I publish the fix for downloading discovery
images [1] but it will take some time before it gets in, so until that another
workaround is needed - precreate /var/lib/tftpboot/boot before running
installer for the first time (or rerun installer after first failed
installation).

If you update existing installation, be sure to rerun installer after yum
upgrade. Also feel free to stop by at #theforeman-dev channel on freenode and
share your experience.

··· -- Marek

[1] https://github.com/theforeman/puppet-foreman/pull/217

On Monday 25 of August 2014 10:39:42 Marek Hulan wrote:

Hi,

you’re right, it seems like some ordering issue in upstream puppet module.
I’ll try to look at that. Meanwhile the workaround is what you say, you can
use shorter variant

mkdir -p /var/lib/tftpboot/boot

Thanks for the report

Marek,

Thanks for the patch. I updated the file with your details, but the
installation was still failing during the image retrieval phase. I thought
that the issue was related to the files being downloaded before the
directory was created, so I created the directory and it worked! Here is
the script that I was using for the installation:

==Install==
yum install -y
http://yum.theforeman.org/nightly/el6/x86_64/foreman-release.rpm
yum -y localinstall
http://mirror.pnl.gov/epel/6/x86_64/epel-release-6-8.noarch.rpm
yum -y install centos-release-SCL
yum -y install foreman-plugin-staypuft
yum -y install foreman-installer-staypuft

mkdir /var/lib/tftpboot
mkdir /var/lib/tftpboot/boot

Now modify the file:

/opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/models/staypu
ft /deployment.rb based on your instructions and then:

staypuft-installer --foreman-configure-epel-repo=true
–foreman-plugin-discovery-install-images=true

== end ====

It appears that the module with problems is this one:
/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp

On Friday, August 22, 2014 12:26:55 PM UTC-4, Nomnom Fuser wrote:

Thx for patch. I’ll try it next week and share my results.

You refer to issue with CentOS 7 and I use Centos 6.5 as host and for
provisioning.
Did I understand correctly that new CentOS versioning (eg. 7.0.1234)
brakes some puppet scripts that runs with staypuft-installer?

On Friday, August 22, 2014 6:46:44 PM UTC+3, Marek Hulan wrote:

Hello,

since CentOS is not supported, noone noticed that it stopped working on
CentOS. I made a fix on both installer and staypuft side today. You’ll
have to
update and rerun the installer (0.3.2 should get to nightly tomorrow)
to
precreate staypuft OS v 7. Staypuft fix was not yet merged but you can
apply it
manually, see [1] for the patch.

Before creating new deploy you must also fix one issue with CentOS
installation
media. For some reason, CentOS 7.0 mirror does not follow standard path
so as
a workaround you have to navigate to Hosts -> Installation Media ->
CentOS
Mirror and update path to
http://mirror.centos.org/centos/$major/os/$arch.

More information at [2].

Fixes should get in and be packaged in the beginning of next week. I’d
be
happy if you tested them and shared your results.

[1] https://github.com/theforeman/staypuft/pull/271
[2] Bug #6884: Centos mirror in "Installation Media" doesn't work for versions without $minor - Foreman

On Thursday 21 of August 2014 17:49:52 Marek Hulan wrote:

I was able to reproduce it. I’ll look have a look at it tomorrow and

will

let you know.

After downloading images I rerun installer and he told me that

everything

success:

Starting configuration…
Starting to seed provisioning data
Use ‘base_CentOS_6’ hostgroup for provisioning

Success!

But UI told me same "missing Operatingsystem"
and in foreman/production.log same error:

Started GET “/deployments/new” for 10.10.10.1 at 2014-08-21
17:14:12

+0300

Processing by Staypuft::DeploymentsController#new as HTML
Operation FAILED: missing Operatingsystem

Rendered common/500.html.erb within staypuft/layouts/staypuft

(3.8ms)

Rendered layouts/base.html.erb (0.9ms)
Rendered

/opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/

views/staypuft/layouts/application.html.erb (1.2ms)
Completed 500 Internal Server Error in 79ms (Views: 6.7ms |

ActiveRecord:

13.4ms)

On Thursday, August 21, 2014 5:01:15 PM UTC+3, Marek Hulan wrote:

There seems to be some ordering issue with discovery images,
target
directory
did not exist yet. When there appears an error during
installation,
configuration won’t continue. Your staypuft installation resulted

in

being
installed but not configured.

Now when you fixed discovery images issue, can you rerun
installer

once

more?
Until it shows that you can use some hostgroup for provisioning
you
can’t
consider staypuft to be ready. But if discovery images was the
only
issue
it
should pass now.

On Thursday 21 of August 2014 03:20:15 Nomnom Fuser wrote:

Ok. Yet another fresh install. Centos 6.5 x86_64
repos:

base

                     CentOS-6 - Base

epel

                     Extra Packages for Enterprise Linux 6
                     -
                     x86_64

extras

                     CentOS-6 - Extras

foreman

                    Foreman nightly

foreman-plugins

                    Foreman plugins nightly

scl

                    CentOS-6 - SCL

updates

                    CentOS-6 - Updates

rpm -qa | grep foreman

foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noar
ch
foreman-installer-staypuft-0.3.0-1.el6.noarch
foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noar
ch
rubygem-foreman_api-0.1.11-1.el6.noarch

foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch

First install with "staypuft-installer

–foreman-configure-epel-repo=true

–foreman-plugin-discovery-install-images=true" :

[ INFO 2014-08-21 13:06:14 main] Executing hooks in group post
[DEBUG 2014-08-21 13:06:14 main] Hook

/usr/share/foreman-installer/hooks/

post/10-setup_provisioning.rb returned false
[ INFO 2014-08-21 13:06:14 main] All hooks in group post
finished
[DEBUG 2014-08-21 13:06:14 main] Exit with status code: 6
(signal

was

[ERROR 2014-08-21 13:06:14 main] Repeating errors encountered

during

run:

[ERROR 2014-08-21 13:06:14 main]

/Stage[main]/Foreman::Plugin::Discovery/

Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-lat
e

st
.

el6.iso-vmlinuz]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery
->

image-latest.el6.iso-vmlinuz]/returns: change from notrun to 0

failed:

/usr/

bin/curl -s

http://downloads.theforeman.org/discovery/releases/latest/foreman-disco
v

er
y

-image-latest.el6.iso-vmlinuz -o

/var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-vmlinuz

returned 23 instead of one of [0] at

/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp:

6
[ERROR 2014-08-21 13:06:14 main]

/Stage[main]/Foreman::Plugin::Discovery/

Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-lat
e

st
.

el6.iso-img]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery-ima
g

e-> > > latest.el6.iso-img]/returns: change from notrun to 0

failed:

/usr/bin/curl> >

-s

http://downloads.theforeman.org/discovery/releases/latest/foreman-disco
v

er
y

-image-latest.el6.iso-img -o
/var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-i
mg

returned

23 instead of one of [0] at

/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp:

Ok, I download images manually using same curl command.
Second run with same options. Seems ok. “Success!”

Login to UI, go to new deployment of openstack,
aaaannnndddd…

Started GET “/deployments/new” for 10.10.10.1 at 2014-08-21

13:14:11

+0300

Processing by Staypuft::DeploymentsController#new as HTML
Operation FAILED: missing Operatingsystem

Rendered common/500.html.erb within staypuft/layouts/staypuft
(7.9ms)
Rendered layouts/base.html.erb (2.0ms)
Rendered
/opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/

views/staypuft/layouts/application.html.erb (3.6ms)
Completed 500 Internal Server Error in 508ms (Views: 16.6ms |

ActiveRecord:

34.1ms)

after installer:

rpm -qa | grep foreman

foreman-installer-staypuft-0.3.0-1.el6.noarch
ruby193-rubygem-foreman-tasks-0.6.4-1.el6.noarch

foreman-postgresql-1.7.0-0.develop.201408201145git87a42cb.el6.noarch

foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noar
ch
rubygem-foreman_api-0.1.11-1.el6.noarch

foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch

foreman-1.7.0-0.develop.201408201145git87a42cb.el6.noarch
foreman-proxy-1.7.0-0.develop.201408191935git7124635.el6.noarch
foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noar
ch
ruby193-rubygem-foreman_discovery-1.3.0-1.el6.noarch

Seems like I need install openstack manually from source or
using

rdo

without provisioning %(

On Thursday, August 21, 2014 10:33:16 AM UTC+3, Marek Hulan > > >> > > >> wrote:

Hello,

This error would indicate that installer did not finish all

steps

successfully.
Could you rerun the installer and try again?


Marek

I'm afraid using 6.5 for hosts is not a good idea, definitely not for
production. AFAIK noone tested it and it's not even an aim to support it so I
guess it will break tomorrow if not today. From staypuft point of view, I know
that there's at least one issue that you'll encounter. It could be probably
fixed but I think there will be other problems. I'm sorry to disappoint you but
I rather invest time to properly support centos 7.

··· -- Marek

On Tuesday 26 of August 2014 07:42:11 Nomnom Fuser wrote:

On Monday, August 25, 2014 10:46:48 AM UTC+3, Marek Hulan wrote:

Hello

You refer to issue with CentOS 7 and I use Centos 6.5 as host and for
provisioning.

It’s a bit confusing but we require CentOS 6.5 for provisioning host
(where we
install Foreman) but we provision CetnOS 7 on other machines on which you
deploy OpenStack.

