Bug 14847

Summary:	NFS lockd is not stopped properly by init script
Product:	[Retired] Red Hat Linux	Reporter:	Jos Vos <jos>
Component:	kernel	Assignee:	Michael K. Johnson <johnsonm>
Status:	CLOSED RAWHIDE	QA Contact:
Severity:	medium	Docs Contact:
Priority:	high
Version:	7.1	CC:	alan, henris, johnsonm, katzj, notting, twaugh
Target Milestone:	---
Target Release:	---
Hardware:	i386
OS:	Linux
Whiteboard:	Winston rc1
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2000-08-23 15:27:38 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Jos Vos 2000-07-30 15:09:46 UTC

/etc/rc.d/init.d/nfslock stop never stops NFS' lockd properly, so you
always see a red FAILED indication.

Comment 1 Glen Foster 2000-07-30 22:09:26 UTC

This defect is considered MUST-FIX for Winston Release-Candidate #1

Comment 2 Glen Foster 2000-07-30 22:27:27 UTC

*** Bug 12316 has been marked as a duplicate of this bug. ***

Comment 3 Bill Nottingham 2000-07-31 14:54:57 UTC

This a kernel problem, unfortunately.

Comment 4 Bill Nottingham 2000-08-01 18:27:51 UTC

*** Bug 14994 has been marked as a duplicate of this bug. ***

Comment 5 Bill Nottingham 2000-08-01 20:27:35 UTC

*** Bug 14959 has been marked as a duplicate of this bug. ***

Comment 6 Jeremy Katz 2000-08-11 20:35:20 UTC

Problem still exists in RC1

Comment 7 Henri Schlereth 2000-08-15 16:28:18 UTC

Additional verification RC1/TUI Upgrade/Install -- Server still fails.

Comment 8 Tim Waugh 2000-08-16 15:15:56 UTC

Getting /etc/rc.d/init.d/functions to wait a little while (udelay 100000) after
sending SIGKILL seems to work around it here.  Should lockd catch SIGTERM
instead though?

Comment 9 Preston Brown 2000-08-16 17:38:57 UTC

*** Bug 16344 has been marked as a duplicate of this bug. ***

Comment 10 Alan Cox 2000-08-16 20:03:15 UTC

Doesnt happen with a standard kernel

Comment 11 Michael K. Johnson 2000-08-20 21:13:29 UTC

Bill, let's implement the workaround Tim found for now, and revert
it if Doug finds the problem in time.

Comment 12 Bill Nottingham 2000-08-21 16:03:28 UTC

Yes, lockd should catch sigterm, I would think.
The 'not waiting after sigkill is actually a real bug, more
or less. Added in initscripts, will be in 5.47 or so.

Comment 13 Doug Ledford 2000-08-22 04:29:29 UTC

Solved by a patch checked into CVS as of 12:30AM, Tue Aug 22, 2000
(Patch29: linux-2.2.16-lockd.patch in the spec file)

Comment 14 Michael K. Johnson 2000-08-22 13:18:35 UTC

Fixed resolution -- "CURRENTRELEASE" would be Red Hat Linux 6.2.

Comment 15 Gene Czarcinski 2000-08-23 15:27:35 UTC

On my dual PII system (only one tested so far), the nfslock shutdown FAILS is
still there.

Comment 16 Doug Ledford 2000-08-23 15:58:41 UTC

Watch the dates in the notes.  I pointed out that the fix was checked into CVS
on 8/22.  That's *after* RC2 so obviously the problem *would* still exist in
RC2, so no reason to reopen the bug.