list all vmdb directories in the gem

[kbrock]

Overview

Fixes #332

The goal is to remove a warning:

warning: File listed twice: /var/www/miq/vmdb/certs
(a few hundred files do this)

Before

%defattr(-,root,root,-)
%{app_root}
%attr(-,manageiq,manageiq) %{app_root}/certs

We stated that we included all files recursively under app_root (/var/www/miq/vmdb).
Then we stated a few directories (with different privileges) under app_root.
So we ended up with duplicates.

Why?

Most app_root subdirectories are owned by root, which is the default, so using the app_root line works.

There are a few config directories that need to be editable and owned by the end user. That is why they are listed with the %attr() syntax.

After

We explicitly listed each folder in the rails root.

I would have liked to keep on using the app_root rule so any additional files would be auto added.

This is tricky because we tend to not debug the directory listing of the rpm So there is a good chance that we will add a directory to app_root (vmdb / manageiq-core root) and will think it is all set but will fail on the appliance.

[bdunne]

Do we also need to add a %dir %{app_root} for the directory itself?

[kbrock]

so %dir says to add a directory only and not the files underneath it.

I was pretty sure that adding the file app_root/AUTHORS (in addition to the next 10
lines) would add app_root with the default attributes.

[Fryguy]

Based on what I'm reading we probably need %dir for app_root, since we want the rpm to own that dir.

[Fryguy]

I'm concerned keeping this up to date will be a pain - I wonder if we can leverage %exclude to make it work? Something like the following?

%files core
%defattr(-,root,root,-)
%{app_root}
%exclude %{app_root}/certs
%attr(-,manageiq,manageiq) %{app_root}/certs
%exclude %{app_root}/config
%attr(-,manageiq,manageiq) %{app_root}/config
%exclude %{app_root}/log
%attr(-,manageiq,manageiq) %{app_root}/log
%exclude %{app_root}/tmp
%attr(-,manageiq,manageiq) %{app_root}/tmp
%exclude %{app_root}/data
%attr(-,manageiq,manageiq) %{app_root}/data
%config(noreplace) %{app_root}/config/cable.yml
%exclude %{app_root}/public/pictures
%exclude %{app_root}/public/assets
...

[kbrock]

a. I'm very concerned about keeping this up to date
b. the exclude rule is the first one I tried

I'll try one more time with the %exclude operations.
The rpms were missing the directories last time I tries
The last I'm pretty sure it did not include any of those directories

Alternatives

Our big issue is that we are trying to put down a directory with both code and user data there.

1. Put the user files elsewhere (e.g.: copy daemons like httpd and put config in /etc/httpd)
2. Just set everything owned by manageiq user (code and config files) and be done with it
3. Put the user files into a separate rpm (where everything is miq owned)

[kbrock]

yea, just tried the exclude example again. it is so clean but it does not work

%defattr(-,root,root,-)
%{app_root}
%exclude %{app_root}/certs
%exclude %{app_root}/config
%exclude %{app_root}/data
%exclude %{app_root}/log
%exclude %{app_root}/tmp
%attr(-,manageiq,manageiq) %{app_root}/certs
%attr(-,manageiq,manageiq) %{app_root}/config
%attr(-,manageiq,manageiq) %{app_root}/data
%attr(-,manageiq,manageiq) %{app_root}/log
%attr(-,manageiq,manageiq) %{app_root}/tmp
%config(noreplace) %{app_root}/config/cable.yml
%exclude %{app_root}/public/pictures
%exclude %{app_root}/public/assets

[miq-bot]

This pull request has been automatically marked as stale because it has not been updated for at least 3 months.

If these changes are still valid, please remove the stale label, make any changes requested by reviewers (if any), and ensure that this issue is being looked at by the assigned/reviewer(s)

Thank you for all your contributions! More information about the ManageIQ triage process can be found in the triage process documentation.

[miq-bot]

This pull request has been automatically closed because it has not been updated for at least 3 months.

Feel free to reopen this pull request if these changes are still valid.

Thank you for all your contributions! More information about the ManageIQ triage process can be found in the triage process documentation.

[kbrock]

think our current suggestion is to extract user content into its own gem and have all of those files owned by manageiq user

[miq-bot]

Checked commit kbrock@80b19f9 with ruby 2.6.10, rubocop 1.28.2, haml-lint 0.35.0, and yamllint
0 files checked, 0 offenses detected
Everything looks fine. 👍

[miq-bot]

This pull request has been automatically marked as stale because it has not been updated for at least 3 months.

If these changes are still valid, please remove the stale label, make any changes requested by reviewers (if any), and ensure that this issue is being looked at by the assigned/reviewer(s)

Thank you for all your contributions! More information about the ManageIQ triage process can be found in the triage process documentation.

[miq-bot]

This pull request has been automatically marked as stale because it has not been updated for at least 3 months.

If these changes are still valid, please remove the stale label, make any changes requested by reviewers (if any), and ensure that this issue is being looked at by the assigned/reviewer(s).

[miq-bot]

This pull request has been automatically marked as stale because it has not been updated for at least 3 months.

If these changes are still valid, please remove the stale label, make any changes requested by reviewers (if any), and ensure that this issue is being looked at by the assigned/reviewer(s).

[miq-bot]

This pull request has been automatically marked as stale because it has not been updated for at least 3 months.

If these changes are still valid, please remove the stale label, make any changes requested by reviewers (if any), and ensure that this issue is being looked at by the assigned/reviewer(s).