Ah. I see. Is it possible to use CentOS 6.5 for host node AND openstack
deployment?
CentOS 7 is so new that I’m afraid install it on prod right now + I don’t
know is Ceph storage already compatible with CentOS 7

Did I understand correctly that new CentOS versioning (eg. 7.0.1234)

brakes

some puppet scripts that runs with staypuft-installer?

More precisely, it breaks default foreman installation media which are
used
during provisioning. The revision number in CentOS path is not standard,
previously they had only $major.$minor. As a workaround staypuft-installer
could modify it but I’d like to see the fix for upstream foreman first,
there
should be some better way. Or CentOS guys will fix their mirrors :slight_smile:

I think that CenOS guys won’t fix that :slight_smile: . Seems that they decided to use
new version numbering.
For now I use dirty hack with mirror url
like http://mirror.centos.org/centos/7/os/$arch

Anyway I built installer 0.3.3 today which contains synxtax error fix, so
if
testing, use this (should get into nightly repo shortly). I’ll try to get
staypuft fix in today as well.

On Friday 22 of August 2014 09:26:55 Nomnom Fuser wrote:

Thx for patch. I’ll try it next week and share my results.

On Friday, August 22, 2014 6:46:44 PM UTC+3, Marek Hulan wrote:

Hello,

since CentOS is not supported, noone noticed that it stopped working

on

CentOS. I made a fix on both installer and staypuft side today. You’ll
have to
update and rerun the installer (0.3.2 should get to nightly tomorrow)

to

precreate staypuft OS v 7. Staypuft fix was not yet merged but you can
apply it
manually, see [1] for the patch.

Before creating new deploy you must also fix one issue with CentOS
installation
media. For some reason, CentOS 7.0 mirror does not follow standard

path so

as
a workaround you have to navigate to Hosts -> Installation Media ->

CentOS

Mirror and update path to

http://mirror.centos.org/centos/$major/os/$arch.

More information at [2].

Fixes should get in and be packaged in the beginning of next week. I’d

be

happy if you tested them and shared your results.

[1] https://github.com/theforeman/staypuft/pull/271
[2] Bug #6884: Centos mirror in "Installation Media" doesn't work for versions without $minor - Foreman

On Thursday 21 of August 2014 17:49:52 Marek Hulan wrote:

I was able to reproduce it. I’ll look have a look at it tomorrow and

will

let you know.

After downloading images I rerun installer and he told me that

everything

success:

Starting configuration…
Starting to seed provisioning data
Use ‘base_CentOS_6’ hostgroup for provisioning

Success!

/var/log/foreman-installer/foreman-installer.log

But UI told me same "missing Operatingsystem"
and in foreman/production.log same error:

Started GET “/deployments/new” for 10.10.10.1 at 2014-08-21

17:14:12

+0300

Processing by Staypuft::DeploymentsController#new as HTML
Operation FAILED: missing Operatingsystem

Rendered common/500.html.erb within staypuft/layouts/staypuft

(3.8ms)

Rendered layouts/base.html.erb (0.9ms)
Rendered

/opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/

views/staypuft/layouts/application.html.erb (1.2ms)
Completed 500 Internal Server Error in 79ms (Views: 6.7ms |

ActiveRecord:

13.4ms)

On Thursday, August 21, 2014 5:01:15 PM UTC+3, Marek Hulan wrote:

There seems to be some ordering issue with discovery images,

target

directory
did not exist yet. When there appears an error during

installation,

configuration won’t continue. Your staypuft installation

resulted in

being
installed but not configured.

Now when you fixed discovery images issue, can you rerun

installer

once

more?
Until it shows that you can use some hostgroup for provisioning

you

can’t
consider staypuft to be ready. But if discovery images was the

only

issue
it
should pass now.

On Thursday 21 of August 2014 03:20:15 Nomnom Fuser wrote:

Ok. Yet another fresh install. Centos 6.5 x86_64
repos:

base

                     CentOS-6 - Base

epel

                     Extra Packages for Enterprise Linux 6
                     x86_64

extras

                     CentOS-6 - Extras

foreman

                    Foreman nightly

foreman-plugins

                    Foreman plugins nightly

scl

                    CentOS-6 - SCL

updates

                    CentOS-6 - Updates

rpm -qa | grep foreman

foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noarch

foreman-installer-staypuft-0.3.0-1.el6.noarch

foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noarch

rubygem-foreman_api-0.1.11-1.el6.noarch

foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch

First install with "staypuft-installer

–foreman-configure-epel-repo=true

–foreman-plugin-discovery-install-images=true" :

[ INFO 2014-08-21 13:06:14 main] Executing hooks in group post
[DEBUG 2014-08-21 13:06:14 main] Hook

/usr/share/foreman-installer/hooks/

post/10-setup_provisioning.rb returned false
[ INFO 2014-08-21 13:06:14 main] All hooks in group post

finished

[DEBUG 2014-08-21 13:06:14 main] Exit with status code: 6

(signal

was

[ERROR 2014-08-21 13:06:14 main] Repeating errors encountered

during

run:

[ERROR 2014-08-21 13:06:14 main]

/Stage[main]/Foreman::Plugin::Discovery/

Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-late

st
.

el6.iso-vmlinuz]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery->

image-latest.el6.iso-vmlinuz]/returns: change from notrun to 0

failed:

/usr/

bin/curl -s

http://downloads.theforeman.org/discovery/releases/latest/foreman-discov

er
y

-image-latest.el6.iso-vmlinuz -o

/var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-vmlinuz

returned 23 instead of one of [0] at

/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp:

6
[ERROR 2014-08-21 13:06:14 main]

/Stage[main]/Foreman::Plugin::Discovery/

Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-late

st
.

el6.iso-img]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery-imag

e-> > > latest.el6.iso-img]/returns: change from notrun to 0

failed:

/usr/bin/curl> >

-s

http://downloads.theforeman.org/discovery/releases/latest/foreman-discov

er
y

-image-latest.el6.iso-img -o

/var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-img

returned

23 instead of one of [0] at

/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp:

Ok, I download images manually using same curl command.
Second run with same options. Seems ok. “Success!”

Login to UI, go to new deployment of openstack,

aaaannnndddd…

Started GET “/deployments/new” for 10.10.10.1 at 2014-08-21

13:14:11

+0300

Processing by Staypuft::DeploymentsController#new as HTML
Operation FAILED: missing Operatingsystem

Rendered common/500.html.erb within

staypuft/layouts/staypuft

(7.9ms)
Rendered layouts/base.html.erb (2.0ms)
Rendered

/opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/

views/staypuft/layouts/application.html.erb (3.6ms)
Completed 500 Internal Server Error in 508ms (Views: 16.6ms |

ActiveRecord:

34.1ms)

after installer:

rpm -qa | grep foreman

foreman-installer-staypuft-0.3.0-1.el6.noarch
ruby193-rubygem-foreman-tasks-0.6.4-1.el6.noarch

foreman-postgresql-1.7.0-0.develop.201408201145git87a42cb.el6.noarch

foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noarch

rubygem-foreman_api-0.1.11-1.el6.noarch

foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch

foreman-1.7.0-0.develop.201408201145git87a42cb.el6.noarch

foreman-proxy-1.7.0-0.develop.201408191935git7124635.el6.noarch

foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noarch

ruby193-rubygem-foreman_discovery-1.3.0-1.el6.noarch

Seems like I need install openstack manually from source or

using

rdo

without provisioning %(

On Thursday, August 21, 2014 10:33:16 AM UTC+3, Marek Hulan > > > > wrote:

Hello,

This error would indicate that installer did not finish all

steps

successfully.
Could you rerun the installer and try again?


Marek

For now it seems like foreman-installer-staypuft-0.3.3-1

  • foreman-1.7.0-0.develop.201408201145git87a42cb work fine. at least now I
    can create deployments.

>
> I'm afraid using 6.5 for hosts is not a good idea, definitely not for
> production. AFAIK noone tested it and it's not even an aim to support it
> so I
> guess it will break tomorrow if not today. From staypuft point of view, I
> know
> that there's at least one issue that you'll encounter. It could be
> probably
> fixed but I think there will be other problems. I'm sorry to disappoint
> you but
> I rather invest time to properly support centos 7.
>

It's fine. Thanks for such great plugin!

··· On Tuesday, August 26, 2014 7:27:41 PM UTC+3, Marek Hulan wrote:


Marek

On Tuesday 26 of August 2014 07:42:11 Nomnom Fuser wrote:

On Monday, August 25, 2014 10:46:48 AM UTC+3, Marek Hulan wrote:

Hello

You refer to issue with CentOS 7 and I use Centos 6.5 as host and
for

provisioning.

It’s a bit confusing but we require CentOS 6.5 for provisioning host
(where we
install Foreman) but we provision CetnOS 7 on other machines on which
you

deploy OpenStack.

Ah. I see. Is it possible to use CentOS 6.5 for host node AND openstack
deployment?
CentOS 7 is so new that I’m afraid install it on prod right now + I
don’t
know is Ceph storage already compatible with CentOS 7

Did I understand correctly that new CentOS versioning (eg. 7.0.1234)

brakes

some puppet scripts that runs with staypuft-installer?

