Bug 126970

Summary:	lock_gulmd seg faults if given invalid server name
Product:	[Retired] Red Hat Cluster Suite	Reporter:	Corey Marthaler <cmarthal>
Component:	gfs	Assignee:	michael conrad tadpol tilstra <mtilstra>
Status:	CLOSED CURRENTRELEASE	QA Contact:	Cluster QE <mspqa-list>
Severity:	medium	Docs Contact:
Priority:	medium
Version:	3
Target Milestone:	---
Target Release:	---
Hardware:	i686
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2004-09-14 21:54:09 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 Corey Marthaler 2004-06-29 21:29:56 UTC

From Bugzilla Helper:
User-Agent: Mozilla/5.0 (compatible; Konqueror/3.1; Linux)

Description of problem:
[root@morph-01 root]# lock_gulmd -s morph-foo,morph-03,morph-05
Segmentation fault


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


How reproducible:
Always

Steps to Reproduce:
see above

Additional info:

Comment 1 michael conrad tadpol tilstra 2004-07-09 16:29:59 UTC

fixed.
ifelseifelseifelse doesn't work when you leave out the elses.

Comment 2 Corey Marthaler 2004-09-02 22:09:51 UTC

Lock_gulm asserts now:

[root@morph-02 root]# lock_gulmd -s morph-foo,morph-03,morph-05
I cannot find the name for ip "morph-foo".
gf->node_cnt = 0
In src/config_main.c:332 (DEVEL.1093621899) death by:
ASSERTION FAILED: gf->node_cnt > 0 && gf->node_cnt < 5 && gf->node_cnt
!= 2
I cannot find the name for ip "morph-foo".
gf->node_cnt = 0
In src/config_main.c:332 (DEVEL.1093621899) death by:
ASSERTION FAILED: gf->node_cnt > 0 && gf->node_cnt < 5 && gf->node_cnt
!= 2
[root@morph-02 root]# I cannot find the name for ip "morph-foo".
gf->node_cnt = 0
In src/config_main.c:332 (DEVEL.1093621899) death by:
ASSERTION FAILED: gf->node_cnt > 0 && gf->node_cnt < 5 && gf->node_cnt
!= 2

Comment 3 michael conrad tadpol tilstra 2004-09-08 18:56:58 UTC

which is technically correct, but I prettied it up anyways.  cvs up &&
make, better?

Comment 4 Corey Marthaler 2004-09-08 21:58:42 UTC

better.  
 
However... :)  
It could be even better if the cmdline parsing also happened before 
the forking of core, LT, and LTPX in main_main.c and if  
parse_cmdline_servers returned an error (for invalid node, null 
server list...). That way it could be checked and could exit right 
away before daemonizing and then the user would have the exit code 
and error right away instead of getting it at random times from the  
three existing daemons.

Comment 5 michael conrad tadpol tilstra 2004-09-13 22:10:02 UTC

should have this now. give'er'a'go.

Comment 6 Corey Marthaler 2004-09-14 21:54:09 UTC

I like it. :)