(MODULES-8648) - Fix for failures on SLES 11 #816
Conversation
This comment has been minimized.
This comment has been minimized.
david22swan
changed the title
There was a problem hiding this comment.
Couple of things I'm not sure about @david22swan , comments in code 👍
| @@ -140,6 +140,11 @@ | |||
| case $::operatingsystem { | |||
| 'CentOS': { | |||
| case $::operatingsystemrelease { | |||
| /^5\..*/: { | |||
There was a problem hiding this comment.
Why is this part of this PR? Has nothing to do with SLES 11.
There was a problem hiding this comment.
This was removed by accident in an earlier pr
There was a problem hiding this comment.
Still has nothing to do with this PR. Either revert the other PR that introduced it or separate it into its own commit. This doesn't make for good git history and would be hard to track.
There was a problem hiding this comment.
Will separate it out into a 2nd commit
spec/acceptance/rules_spec.rb
Outdated
| @@ -241,7 +240,7 @@ class { '::firewall': } | |||
| it 'applies cleanly' do | |||
| # Run it twice and test for idempotency | |||
| apply_manifest(pp2, catch_failures: true) | |||
| apply_manifest(pp2, catch_changes: do_catch_changes) | |||
| apply_manifest(pp2, catch_changes: true) unless fact('operatingsystem') == 'SLES' && fact('operatingsystemmajrelease') == '11' | |||
There was a problem hiding this comment.
You seem to skip the idempotent checks, is there a reason for this?
There was a problem hiding this comment.
on sles 11 sp4 State has been deprecated and given a warning to use ctstate instead. It seems to do this for you automatically so everytime the manifest runs it applies as state, but then the firewall changes this to ctstate
There was a problem hiding this comment.
I feel like this needs to be a code change instead? If we are passing through the wrong flag we are always going to get those change 'failures', whereas I think the commands being ran need updated to use the -ctstate flag instead of -state?
There was a problem hiding this comment.
Have been looking at using an if to swap between the two for sless 11.4 but replacing all mentions of state with ctstate might be good as well. They are simply different packages you can choose between. Ctstate just has more features and complexity.
david22swan
changed the title
|
Adhoc running |
Code was removed by mistake during removal of Scientific 5 support
Failures caused by the deprecation of `state` on SLES 11 SP4. Warning has been placed in readme and test's have been updated to use `ctstate` instead.
|
|
|
👍 |
SLES 11 Sp4 seems to default to ‘-m conntrack’ rather than ‘-m state’
In addition, adding back a piece of code for Centos 5