More precisely, it breaks default foreman installation media which are
used
during provisioning. The revision number in CentOS path is not
standard,

previously they had only $major.$minor. As a workaround
staypuft-installer

could modify it but I’d like to see the fix for upstream foreman
first,

there
should be some better way. Or CentOS guys will fix their mirrors :slight_smile:

I think that CenOS guys won’t fix that :slight_smile: . Seems that they decided to
use
new version numbering.
For now I use dirty hack with mirror url
like http://mirror.centos.org/centos/7/os/$arch

Anyway I built installer 0.3.3 today which contains synxtax error fix,
so

if
testing, use this (should get into nightly repo shortly). I’ll try to
get

staypuft fix in today as well.

On Friday 22 of August 2014 09:26:55 Nomnom Fuser wrote:

Thx for patch. I’ll try it next week and share my results.

On Friday, August 22, 2014 6:46:44 PM UTC+3, Marek Hulan wrote:

Hello,

since CentOS is not supported, noone noticed that it stopped
working

on

CentOS. I made a fix on both installer and staypuft side today.
You’ll

have to
update and rerun the installer (0.3.2 should get to nightly
tomorrow)

to

precreate staypuft OS v 7. Staypuft fix was not yet merged but you
can

apply it
manually, see [1] for the patch.

Before creating new deploy you must also fix one issue with CentOS
installation
media. For some reason, CentOS 7.0 mirror does not follow standard

path so

as
a workaround you have to navigate to Hosts -> Installation Media
->

CentOS

Mirror and update path to

http://mirror.centos.org/centos/$major/os/$arch.

More information at [2].

Fixes should get in and be packaged in the beginning of next week.
I’d

be

happy if you tested them and shared your results.

[1] https://github.com/theforeman/staypuft/pull/271
[2] Bug #6884: Centos mirror in "Installation Media" doesn't work for versions without $minor - Foreman

On Thursday 21 of August 2014 17:49:52 Marek Hulan wrote:

I was able to reproduce it. I’ll look have a look at it tomorrow
and

will

let you know.

After downloading images I rerun installer and he told me that

everything

success:

Starting configuration…
Starting to seed provisioning data
Use ‘base_CentOS_6’ hostgroup for provisioning

Success!

  • Puppetmaster is running at port 8140
    The full log is at

/var/log/foreman-installer/foreman-installer.log

But UI told me same "missing Operatingsystem"
and in foreman/production.log same error:

Started GET “/deployments/new” for 10.10.10.1 at 2014-08-21

17:14:12

+0300

Processing by Staypuft::DeploymentsController#new as HTML
Operation FAILED: missing Operatingsystem

Rendered common/500.html.erb within
staypuft/layouts/staypuft

(3.8ms)

Rendered layouts/base.html.erb (0.9ms)
Rendered

/opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/

views/staypuft/layouts/application.html.erb (1.2ms)
Completed 500 Internal Server Error in 79ms (Views: 6.7ms |

ActiveRecord:

13.4ms)

On Thursday, August 21, 2014 5:01:15 PM UTC+3, Marek Hulan > wrote:

There seems to be some ordering issue with discovery images,

target

directory
did not exist yet. When there appears an error during

installation,

configuration won’t continue. Your staypuft installation

resulted in

being
installed but not configured.

Now when you fixed discovery images issue, can you rerun

installer

once

more?
Until it shows that you can use some hostgroup for
provisioning

you

can’t
consider staypuft to be ready. But if discovery images was
the

only

issue
it
should pass now.

On Thursday 21 of August 2014 03:20:15 Nomnom Fuser wrote:

Ok. Yet another fresh install. Centos 6.5 x86_64
repos:

base

                     CentOS-6 - Base 

epel

                     Extra Packages for Enterprise 

Linux 6

                     x86_64 

extras

                     CentOS-6 - Extras 

foreman

                    Foreman nightly 

foreman-plugins

                    Foreman plugins nightly 

scl

                    CentOS-6 - SCL 

updates

                    CentOS-6 - Updates 

rpm -qa | grep foreman

foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noarch

foreman-installer-staypuft-0.3.0-1.el6.noarch

foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noarch

rubygem-foreman_api-0.1.11-1.el6.noarch

foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch

First install with "staypuft-installer

–foreman-configure-epel-repo=true

–foreman-plugin-discovery-install-images=true" :

[ INFO 2014-08-21 13:06:14 main] Executing hooks in group
post

[DEBUG 2014-08-21 13:06:14 main] Hook

/usr/share/foreman-installer/hooks/

post/10-setup_provisioning.rb returned false
[ INFO 2014-08-21 13:06:14 main] All hooks in group post

finished

[DEBUG 2014-08-21 13:06:14 main] Exit with status code: 6

(signal

was

[ERROR 2014-08-21 13:06:14 main] Repeating errors
encountered

during

run:

[ERROR 2014-08-21 13:06:14 main]

/Stage[main]/Foreman::Plugin::Discovery/

Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-late

st
.

el6.iso-vmlinuz]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery->

image-latest.el6.iso-vmlinuz]/returns: change from notrun to
0

failed:

/usr/

bin/curl -s

http://downloads.theforeman.org/discovery/releases/latest/foreman-discov

er
y

-image-latest.el6.iso-vmlinuz -o

/var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-vmlinuz

returned 23 instead of one of [0] at

/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp:

6
[ERROR 2014-08-21 13:06:14 main]

/Stage[main]/Foreman::Plugin::Discovery/

Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-late

st
.

el6.iso-img]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery-imag

e-> > > latest.el6.iso-img]/returns: change from notrun to 0

failed:

/usr/bin/curl> >

-s

http://downloads.theforeman.org/discovery/releases/latest/foreman-discov

er
y

-image-latest.el6.iso-img -o

/var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-img

returned

23 instead of one of [0] at

/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp:

Ok, I download images manually using same curl command.
Second run with same options. Seems ok. “Success!”

Login to UI, go to new deployment of openstack,

aaaannnndddd…

Started GET “/deployments/new” for 10.10.10.1 at
2014-08-21

13:14:11

+0300

Processing by Staypuft::DeploymentsController#new as HTML
Operation FAILED: missing Operatingsystem

Rendered common/500.html.erb within

staypuft/layouts/staypuft

(7.9ms)
Rendered layouts/base.html.erb (2.0ms)
Rendered

/opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/

views/staypuft/layouts/application.html.erb (3.6ms)
Completed 500 Internal Server Error in 508ms (Views:
16.6ms |

ActiveRecord:

34.1ms)

after installer:

rpm -qa | grep foreman

foreman-installer-staypuft-0.3.0-1.el6.noarch
ruby193-rubygem-foreman-tasks-0.6.4-1.el6.noarch

foreman-postgresql-1.7.0-0.develop.201408201145git87a42cb.el6.noarch

foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noarch

rubygem-foreman_api-0.1.11-1.el6.noarch

foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch

foreman-1.7.0-0.develop.201408201145git87a42cb.el6.noarch

foreman-proxy-1.7.0-0.develop.201408191935git7124635.el6.noarch

foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noarch

ruby193-rubygem-foreman_discovery-1.3.0-1.el6.noarch

Seems like I need install openstack manually from source
or

using

rdo

without provisioning %(

On Thursday, August 21, 2014 10:33:16 AM UTC+3, Marek > Hulan > > > > > > wrote:

Hello,

This error would indicate that installer did not finish
all

steps

successfully.
Could you rerun the installer and try again?


Marek

Marek,

Can you please create a new staypuft RPM version from the github
repositories that includes the latest change to incorporate 1.4.0.rc1 as
the new required version for foreman-discovery plugin vs the 1.3.0.rc1
that is in included in the latest rpm.

I was able to trace the error in the installation back to that change
but my technical skills are not yet there to deploy the github version
into my machine. :wink:

https://github.com/theforeman/staypuft/commit/87ceae281a3ce864030bbd608c75c76a2bb65124#diff-32c0f0d9976a9b49506d42ab8f4e0520

Thanks.

