So that pupet doesn't try to do it on its own (parallel processes
could be racy).
Also, make grep quiet when searching for nodes to apply.
Change-Id: I5012278cfd98c7e3d39ca75794ce97366289d7a1
Double [[]] are needed by bash for the regex operator. Use them
universally for safety.
And make gate-config-puppet-apply-centos6 voting.
Change-Id: I8f245a86381d24d8938df444e0268343b46fe09f
Associate an OS with each node defined in site.pp so that the puppet
apply test will only test nodes apprapriate to the node the test
is running on.
Fix remaining errors related to this test, and make it vote.
Also, start running the centos6 version regularly
(which is still non-voting).
Change-Id: I26722dce15589c982af5de36f8ccf42666d29f2b
Put the variables defined in manifests/site.pp at the top of the file so
that they are in a known location after running csplit in test.sh and we
can prepend them to the puppet-apply top files.
A better solution would be to move this data into hiera, but this is not
sensitive data, so a move to hiera should wait until we a have solution
for a public hiera data repo separate from the private hiera data.
Change-Id: I509a8266462dfdf53e1727938e4fb043241166b6
Puppet apply is single threaded and slow. We are testing many puppet
applies in noop mode so it should be safe to run them in parallel using
the number of cpus available to us. Make it so.
Change-Id: I66250d21258de7791e4f5fc0b013100efae63603
The test.sh script is not currently being run in any jobs, this change
removes the redundant validation code that's also in the puppet-syntax
job and creates a puppet-apply-test job that runs the test.sh script.
Running `puppet apply --noop` requires sudo, otherwise it will give
errors about refusing to run commands as other users.
Change-Id: Ie6b278d98390a8a5dd8bb24899c8c4083f5755c9
This allows hiera data to be split into two buckets: sensitive and
non-sensitive. Sensitive data is put where it always has been. Non-
sensitive data, such as users ssh keys and lists of packages can be
put into the data directory contained within the
openstack-infra/config git repository.
This also introduces a more complex hierarchy: data is now split on
osfamily and lsbdistcondename. Operatingsystem is left for
compatibility.
Change-Id: I3b3faeab837aa5732da01e96bbe0c19aa936a58d
Pass the sysadmins list into each node definition. This allows us to
retrieve the data from hiera rather than hard coding it in the puppet
manifests. Also, update test script to use bogus sysadmin data when
testing.
Change-Id: Ide3560f16bce4d66fb95cc5021fc879476e6a712
Reviewed-on: https://review.openstack.org/12512
Reviewed-by: James E. Blair <corvus@inaugust.com>
Approved: Monty Taylor <mordred@inaugust.com>
Reviewed-by: Monty Taylor <mordred@inaugust.com>
Tested-by: Jenkins
This way, I can easily run it locally before submitting. Also, updated a
little bit to put in some hacks around hiera calls.
Change-Id: Ie4f3de6d81d915fbf06e896f31362c836fd244a9
Reviewed-on: https://review.openstack.org/10834
Approved: Monty Taylor <mordred@inaugust.com>
Reviewed-by: Monty Taylor <mordred@inaugust.com>
Tested-by: Jenkins
