Bug 236112

Summary: net/glusterfs: Startup should happen before mountlate
Product: Ports & Packages Reporter: Daniel Morante <daniel>
Component: Individual Port(s)Assignee: Kubilay Kocak <koobs>
Status: Open ---    
Severity: Affects Many People CC: craig001
Priority: --- Keywords: needs-qa
Version: LatestFlags: bugzilla: maintainer-feedback? (craig001)
koobs: merge-quarterly?
Hardware: Any   
OS: Any   
Attachments:
Description Flags
Before changing rc script
none
After changing rc script
none
Unified diff to fix the mentioned issue none

Description Daniel Morante 2019-02-28 18:45:14 UTC
Created attachment 202448 [details]
Before changing rc script

To support the ability to have replicated volumes auto mount on boot, glusterd should start up before the system starts mounting late filessystems.

In `/usr/local/etc/rc.d/glusterd`

Change:

```
# PROVIDE: glusterd
# REQUIRE: LOGIN
# KEYWORD: shutdown
```

To:

```
# PROVIDE: glusterd
# BEFORE: mountlate
# REQUIRE: NETWORKING
# KEYWORD: shutdown
```

For context, `/etc/fstab` has (example):

```
localhost:replicated	/path/to/replicated	fusefs	rw,_netdev,backupvolfile-server=host1:host2:host3,mountprog=/usr/local/sbin/mount_glusterfs,late	0	0
```
Comment 1 Daniel Morante 2019-02-28 18:45:40 UTC
Created attachment 202449 [details]
After changing rc script
Comment 2 craig001 2019-03-04 09:51:19 UTC
Folks,  please change maintainership of the port.
Comment 3 Daniel Morante 2019-08-21 01:58:14 UTC
Created attachment 206744 [details]
Unified diff to fix the mentioned issue

Created patch to fix the problem.
Comment 4 Kubilay Kocak freebsd_committer freebsd_triage 2019-08-21 05:30:05 UTC
Thank you for the report and patch Daniel

^Triage: Assign myself to take care of the maintainership reset (comment 2) and open issues (for glusterfs) re-triage