··· On 08/26/2014 03:49 AM, Marek Hulan wrote: > Hello guys, > > staypuft 0.3.0 and foreman-installer-staypuft 0.3.3 were uploaded to nightly > repos containing all CentOS repos up to date. There's still workaround needed > regarding installation media. I publish the fix for downloading discovery > images [1] but it will take some time before it gets in, so until that another > workaround is needed - precreate /var/lib/tftpboot/boot before running > installer for the first time (or rerun installer after first failed > installation). > > If you update existing installation, be sure to rerun installer after yum > upgrade. Also feel free to stop by at #theforeman-dev channel on freenode and > share your experience. > > -- > Marek > > [1] https://github.com/theforeman/puppet-foreman/pull/217 > > > On Monday 25 of August 2014 10:39:42 Marek Hulan wrote: >> Hi, >> >> you're right, it seems like some ordering issue in upstream puppet module. >> I'll try to look at that. Meanwhile the workaround is what you say, you can >> use shorter variant >> >> mkdir -p /var/lib/tftpboot/boot >> >> Thanks for the report >> >>> Marek, >>> >>> Thanks for the patch. I updated the file with your details, but the >>> installation was still failing during the image retrieval phase. I thought >>> that the issue was related to the files being downloaded before the >>> directory was created, so I created the directory and it worked! Here is >>> the script that I was using for the installation: >>> >>> ==Install== >>> yum install -y >>> http://yum.theforeman.org/nightly/el6/x86_64/foreman-release.rpm >>> yum -y localinstall >>> http://mirror.pnl.gov/epel/6/x86_64/epel-release-6-8.noarch.rpm >>> yum -y install centos-release-SCL >>> yum -y install foreman-plugin-staypuft >>> yum -y install foreman-installer-staypuft >>> >>> mkdir /var/lib/tftpboot >>> mkdir /var/lib/tftpboot/boot >>> >>> # Now modify the file: >>> /opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/models/staypu >>> ft /deployment.rb based on your instructions and then: >>> >>> staypuft-installer --foreman-configure-epel-repo=true >>> --foreman-plugin-discovery-install-images=true >>> >>> == end ==== >>> >>> It appears that the module with problems is this one: >>> /usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp >>> >>> On Friday, August 22, 2014 12:26:55 PM UTC-4, Nomnom Fuser wrote: >>>> Thx for patch. I'll try it next week and share my results. >>>> >>>> You refer to issue with CentOS 7 and I use Centos 6.5 as host and for >>>> provisioning. >>>> Did I understand correctly that new CentOS versioning (eg. 7.0.1234) >>>> brakes some puppet scripts that runs with staypuft-installer? >>>> >>>> On Friday, August 22, 2014 6:46:44 PM UTC+3, Marek Hulan wrote: >>>>> Hello, >>>>> >>>>> since CentOS is not supported, noone noticed that it stopped working on >>>>> CentOS. I made a fix on both installer and staypuft side today. You'll >>>>> have to >>>>> update and rerun the installer (0.3.2 should get to nightly tomorrow) >>>>> to >>>>> precreate staypuft OS v 7. Staypuft fix was not yet merged but you can >>>>> apply it >>>>> manually, see [1] for the patch. >>>>> >>>>> Before creating new deploy you must also fix one issue with CentOS >>>>> installation >>>>> media. For some reason, CentOS 7.0 mirror does not follow standard path >>>>> so as >>>>> a workaround you have to navigate to Hosts -> Installation Media -> >>>>> CentOS >>>>> Mirror and update path to >>>>> http://mirror.centos.org/centos/$major/os/$arch. >>>>> >>>>> More information at [2]. >>>>> >>>>> Fixes should get in and be packaged in the beginning of next week. I'd >>>>> be >>>>> happy if you tested them and shared your results. >>>>> >>>>> [1] https://github.com/theforeman/staypuft/pull/271 >>>>> [2] http://projects.theforeman.org/issues/6884 >>>>> >>>>> On Thursday 21 of August 2014 17:49:52 Marek Hulan wrote: >>>>>> I was able to reproduce it. I'll look have a look at it tomorrow and >>>>> will >>>>> >>>>>> let you know. >>>>>> >>>>>>> After downloading images I rerun installer and he told me that >>>>> everything >>>>> >>>>>>> success: >>>>>>> >>>>>>> Starting configuration... >>>>>>> Starting to seed provisioning data >>>>>>> Use 'base_CentOS_6' hostgroup for provisioning >>>>>>> >>>>>>> Success! >>>>>>> * Foreman is running at https://openstack-core.cloud >>>>>>> >>>>>>> Initial credentials are admin / df4Rauujuwxdede6 >>>>>>> >>>>>>> * Foreman Proxy is running at https://openstack-core.cloud:8443 >>>>>>> * Puppetmaster is running at port 8140 >>>>>>> The full log is at >>>>>>> /var/log/foreman-installer/foreman-installer.log >>>>>>> >>>>>>> But UI told me same "missing Operatingsystem" >>>>>>> and in foreman/production.log same error: >>>>>>> >>>>>>> Started GET "/deployments/new" for 10.10.10.1 at 2014-08-21 >>>>>>> 17:14:12 >>>>> +0300 >>>>> >>>>>>> Processing by Staypuft::DeploymentsController#new as HTML >>>>>>> Operation FAILED: missing Operatingsystem >>>>>>> >>>>>>> Rendered common/500.html.erb within staypuft/layouts/staypuft >>>>> (3.8ms) >>>>> >>>>>>> Rendered layouts/base.html.erb (0.9ms) >>>>>>> Rendered >>>>> /opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/ >>>>> >>>>>>> views/staypuft/layouts/application.html.erb (1.2ms) >>>>>>> Completed 500 Internal Server Error in 79ms (Views: 6.7ms | >>>>> ActiveRecord: >>>>>>> 13.4ms) >>>>>>> >>>>>>> On Thursday, August 21, 2014 5:01:15 PM UTC+3, Marek Hulan wrote: >>>>>>>> There seems to be some ordering issue with discovery images, >>>>>>>> target >>>>>>>> directory >>>>>>>> did not exist yet. When there appears an error during >>>>>>>> installation, >>>>>>>> configuration won't continue. Your staypuft installation resulted >>>>> in >>>>> >>>>>>>> being >>>>>>>> installed but not configured. >>>>>>>> >>>>>>>> Now when you fixed discovery images issue, can you rerun >>>>>>>> installer >>>>> once >>>>> >>>>>>>> more? >>>>>>>> Until it shows that you can use some hostgroup for provisioning >>>>>>>> you >>>>>>>> can't >>>>>>>> consider staypuft to be ready. But if discovery images was the >>>>>>>> only >>>>>>>> issue >>>>>>>> it >>>>>>>> should pass now. >>>>>>>> >>>>>>>> On Thursday 21 of August 2014 03:20:15 Nomnom Fuser wrote: >>>>>>>>> Ok. Yet another fresh install. Centos 6.5 x86_64 >>>>>>>>> repos: >>>>>>>>> >>>>>>>>> >>>>>>>>> base >>>>>>>>> >>>>>>>>> CentOS-6 - Base >>>>>>>>> >>>>>>>>> epel >>>>>>>>> >>>>>>>>> Extra Packages for Enterprise Linux 6 >>>>>>>>> - >>>>>>>>> x86_64 >>>>>>>>> >>>>>>>>> extras >>>>>>>>> >>>>>>>>> CentOS-6 - Extras >>>>>>>>> >>>>>>>>> foreman >>>>>>>>> >>>>>>>>> Foreman nightly >>>>>>>>> >>>>>>>>> foreman-plugins >>>>>>>>> >>>>>>>>> Foreman plugins nightly >>>>>>>>> >>>>>>>>> scl >>>>>>>>> >>>>>>>>> CentOS-6 - SCL >>>>>>>>> >>>>>>>>> updates >>>>>>>>> >>>>>>>>> CentOS-6 - Updates >>>>>>>>> >>>>>>>>> # rpm -qa | grep foreman >>>>>>>>> foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noar >>>>>>>>> ch >>>>>>>>> foreman-installer-staypuft-0.3.0-1.el6.noarch >>>>>>>>> foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noar >>>>>>>>> ch >>>>>>>>> rubygem-foreman_api-0.1.11-1.el6.noarch >>>>> foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch >>>>> >>>>>>>>> First install with "staypuft-installer >>>>>>>> --foreman-configure-epel-repo=true >>>>>>>> >>>>>>>>> --foreman-plugin-discovery-install-images=true" : >>>>>>>>> >>>>>>>>> [ INFO 2014-08-21 13:06:14 main] Executing hooks in group post >>>>>>>>> [DEBUG 2014-08-21 13:06:14 main] Hook >>>>>>>> /usr/share/foreman-installer/hooks/ >>>>>>>> >>>>>>>>> post/10-setup_provisioning.rb returned false >>>>>>>>> [ INFO 2014-08-21 13:06:14 main] All hooks in group post >>>>>>>>> finished >>>>>>>>> [DEBUG 2014-08-21 13:06:14 main] Exit with status code: 6 >>>>>>>>> (signal >>>>> was >>>>> >>>>>>>>> 6) >>>>>>>>> [ERROR 2014-08-21 13:06:14 main] Repeating errors encountered >>>>> during >>>>> >>>>>>>> run: >>>>>>>>> [ERROR 2014-08-21 13:06:14 main] >>>>>>>> >>>>>>>> /Stage[main]/Foreman::Plugin::Discovery/ >>>>> Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-lat >>>>> e >>>>> >>>>>>>> st >>>>>>>> . >>>>> el6.iso-vmlinuz]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery >>>>> -> >>>>> >>>>>>>> image-latest.el6.iso-vmlinuz]/returns: change from notrun to 0 >>>>> failed: >>>>>>>> /usr/ >>>>>>>> >>>>>>>>> bin/curl -s >>>>> http://downloads.theforeman.org/discovery/releases/latest/foreman-disco >>>>> v >>>>> >>>>>>>> er >>>>>>>> y >>>>>>>> >>>>>>>>> -image-latest.el6.iso-vmlinuz -o >>>>> /var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-vmlinuz >>>>> >>>>>>>>> returned 23 instead of one of [0] at >>>>> /usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp: >>>>>>>>> 6 >>>>>>>>> [ERROR 2014-08-21 13:06:14 main] >>>>>>>> >>>>>>>> /Stage[main]/Foreman::Plugin::Discovery/ >>>>> Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-lat >>>>> e >>>>> >>>>>>>> st >>>>>>>> . >>>>> el6.iso-img]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery-ima >>>>> g >>>>> >>>>>>>> e-> > > latest.el6.iso-img]/returns: change from notrun to 0 >>>>> failed: >>>>>>>> /usr/bin/curl> > >>>>>>>> >>>>>>>>> -s >>>>> http://downloads.theforeman.org/discovery/releases/latest/foreman-disco >>>>> v >>>>> >>>>>>>> er >>>>>>>> y >>>>>>>> >>>>>>>>> -image-latest.el6.iso-img -o >>>>>>>>> /var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-i >>>>>>>>> mg >>>>>>>> returned >>>>>>>> >>>>>>>>> 23 instead of one of [0] at >>>>> /usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp: >>>>>>>>> Ok, I download images manually using same curl command. >>>>>>>>> Second run with same options. Seems ok. "Success!" >>>>>>>>> >>>>>>>>> Login to UI, go to new deployment of openstack, >>>>>>>>> aaaannnndddd..... >>>>>>>>> >>>>>>>>> Started GET "/deployments/new" for 10.10.10.1 at 2014-08-21 >>>>> 13:14:11 >>>>> >>>>>>>> +0300 >>>>>>>> >>>>>>>>> Processing by Staypuft::DeploymentsController#new as HTML >>>>>>>>> Operation FAILED: missing Operatingsystem >>>>>>>>> >>>>>>>>> Rendered common/500.html.erb within staypuft/layouts/staypuft >>>>>>>>> (7.9ms) >>>>>>>>> Rendered layouts/base.html.erb (2.0ms) >>>>>>>>> Rendered >>>>>>>>> /opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/ >>>>>>>>> >>>>>>>>> views/staypuft/layouts/application.html.erb (3.6ms) >>>>>>>>> Completed 500 Internal Server Error in 508ms (Views: 16.6ms | >>>>>>>> ActiveRecord: >>>>>>>>> 34.1ms) >>>>>>>>> >>>>>>>>> after installer: >>>>>>>>> >>>>>>>>> # rpm -qa | grep foreman >>>>>>>>> foreman-installer-staypuft-0.3.0-1.el6.noarch >>>>>>>>> ruby193-rubygem-foreman-tasks-0.6.4-1.el6.noarch >>>>> foreman-postgresql-1.7.0-0.develop.201408201145git87a42cb.el6.noarch >>>>> >>>>>>>>> foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noar >>>>>>>>> ch >>>>>>>>> rubygem-foreman_api-0.1.11-1.el6.noarch >>>>> foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch >>>>> >>>>>>>>> foreman-1.7.0-0.develop.201408201145git87a42cb.el6.noarch >>>>>>>>> foreman-proxy-1.7.0-0.develop.201408191935git7124635.el6.noarch >>>>>>>>> foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noar >>>>>>>>> ch >>>>>>>>> ruby193-rubygem-foreman_discovery-1.3.0-1.el6.noarch >>>>>>>>> >>>>>>>>> Seems like I need install openstack manually from source or >>>>>>>>> using >>>>> rdo >>>>> >>>>>>>>> without provisioning %( >>>>>>>>> >>>>>>>>> On Thursday, August 21, 2014 10:33:16 AM UTC+3, Marek Hulan >>>>> wrote: >>>>>>>>>> Hello, >>>>>>>>>> >>>>>>>>>> This error would indicate that installer did not finish all >>>>> steps >>>>> >>>>>>>>>> successfully. >>>>>>>>>> Could you rerun the installer and try again?

