QA had asked me a while ago to clean up the way the precedence list for 1.x was worded,
as the intro from the list started with "then comes", as if something should preceed
it. The comments from OxABAB were not helpful themselves, but his issue reminded me that
this was on my to do list to make a little cleaner and clearer. Edits made to remove the
"then comes" intros for each list line, to help with clarity.
Now that Github supports separate issue and PR templates, we can have a
separate cut-down version for PRs without all the things we ask for in a
new issue. The PR types are also removed from the ISSUE_TEMPLATE.
The old template was effusive at the expense of making the text harder
to read and easier to miss things in. This one is more direct, and easy
to scan quickly.
will display on certain verbosity levels, both playbook/file info
and non empty options with which it's running.
avoid errors when not using CLI classes
The setup module calls /bin/lsblk once for each device appearing in the /etc/mtab file. However, the same device appears there mutliple times when the system uses bind-mounts. As a result, /bin/lsblk is being called repeatedly to get the uuid of the same device.
On a system with many mounts, this leads to a TimeoutError in the get_mount_facts function of the setup module as described in #14551.
Fixes#14551
This commit fixes a situation where connection errors would be caught
but no useful information display. The connection error is now caught
and emitted in a call to fail_json
This commit fixes a situation where connection errors would be caught
but no useful information display. The connection error is now caught
and emitted in a call to fail_json
This commit fixes a situation where connection errors would be caught
but no useful information display. The connection error is now caught
and emitted in a call to fail_json
This commit fixes a situation where connection errors would be caught
but no useful information display. The connection error is now caught
and emitted in a call to fail_json
This commit fixes a situation where connection errors would be caught
but no useful information display. The connection error is now caught
and emitted in a call to fail_json
