Bug 654431

Summary: kexec-tools really breaks /sbin/mkdumprd
Product: [Fedora] Fedora Reporter: Michal Jaegermann <michal>
Component: kexec-toolsAssignee: Neil Horman <nhorman>
Status: CLOSED DUPLICATE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: medium Docs Contact:
Priority: low    
Version: rawhideCC: nhorman, qcai
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2010-11-18 01:35:37 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
gzip compressed traces from an attempt to use /sbin/mkdumprd none

Description Michal Jaegermann 2010-11-17 20:37:59 UTC 
Created attachment 461159 [details]
gzip compressed traces from an attempt to use /sbin/mkdumprd

Description of problem:

After fixing for an umpteenth time bug 635893, just to get anywhere instead of immediately starting an inifinite loop (why this still lingers unfixed?), an attempt to use /sbin/mkdumprd from kexec-tools-2.0.0-41.fc15 gets into another infinite loop.  This time when going through kernel modules.

While trying to trace that shell script /sbin/mkdumprd a 10 minutes wait
produced 123M of trace but this was not getting any closer to a desired
kdump image.  Trying to grep, for example, in this trace for a pattern:
' echo /lib/modules/2.6.36-1.fc15.x86_64/kernel/.*nat'
produced 296 lines.  Seems tad excessive.  Similar with other cases.

An attempt to boot a new kernel with a service kdump active simply gets
stuck.

Version-Release number of selected component (if applicable):
kexec-tools-2.0.0-41.fc15

How reproducible:
Every time.
Comment 1 Neil Horman 2010-11-18 01:35:37 UTC 
Its pointless to open a new bug that does nothing but gripe, please don't waste your time or mine.  bz 635893 is open to track this, and I'll get to it when I can.

*** This bug has been marked as a duplicate of bug 635893 ***
Comment 2 Michal Jaegermann 2010-11-18 07:52:27 UTC 
(In reply to comment #1)
> Its pointless to open a new bug that does nothing but gripe,

In case you failed to notice, despite that I wrote explicitely "another infinite loop", this looks like a new bug.  The problem in bug 635893 is an infinite loop but not the same one and whatever you can find in comments there does not fix a loop from _this_ report.

If I would have a patch for that one I would attach it but I got lost in this huge trace and I do not know what really changed.  I did not try very long but I failed to work around here.  It is up to you where you want to track it but I was told a number of times that different bugs should NOT be conflated in the same report.  I am not sure what "nothing but gripe" refers too.