The new version does allow for the creation of deployments. Great!

Now the problem arises when assigning the functions to the discovered nodes.
a. After assigning them, the nodes are not rebooted out of the ovirt image
and into the new CentOS 7 image. I need to manually reboot the nodes for
them to take the new image. That is not much of a problem but more of a
feature that was included in the discovered hosts before.

b. After loading the CentOS image and booting up, the nodes finish with an
error:

anaconda 19.31.79-1 for CentOS 7 started.
Running pre-installation scripts
The following problem occurred on line 0 of the kickstart file:

Unable to open input kickstart file: curl#37 - "Could't open file
/tmp/diskpart.cfg"

If I open a shell on the node, on the /tmp directory there are
ks-script-JpfOS3 and ks-script-JpfOS3.log. The first one reads:

#Dynamic
zerombr
clearpart --all --initlabel

the second one reads:

line 2: zerombr: command not found
line 3: clearpart: command not found

BTW, I am installing all into a VirtualBox interfase

··· On Wednesday, August 27, 2014 4:49:33 AM UTC-4, Nomnom Fuser wrote: > > For now it seems like foreman-installer-staypuft-0.3.3-1 > + foreman-1.7.0-0.develop.201408201145git87a42cb work fine. at least now I > can create deployments. > > On Tuesday, August 26, 2014 7:27:41 PM UTC+3, Marek Hulan wrote: >> >> I'm afraid using 6.5 for hosts is not a good idea, definitely not for >> production. AFAIK noone tested it and it's not even an aim to support it >> so I >> guess it will break tomorrow if not today. From staypuft point of view, I >> know >> that there's at least one issue that you'll encounter. It could be >> probably >> fixed but I think there will be other problems. I'm sorry to disappoint >> you but >> I rather invest time to properly support centos 7. >> > > It's fine. Thanks for such great plugin! > > >> >> -- >> Marek >> >> On Tuesday 26 of August 2014 07:42:11 Nomnom Fuser wrote: >> > On Monday, August 25, 2014 10:46:48 AM UTC+3, Marek Hulan wrote: >> > > Hello >> > > >> > > > You refer to issue with CentOS 7 and I use Centos 6.5 as host and >> for >> > > > provisioning. >> > > >> > > It's a bit confusing but we require CentOS 6.5 for provisioning host >> > > (where we >> > > install Foreman) but we provision CetnOS 7 on other machines on which >> you >> > > deploy OpenStack. >> > >> > Ah. I see. Is it possible to use CentOS 6.5 for host node AND openstack >> > deployment? >> > CentOS 7 is so new that I'm afraid install it on prod right now + I >> don't >> > know is Ceph storage already compatible with CentOS 7 >> > >> > > > Did I understand correctly that new CentOS versioning (eg. >> 7.0.1234) >> > > >> > > brakes >> > > >> > > > some puppet scripts that runs with staypuft-installer? >> > > >> > > More precisely, it breaks default foreman installation media which >> are >> > > used >> > > during provisioning. The revision number in CentOS path is not >> standard, >> > > previously they had only $major.$minor. As a workaround >> staypuft-installer >> > > could modify it but I'd like to see the fix for upstream foreman >> first, >> > > there >> > > should be some better way. Or CentOS guys will fix their mirrors :-) >> > >> > I think that CenOS guys won't fix that :-) . Seems that they decided to >> use >> > new version numbering. >> > For now I use dirty hack with mirror url >> > like http://mirror.centos.org/centos/7/os/$arch >> > >> > > Anyway I built installer 0.3.3 today which contains synxtax error >> fix, so >> > > if >> > > testing, use this (should get into nightly repo shortly). I'll try to >> get >> > > staypuft fix in today as well. >> > > >> > > On Friday 22 of August 2014 09:26:55 Nomnom Fuser wrote: >> > > > Thx for patch. I'll try it next week and share my results. >> > > > >> > > > On Friday, August 22, 2014 6:46:44 PM UTC+3, Marek Hulan wrote: >> > > > > Hello, >> > > > > >> > > > > since CentOS is not supported, noone noticed that it stopped >> working >> > > >> > > on >> > > >> > > > > CentOS. I made a fix on both installer and staypuft side today. >> You'll >> > > > > have to >> > > > > update and rerun the installer (0.3.2 should get to nightly >> tomorrow) >> > > >> > > to >> > > >> > > > > precreate staypuft OS v 7. Staypuft fix was not yet merged but >> you can >> > > > > apply it >> > > > > manually, see [1] for the patch. >> > > > > >> > > > > Before creating new deploy you must also fix one issue with >> CentOS >> > > > > installation >> > > > > media. For some reason, CentOS 7.0 mirror does not follow >> standard >> > > >> > > path so >> > > >> > > > > as >> > > > > a workaround you have to navigate to Hosts -> Installation Media >> -> >> > > >> > > CentOS >> > > >> > > > > Mirror and update path to >> > > >> > > http://mirror.centos.org/centos/$major/os/$arch. >> > > >> > > > > More information at [2]. >> > > > > >> > > > > Fixes should get in and be packaged in the beginning of next >> week. I'd >> > > >> > > be >> > > >> > > > > happy if you tested them and shared your results. >> > > > > >> > > > > [1] https://github.com/theforeman/staypuft/pull/271 >> > > > > [2] http://projects.theforeman.org/issues/6884 >> > > > > >> > > > > On Thursday 21 of August 2014 17:49:52 Marek Hulan wrote: >> > > > > > I was able to reproduce it. I'll look have a look at it >> tomorrow and >> > > > > >> > > > > will >> > > > > >> > > > > > let you know. >> > > > > > >> > > > > > > After downloading images I rerun installer and he told me >> that >> > > > > >> > > > > everything >> > > > > >> > > > > > > success: >> > > > > > > >> > > > > > > Starting configuration... >> > > > > > > Starting to seed provisioning data >> > > > > > > Use 'base_CentOS_6' hostgroup for provisioning >> > > > > > > >> > > > > > > Success! >> > > > > > > * Foreman is running at https://openstack-core.cloud >> > > > > > > >> > > > > > > Initial credentials are admin / df4Rauujuwxdede6 >> > > > > > > >> > > > > > > * Foreman Proxy is running at >> https://openstack-core.cloud:8443 >> > > > > > > * Puppetmaster is running at port 8140 >> > > > > > > The full log is at >> > > >> > > /var/log/foreman-installer/foreman-installer.log >> > > >> > > > > > > But UI told me same "missing Operatingsystem" >> > > > > > > and in foreman/production.log same error: >> > > > > > > >> > > > > > > Started GET "/deployments/new" for 10.10.10.1 at 2014-08-21 >> > > >> > > 17:14:12 >> > > >> > > > > +0300 >> > > > > >> > > > > > > Processing by Staypuft::DeploymentsController#new as HTML >> > > > > > > Operation FAILED: missing Operatingsystem >> > > > > > > >> > > > > > > Rendered common/500.html.erb within >> staypuft/layouts/staypuft >> > > > > >> > > > > (3.8ms) >> > > > > >> > > > > > > Rendered layouts/base.html.erb (0.9ms) >> > > > > > > Rendered >> > > > > >> > > > > /opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/ >> > > > > >> > > > > > > views/staypuft/layouts/application.html.erb (1.2ms) >> > > > > > > Completed 500 Internal Server Error in 79ms (Views: 6.7ms | >> > > > > >> > > > > ActiveRecord: >> > > > > > > 13.4ms) >> > > > > > > >> > > > > > > On Thursday, August 21, 2014 5:01:15 PM UTC+3, Marek Hulan >> wrote: >> > > > > > > > There seems to be some ordering issue with discovery >> images, >> > > >> > > target >> > > >> > > > > > > > directory >> > > > > > > > did not exist yet. When there appears an error during >> > > >> > > installation, >> > > >> > > > > > > > configuration won't continue. Your staypuft installation >> > > >> > > resulted in >> > > >> > > > > > > > being >> > > > > > > > installed but not configured. >> > > > > > > > >> > > > > > > > Now when you fixed discovery images issue, can you rerun >> > > >> > > installer >> > > >> > > > > once >> > > > > >> > > > > > > > more? >> > > > > > > > Until it shows that you can use some hostgroup for >> provisioning >> > > >> > > you >> > > >> > > > > > > > can't >> > > > > > > > consider staypuft to be ready. But if discovery images was >> the >> > > >> > > only >> > > >> > > > > > > > issue >> > > > > > > > it >> > > > > > > > should pass now. >> > > > > > > > >> > > > > > > > On Thursday 21 of August 2014 03:20:15 Nomnom Fuser wrote: >> > > > > > > > > Ok. Yet another fresh install. Centos 6.5 x86_64 >> > > > > > > > > repos: >> > > > > > > > > >> > > > > > > > > >> > > > > > > > > base >> > > > > > > > > >> > > > > > > > > CentOS-6 - Base >> > > > > > > > > >> > > > > > > > > epel >> > > > > > > > > >> > > > > > > > > Extra Packages for Enterprise >> Linux 6 >> > > >> > > - >> > > >> > > > > > > > > x86_64 >> > > > > > > > > >> > > > > > > > > extras >> > > > > > > > > >> > > > > > > > > CentOS-6 - Extras >> > > > > > > > > >> > > > > > > > > foreman >> > > > > > > > > >> > > > > > > > > Foreman nightly >> > > > > > > > > >> > > > > > > > > foreman-plugins >> > > > > > > > > >> > > > > > > > > Foreman plugins nightly >> > > > > > > > > >> > > > > > > > > scl >> > > > > > > > > >> > > > > > > > > CentOS-6 - SCL >> > > > > > > > > >> > > > > > > > > updates >> > > > > > > > > >> > > > > > > > > CentOS-6 - Updates >> > > > > > > > > >> > > > > > > > > # rpm -qa | grep foreman >> > > >> > > foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noarch >> > > >> > > > > > > > > foreman-installer-staypuft-0.3.0-1.el6.noarch >> > > >> > > foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noarch >> > > >> > > > > > > > > rubygem-foreman_api-0.1.11-1.el6.noarch >> > > > > >> > > > > >> foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch >> > > > > >> > > > > > > > > First install with "staypuft-installer >> > > > > > > > >> > > > > > > > --foreman-configure-epel-repo=true >> > > > > > > > >> > > > > > > > > --foreman-plugin-discovery-install-images=true" : >> > > > > > > > > >> > > > > > > > > [ INFO 2014-08-21 13:06:14 main] Executing hooks in group >> post >> > > > > > > > > [DEBUG 2014-08-21 13:06:14 main] Hook >> > > > > > > > >> > > > > > > > /usr/share/foreman-installer/hooks/ >> > > > > > > > >> > > > > > > > > post/10-setup_provisioning.rb returned false >> > > > > > > > > [ INFO 2014-08-21 13:06:14 main] All hooks in group post >> > > >> > > finished >> > > >> > > > > > > > > [DEBUG 2014-08-21 13:06:14 main] Exit with status code: 6 >> > > >> > > (signal >> > > >> > > > > was >> > > > > >> > > > > > > > > 6) >> > > > > > > > > [ERROR 2014-08-21 13:06:14 main] Repeating errors >> encountered >> > > > > >> > > > > during >> > > > > >> > > > > > > > run: >> > > > > > > > > [ERROR 2014-08-21 13:06:14 main] >> > > > > > > > >> > > > > > > > /Stage[main]/Foreman::Plugin::Discovery/ >> > > >> > > >> Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-late >> > > >> > > > > > > > st >> > > > > > > > . >> > > >> > > >> el6.iso-vmlinuz]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery-> >> > > >> > > > > > > > image-latest.el6.iso-vmlinuz]/returns: change from notrun >> to 0 >> > > > > >> > > > > failed: >> > > > > > > > /usr/ >> > > > > > > > >> > > > > > > > > bin/curl -s >> > > >> > > >> http://downloads.theforeman.org/discovery/releases/latest/foreman-discov >> > > >> > > > > > > > er >> > > > > > > > y >> > > > > > > > >> > > > > > > > > -image-latest.el6.iso-vmlinuz -o >> > > > > >> > > > > >> /var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-vmlinuz >> > > > > >> > > > > > > > > returned 23 instead of one of [0] at >> > > > > >> > > > > >> /usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp: >> > > > > > > > > 6 >> > > > > > > > > [ERROR 2014-08-21 13:06:14 main] >> > > > > > > > >> > > > > > > > /Stage[main]/Foreman::Plugin::Discovery/ >> > > >> > > >> Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-late >> > > >> > > > > > > > st >> > > > > > > > . >> > > >> > > >> el6.iso-img]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery-imag >> > > >> > > > > > > > e-> > > latest.el6.iso-img]/returns: change from notrun to >> 0 >> > > >> > > failed: >> > > > > > > > /usr/bin/curl> > >> > > > > > > > >> > > > > > > > > -s >> > > >> > > >> http://downloads.theforeman.org/discovery/releases/latest/foreman-discov >> > > >> > > > > > > > er >> > > > > > > > y >> > > > > > > > >> > > > > > > > > -image-latest.el6.iso-img -o >> > > >> > > /var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-img >> > > >> > > > > > > > returned >> > > > > > > > >> > > > > > > > > 23 instead of one of [0] at >> > > > > >> > > > > >> /usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp: >> > > > > > > > > Ok, I download images manually using same curl command. >> > > > > > > > > Second run with same options. Seems ok. "Success!" >> > > > > > > > > >> > > > > > > > > Login to UI, go to new deployment of openstack, >> > > >> > > aaaannnndddd..... >> > > >> > > > > > > > > Started GET "/deployments/new" for 10.10.10.1 at >> 2014-08-21 >> > > > > >> > > > > 13:14:11 >> > > > > >> > > > > > > > +0300 >> > > > > > > > >> > > > > > > > > Processing by Staypuft::DeploymentsController#new as HTML >> > > > > > > > > Operation FAILED: missing Operatingsystem >> > > > > > > > > >> > > > > > > > > Rendered common/500.html.erb within >> > > >> > > staypuft/layouts/staypuft >> > > >> > > > > > > > > (7.9ms) >> > > > > > > > > Rendered layouts/base.html.erb (2.0ms) >> > > > > > > > > Rendered >> > > >> > > /opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/ >> > > >> > > > > > > > > views/staypuft/layouts/application.html.erb (3.6ms) >> > > > > > > > > Completed 500 Internal Server Error in 508ms (Views: >> 16.6ms | >> > > > > > > > >> > > > > > > > ActiveRecord: >> > > > > > > > > 34.1ms) >> > > > > > > > > >> > > > > > > > > after installer: >> > > > > > > > > >> > > > > > > > > # rpm -qa | grep foreman >> > > > > > > > > foreman-installer-staypuft-0.3.0-1.el6.noarch >> > > > > > > > > ruby193-rubygem-foreman-tasks-0.6.4-1.el6.noarch >> > > > > >> > > > > >> foreman-postgresql-1.7.0-0.develop.201408201145git87a42cb.el6.noarch >> > > >> > > foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noarch >> > > >> > > > > > > > > rubygem-foreman_api-0.1.11-1.el6.noarch >> > > > > >> > > > > >> foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch >> > > > > >> > > > > > > > > foreman-1.7.0-0.develop.201408201145git87a42cb.el6.noarch >> > > >> > > foreman-proxy-1.7.0-0.develop.201408191935git7124635.el6.noarch >> > > >> > > foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noarch >> > > >> > > > > > > > > ruby193-rubygem-foreman_discovery-1.3.0-1.el6.noarch >> > > > > > > > > >> > > > > > > > > Seems like I need install openstack manually from source >> or >> > > >> > > using >> > > >> > > > > rdo >> > > > > >> > > > > > > > > without provisioning %( >> > > > > > > > > >> > > > > > > > > On Thursday, August 21, 2014 10:33:16 AM UTC+3, Marek >> Hulan >> > > >> > > wrote: >> > > > > > > > > > Hello, >> > > > > > > > > > >> > > > > > > > > > This error would indicate that installer did not finish >> all >> > > > > >> > > > > steps >> > > > > >> > > > > > > > > > successfully. >> > > > > > > > > > Could you rerun the installer and try again? >> >> -- >> Marek >> >

Hello Ignacio,

unfortunately there are more patches in master branch already and they are not
compatible with current foreman develop branch. We are waiting to get
networking PR into foreman core [1], then I can release. Until then only
workaround is to manually bump discovery version in gem specification (and
install discovery 1.4.0.rc1 unless you've already installed it).

