Bug 466044

Summary: all maps in a multi-map entry should be cached with the same time
Product: Red Hat Enterprise Linux 4 Reporter: Jeff Moyer <jmoyer>
Component: autofsAssignee: Jeff Moyer <jmoyer>
Status: CLOSED ERRATA QA Contact: Martin Jenner <mjenner>
Severity: medium Docs Contact:
Priority: medium    
Version: 4.8CC: cward, duck, ikent, jrfuller, kvolny, tao
Target Milestone: rcKeywords: OtherQA
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
* the multi-map lookup code contained a race that could result in automount reading two maps in quick succession. Reading the second map invalidated the cache entries created by the first map, and in some configurations, this led to problematic or nonsensical results such as duplicate keys, or a wildcard key in the first map not present in subsequent maps. This would result in failed lookups for valid automount keys. The lookup code now passes time into the lookup_ghost functions, and therefore avoids incorrectly clearing the cache. Automount now reliably looks up valid keys.
 Story Points: ---
Clone Of: Environment:
Last Closed: 2009-05-18 20:22:00 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:
Attachments:
Description Flags
Pass time into the lookup_ghost functions none

Description Jeff Moyer 2008-10-07 20:58:52 UTC 
Description of problem:
There is a slight window where we could have one map in a multimap entry read a second after the previous map, thus invalidating all of the cache entries created by the previous map.  This results in incorrect results in some configurations, such as duplicate keys, or a wildcard key in the first map but not in subsequent maps.  This has been tripped over by at least one customer.

I will attempt to create an RHTS test for this, though I think it will be difficult to trigger.
Comment 4 Jeff Moyer 2008-10-30 14:19:51 UTC 
Created attachment 321937 [details]
Pass time into the lookup_ghost functions

This patch should address the problem of incorrectly clearing the cache from one module in a multi-map configuration.
Comment 9 Jeff Moyer 2009-01-21 15:53:01 UTC 
A fix for this is available in autofs versions 4.1.3-235.4 and later.
Comment 11 Ruediger Landmann 2009-01-23 06:55:01 UTC 
Release note added. If any revisions are required, please set the 
"requires_release_notes" flag to "?" and edit the "Release Notes" field accordingly.
All revisions will be proofread by the Engineering Content Services team.

New Contents:
* the multi-map lookup code contained a race that could result in automount reading two maps in quick succession. Reading the second map invalidated the cache entries created by the first map, and in some configurations, this led to problematic or nonsensical results such as duplicate keys, or a wildcard key in the first map not present in subsequent maps. This would result in failed lookups for valid automount keys. The lookup code now passes time into the lookup_ghost functions, and therefore avoids incorrectly clearing the cache. Automount now reliably looks up valid keys.
Comment 16 errata-xmlrpc 2009-05-18 20:22:00 UTC 
An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on therefore solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.

http://rhn.redhat.com/errata/RHBA-2009-0991.html