Bug 694410

Summary: Duplication hash should not be computed from package NVR and executable
Product: Red Hat Enterprise Linux 6 Reporter: Karel Klíč <kklic>
Component: abrtAssignee: Jiri Moskovcak <jmoskovc>
Status: CLOSED ERRATA QA Contact: BaseOS QE - Apps <qe-baseos-apps>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 6.2CC: ahecox, dfediuck, dvlasenk, gavin, kklic, mnowak, npajkovs, rvokal, syeghiay
Target Milestone: rc   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: abrt-1.1.16-3.el6 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2011-05-19 13:41:44 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Attachments:
Description Flags
Proposed patch none

Description Karel Klíč 2011-04-07 09:41:05 UTC
Created attachment 490502 [details]
Proposed patch

Description of problem:
Currently the duplicate hash of a crash is computed from the package NVR (name, version, release), path of the executable, and the backtrace hash.

This is very unfortunate because the package NVR is changed with every update, and path of the executable is not a good differentiator. When multiple executables from a single component crash with similar backtrace, it is better to assume that they share code.

Version-Release number of selected component (if applicable):
1.1.16-2

Steps to Reproduce:
With abrt 1.1.16-2, you can run abrtd under gdb and 
step through CAnalyzerCCpp::CreateReport (lib/Plugins/CCpp.cpp) to see how the hash is computed.

With the patch applied, you can run `abrtd -dvvv` and see the fixed hash input on abrtd's standard output.

Actual results:
Duplicate hash is computed from package nvr, executable path, and backtrace hash. Multiple crashes from older and newer package and single bug in the source code have different duplicate hashes.

Expected results:
Duplicate hash is computed from component name and backtrace hash. Multiple crashes from older and newer package and single bug in the source code have the same duplicate hash.

Comment 2 RHEL Program Management 2011-04-07 09:43:45 UTC
Since RHEL 6.1 External Beta has begun, and this bug remains
unresolved, it has been rejected as it is not proposed as
exception or blocker.

Red Hat invites you to ask your support representative to
propose this request, if appropriate and relevant, in the
next release of Red Hat Enterprise Linux.

Comment 3 Jiri Moskovcak 2011-04-07 12:53:20 UTC
This is quite important change for GSS. The hash is used for matching tickets in bugzilla with cases in customer portal. If this is not working it must be done manually and takes a lot of time to GSS engineers.

Comment 5 Andrew Hecox 2011-04-07 20:47:42 UTC
Thanks Jiri/Radek.

this change will allow us to substantially reduce triage and analysis time for roughly 50% of RHEL 6 based cases which are caused by bugs. Please reach out to me if additional detail is needed on business impact.

Comment 7 Jiri Moskovcak 2011-04-12 14:51:15 UTC
To improve the compatibility with the GSS portal we also fixed the problem with user-agent http header. We will test it together with this bug.

Comment 8 Jiri Moskovcak 2011-04-12 15:22:17 UTC
Created a tested case which has the user-agent set properly:

https://access.redhat.com/support/cases/00450063

how to test the new hash is described in comment #1.

Comment 13 errata-xmlrpc 2011-05-19 13:41:44 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-2011-0619.html