[1] https://github.com/theforeman/foreman/pull/1565

··· -- Marek

On Saturday 30 of August 2014 02:37:18 Ignacio Bravo wrote:

Marek,

Can you please create a new staypuft RPM version from the github
repositories that includes the latest change to incorporate 1.4.0.rc1 as
the new required version for foreman-discovery plugin vs the 1.3.0.rc1
that is in included in the latest rpm.

I was able to trace the error in the installation back to that change
but my technical skills are not yet there to deploy the github version
into my machine. :wink:

Foreman

https://github.com/theforeman/staypuft/commit/87ceae281a3ce864030bbd608c75c7
6a2bb65124#diff-32c0f0d9976a9b49506d42ab8f4e0520

Thanks.

On 08/26/2014 03:49 AM, Marek Hulan wrote:

Hello guys,

staypuft 0.3.0 and foreman-installer-staypuft 0.3.3 were uploaded to
nightly repos containing all CentOS repos up to date. There’s still
workaround needed regarding installation media. I publish the fix for
downloading discovery images [1] but it will take some time before it
gets in, so until that another workaround is needed - precreate
/var/lib/tftpboot/boot before running installer for the first time (or
rerun installer after first failed installation).

If you update existing installation, be sure to rerun installer after yum
upgrade. Also feel free to stop by at #theforeman-dev channel on freenode
and share your experience.


Marek

[1] https://github.com/theforeman/puppet-foreman/pull/217

On Monday 25 of August 2014 10:39:42 Marek Hulan wrote:

Hi,

you’re right, it seems like some ordering issue in upstream puppet
module.
I’ll try to look at that. Meanwhile the workaround is what you say, you
can
use shorter variant

mkdir -p /var/lib/tftpboot/boot

Thanks for the report

Marek,

Thanks for the patch. I updated the file with your details, but the
installation was still failing during the image retrieval phase. I
thought
that the issue was related to the files being downloaded before the
directory was created, so I created the directory and it worked! Here is
the script that I was using for the installation:

==Install==
yum install -y
http://yum.theforeman.org/nightly/el6/x86_64/foreman-release.rpm
yum -y localinstall
http://mirror.pnl.gov/epel/6/x86_64/epel-release-6-8.noarch.rpm
yum -y install centos-release-SCL
yum -y install foreman-plugin-staypuft
yum -y install foreman-installer-staypuft

mkdir /var/lib/tftpboot
mkdir /var/lib/tftpboot/boot

Now modify the file:

/opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/models/stay
pu
ft /deployment.rb based on your instructions and then:

staypuft-installer --foreman-configure-epel-repo=true
–foreman-plugin-discovery-install-images=true

== end ====

It appears that the module with problems is this one:
/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp

On Friday, August 22, 2014 12:26:55 PM UTC-4, Nomnom Fuser wrote:

Thx for patch. I’ll try it next week and share my results.

You refer to issue with CentOS 7 and I use Centos 6.5 as host and for
provisioning.
Did I understand correctly that new CentOS versioning (eg. 7.0.1234)
brakes some puppet scripts that runs with staypuft-installer?

On Friday, August 22, 2014 6:46:44 PM UTC+3, Marek Hulan wrote:

Hello,

since CentOS is not supported, noone noticed that it stopped working
on
CentOS. I made a fix on both installer and staypuft side today. You’ll
have to
update and rerun the installer (0.3.2 should get to nightly tomorrow)
to
precreate staypuft OS v 7. Staypuft fix was not yet merged but you can
apply it
manually, see [1] for the patch.

Before creating new deploy you must also fix one issue with CentOS
installation
media. For some reason, CentOS 7.0 mirror does not follow standard
path
so as
a workaround you have to navigate to Hosts -> Installation Media ->
CentOS
Mirror and update path to
http://mirror.centos.org/centos/$major/os/$arch.

More information at [2].

Fixes should get in and be packaged in the beginning of next week. I’d
be
happy if you tested them and shared your results.

[1] https://github.com/theforeman/staypuft/pull/271
[2] Bug #6884: Centos mirror in "Installation Media" doesn't work for versions without $minor - Foreman

On Thursday 21 of August 2014 17:49:52 Marek Hulan wrote:

I was able to reproduce it. I’ll look have a look at it tomorrow and

will

let you know.

After downloading images I rerun installer and he told me that

everything

success:

Starting configuration…
Starting to seed provisioning data
Use ‘base_CentOS_6’ hostgroup for provisioning

Success!

But UI told me same "missing Operatingsystem"
and in foreman/production.log same error:

Started GET “/deployments/new” for 10.10.10.1 at 2014-08-21
17:14:12

+0300

Processing by Staypuft::DeploymentsController#new as HTML
Operation FAILED: missing Operatingsystem

Rendered common/500.html.erb within staypuft/layouts/staypuft

(3.8ms)

Rendered layouts/base.html.erb (0.9ms)
Rendered

/opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/

views/staypuft/layouts/application.html.erb (1.2ms)
Completed 500 Internal Server Error in 79ms (Views: 6.7ms |

ActiveRecord:

13.4ms)

On Thursday, August 21, 2014 5:01:15 PM UTC+3, Marek Hulan wrote:

There seems to be some ordering issue with discovery images,
target
directory
did not exist yet. When there appears an error during
installation,
configuration won’t continue. Your staypuft installation resulted

in

being
installed but not configured.

Now when you fixed discovery images issue, can you rerun
installer

once

more?
Until it shows that you can use some hostgroup for provisioning
you
can’t
consider staypuft to be ready. But if discovery images was the
only
issue
it
should pass now.

On Thursday 21 of August 2014 03:20:15 Nomnom Fuser wrote:

Ok. Yet another fresh install. Centos 6.5 x86_64
repos:

base

                      CentOS-6 - Base

epel

                      Extra Packages for Enterprise Linux 6
                      -
                      x86_64

extras

                      CentOS-6 - Extras

foreman

                     Foreman nightly

foreman-plugins

                     Foreman plugins nightly

scl

                     CentOS-6 - SCL

updates

                     CentOS-6 - Updates

rpm -qa | grep foreman

foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noar
ch
foreman-installer-staypuft-0.3.0-1.el6.noarch
foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noar
ch
rubygem-foreman_api-0.1.11-1.el6.noarch

foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch

First install with "staypuft-installer

–foreman-configure-epel-repo=true

–foreman-plugin-discovery-install-images=true" :

[ INFO 2014-08-21 13:06:14 main] Executing hooks in group post
[DEBUG 2014-08-21 13:06:14 main] Hook

/usr/share/foreman-installer/hooks/

post/10-setup_provisioning.rb returned false
[ INFO 2014-08-21 13:06:14 main] All hooks in group post
finished
[DEBUG 2014-08-21 13:06:14 main] Exit with status code: 6
(signal

was

[ERROR 2014-08-21 13:06:14 main] Repeating errors encountered

during

run:

[ERROR 2014-08-21 13:06:14 main]

/Stage[main]/Foreman::Plugin::Discovery/

Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-la
t
e

st
.

el6.iso-vmlinuz]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discover
y
->

image-latest.el6.iso-vmlinuz]/returns: change from notrun to 0

failed:

/usr/

bin/curl -s

http://downloads.theforeman.org/discovery/releases/latest/foreman-disc
o
v

er
y

-image-latest.el6.iso-vmlinuz -o

/var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-vmlinuz

returned 23 instead of one of [0] at

/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp:

6
[ERROR 2014-08-21 13:06:14 main]

/Stage[main]/Foreman::Plugin::Discovery/

Foreman::Remote_file[/var/lib/tftpboot/boot/foreman-discovery-image-la
t
e

st
.

el6.iso-img]/Exec[retrieve_/var/lib/tftpboot/boot/foreman-discovery-im
a
g

e-> > > latest.el6.iso-img]/returns: change from notrun to 0

failed:

/usr/bin/curl> >

-s

http://downloads.theforeman.org/discovery/releases/latest/foreman-disc
o
v

er
y

-image-latest.el6.iso-img -o
/var/lib/tftpboot/boot/foreman-discovery-image-latest.el6.iso-i
mg

returned

23 instead of one of [0] at

/usr/share/foreman-installer/modules/foreman/manifests/remote_file.pp:

Ok, I download images manually using same curl command.
Second run with same options. Seems ok. “Success!”

