Bug 764216 (GLUSTER-2484)

Summary: Default start/stop levels in init.d script are wrong for boot mount
Product: [Community] GlusterFS Reporter: Alex/AT <alex>
Component: buildAssignee: Sachidananda Urs <sac>
Status: CLOSED CURRENTRELEASE QA Contact:
Severity: low Docs Contact:
Priority: medium    
Version: 3.1.2CC: amarts, gluster-bugs, sac, vijay
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: glusterfs-3.4.0 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2013-07-24 17:37:01 UTC Type: ---
Regression: --- Mount Type: fuse
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Bug Depends On:    
Bug Blocks: 817967    

Description Alex/AT 2011-03-03 08:29:17 UTC 
The default init script for glusterd says:

# chkconfig: 35 90 12

This makes it start at order 90, and stop at order 12.

At least under RHEL/CentOS and compatible systems it will cause boot problems if we are mounting local node data (or using DNS-RR with only the local node starting), and can also cause shutdown problems as well.

Why is it so:

netfs (network fs mounting script) starts at order 25, and stops at order 75. So at the moment of netfs start/stop local glusterfsd node will not be yet/already available.

Suggestion:

We may change init script as following.

# chkconfig: 35 20 80

So the glusterd will start after network (it is at 10/90), but just before netfs.
Comment 1 Anand Avati 2011-11-24 01:50:50 UTC 
CHANGE: http://review.gluster.com/746 (In cases of client and server sharing the same machine, mounting GlusterFS at) merged in master by Anand Avati (avati)