Login to UI, go to new deployment of openstack,
aaaannnndddd…

Started GET “/deployments/new” for 10.10.10.1 at 2014-08-21

13:14:11

+0300

Processing by Staypuft::DeploymentsController#new as HTML
Operation FAILED: missing Operatingsystem

Rendered common/500.html.erb within staypuft/layouts/staypuft
(7.9ms)
Rendered layouts/base.html.erb (2.0ms)
Rendered
/opt/rh/ruby193/root/usr/share/gems/gems/staypuft-0.1.20/app/

views/staypuft/layouts/application.html.erb (3.6ms)
Completed 500 Internal Server Error in 508ms (Views: 16.6ms |

ActiveRecord:

34.1ms)

after installer:

rpm -qa | grep foreman

foreman-installer-staypuft-0.3.0-1.el6.noarch
ruby193-rubygem-foreman-tasks-0.6.4-1.el6.noarch

foreman-postgresql-1.7.0-0.develop.201408201145git87a42cb.el6.noarch

foreman-release-1.7.0-0.develop.201408201145git87a42cb.el6.noar
ch
rubygem-foreman_api-0.1.11-1.el6.noarch

foreman-installer-1.7.0-0.develop.201408210326git3551ee3.el6.noarch

foreman-1.7.0-0.develop.201408201145git87a42cb.el6.noarch
foreman-proxy-1.7.0-0.develop.201408191935git7124635.el6.noarch
foreman-selinux-1.7.0-0.develop.201408202047gitd5e80cf.el6.noar
ch
ruby193-rubygem-foreman_discovery-1.3.0-1.el6.noarch

Seems like I need install openstack manually from source or
using

rdo

without provisioning %(

On Thursday, August 21, 2014 10:33:16 AM UTC+3, Marek Hulan > >>>>> > >>>>> wrote:

Hello,

This error would indicate that installer did not finish all

steps

successfully.
Could you rerun the installer and try again?


Marek

>
> The new version does allow for the creation of deployments. Great!
>
> Now the problem arises when assigning the functions to the discovered
> nodes.
> a. After assigning them, the nodes are not rebooted out of the ovirt image
> and into the new CentOS 7 image. I need to manually reboot the nodes for
> them to take the new image. That is not much of a problem but more of a
> feature that was included in the discovered hosts before.
>
on bare metal everything reboot normally

>
> b. After loading the CentOS image and booting up, the nodes finish with an
> error:
>
> anaconda 19.31.79-1 for CentOS 7 started.
> Running pre-installation scripts
> The following problem occurred on line 0 of the kickstart file:
>
> Unable to open input kickstart file: curl#37 - "Could't open file
> /tmp/diskpart.cfg"
>
I fix that issue in UI -> Hosts -> Partition tables -> OpenStack default.
If you use another partition table see there.
My file now look like this:

#Dynamic
cat <<EOF > /tmp/diskpart.cfg
zerombr
clearpart --all --initlabel
part /boot --fstype ext3 --size=500 --ondisk=sda
part swap --size=1024 --ondisk=sda
part pv.01 --size=1024 --grow --ondisk=sda
volgroup vg_root pv.01
logvol / --vgname=vg_root --size=1 --grow --name=lv_root
EOF

I've added second and last line.

··· On Wednesday, August 27, 2014 6:32:16 PM UTC+3, Ignacio Bravo wrote:

If I open a shell on the node, on the /tmp directory there are
ks-script-JpfOS3 and ks-script-JpfOS3.log. The first one reads:

#Dynamic
zerombr
clearpart --all --initlabel

the second one reads:

line 2: zerombr: command not found
line 3: clearpart: command not found

BTW, I am installing all into a VirtualBox interfase

And one more thing. After playing wih this for about 10 hours, I came to
the conclusion that puppet is not being loaded via the kickstart and
that causes my servers not to reboot and also not to be configured after
being built.

I belive that this line in the kickstart default is the one not working.

repo --name="Extra Packages for Enterprise Linux"
–mirrorlist=https://mirrors.fedoraproject.org/metalink?repo=epel-<%=
@host.operatingsystem.major %>&arch=<%= @host.architecture %>

What is odd is that if I replace the values, in a web browser, I can get
the file. And also that this configuration works in another machine with
Centos 6.5

Ideas?

IB

··· On 08/27/2014 12:10 PM, Nomnom Fuser wrote: > > > On Wednesday, August 27, 2014 6:32:16 PM UTC+3, Ignacio Bravo wrote: > > The new version does allow for the creation of deployments. Great! > > Now the problem arises when assigning the functions to the > discovered nodes. > a. After assigning them, the nodes are not rebooted out of the > ovirt image and into the new CentOS 7 image. I need to manually > reboot the nodes for them to take the new image. That is not much > of a problem but more of a feature that was included in the > discovered hosts before. > > on bare metal everything reboot normally > > > b. After loading the CentOS image and booting up, the nodes finish > with an error: > > anaconda 19.31.79-1 for CentOS 7 started. > Running pre-installation scripts > The following problem occurred on line 0 of the kickstart file: > > Unable to open input kickstart file: curl#37 - "Could't open file > /tmp/diskpart.cfg" > > I fix that issue in UI -> Hosts -> Partition tables -> OpenStack > default. If you use another partition table see there. > My file now look like this: > > > > #Dynamic > cat </tmp/diskpart.cfg > zerombr > clearpart --all --initlabel > part /boot --fstype ext3 --size=500--ondisk=sda > part swap --size=1024--ondisk=sda > part pv.01--size=1024--grow --ondisk=sda > volgroup vg_root pv.01 > logvol /--vgname=vg_root --size=1--grow --name=lv_root > EOF > > > > > I've added second and last line. > > If I open a shell on the node, on the /tmp directory there are > ks-script-JpfOS3 and ks-script-JpfOS3.log. The first one reads: > > #Dynamic > zerombr > clearpart --all --initlabel > ... > > the second one reads: > > line 2: zerombr: command not found > line 3: clearpart: command not found > ... > > BTW, I am installing all into a VirtualBox interfase > > -- > You received this message because you are subscribed to a topic in the > Google Groups "Foreman users" group. > To unsubscribe from this topic, visit > https://groups.google.com/d/topic/foreman-users/MUM5iAwwVUc/unsubscribe. > To unsubscribe from this group and all its topics, send an email to > foreman-users+unsubscribe@googlegroups.com > . > To post to this group, send email to foreman-users@googlegroups.com > . > Visit this group at http://groups.google.com/group/foreman-users. > For more options, visit https://groups.google.com/d/optout.

Maybe you could try provision a clean CentOS and see whether networking works
correctly. Also I'd double check DHCP configuration (set by staypuft installer,
values in first screen) and make sure you have correct nameserver on
provisioned hosts. I think that DHCP on staypuft machine is not configured to
do recursive queries by default.

··· -- Marek

On Thursday 28 of August 2014 02:17:39 Ignacio Bravo wrote:

And one more thing. After playing wih this for about 10 hours, I came to
the conclusion that puppet is not being loaded via the kickstart and
that causes my servers not to reboot and also not to be configured after
being built.

I belive that this line in the kickstart default is the one not working.

repo --name=“Extra Packages for Enterprise Linux”
–mirrorlist=https://mirrors.fedoraproject.org/metalink?repo=epel-<%=
@host.operatingsystem.major %>&arch=<%= @host.architecture %>

What is odd is that if I replace the values, in a web browser, I can get
the file. And also that this configuration works in another machine with
Centos 6.5

Ideas?

IB

On 08/27/2014 12:10 PM, Nomnom Fuser wrote:

On Wednesday, August 27, 2014 6:32:16 PM UTC+3, Ignacio Bravo wrote:
The new version does allow for the creation of deployments. Great!

Now the problem arises when assigning the functions to the
discovered nodes.
a. After assigning them, the nodes are not rebooted out of the
ovirt image and into the new CentOS 7 image. I need to manually
reboot the nodes for them to take the new image. That is not much
of a problem but more of a feature that was included in the
discovered hosts before.

on bare metal everything reboot normally

b. After loading the CentOS image and booting up, the nodes finish
with an error:

anaconda 19.31.79-1 for CentOS 7 started.
Running pre-installation scripts
The following problem occurred on line 0 of the kickstart file:

Unable to open input kickstart file: curl#37 - "Could't open file
/tmp/diskpart.cfg"

I fix that issue in UI -> Hosts -> Partition tables -> OpenStack
default. If you use another partition table see there.
My file now look like this:

#Dynamic
cat </tmp/diskpart.cfg
zerombr
clearpart --all --initlabel
part /boot --fstype ext3 --size=500–ondisk=sda
part swap --size=1024–ondisk=sda
part pv.01–size=1024–grow --ondisk=sda
volgroup vg_root pv.01
logvol /–vgname=vg_root --size=1–grow --name=lv_root
EOF

I’ve added second and last line.

If I open a shell on the node, on the /tmp directory there are
ks-script-JpfOS3 and ks-script-JpfOS3.log. The first one reads:

#Dynamic
zerombr
clearpart --all --initlabel
...

the second one reads:

line 2: zerombr: command not found
line 3: clearpart: command not found
...

BTW, I am installing all into a VirtualBox interfase


Marek