Bug 73831

Summary: nis group bygid lookup not working properly in 8.0 final
Product: [Retired] Red Hat Linux Reporter: Matt Wilson <msw>
Component: glibcAssignee: Jakub Jelinek <jakub>
Status: CLOSED CURRENTRELEASE QA Contact: Brian Brock <bbrock>
Severity: medium Docs Contact:
Priority: medium    
Version: 8.0CC: fweimer
Target Milestone: ---   
Target Release: ---   
Hardware: i386   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2003-10-09 21:26:08 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 Matt Wilson 2002-09-11 20:36:37 UTC
[msw@sid msw]$ id msw
uid=2195(msw) gid=2195(msw) groups=2195(msw),1076,1070,4001(idevel)

notice that group 1070 did not get resolved to a name.  Only the primary group
and the last supplimental group have names resolved.  The server has the needed
info:

[msw@sid msw]$ ypmatch 1070 group.bygid
devel:x:1070:{list of members}

[msw@sid msw]$ ypmatch devel group.byname
devel:x:1070:{list of members}

It seems that gr_name is not getting properly filled in by the nis module.

Comment 1 Roland McGrath 2002-09-11 20:43:12 UTC
I will set up a testing NIS server at home and debug this today.

Comment 2 Nalin Dahyabhai 2002-09-11 22:44:21 UTC
No idea if this is useful information or not, but it appears to work with nscd
running.

Comment 3 Roland McGrath 2002-09-12 02:57:19 UTC
The bug was with very long results that cause expansion of the initial buffer.
For future testing reference, be sure to have some such cases in the NIS test
environment (e.g. group lines with many members--group 1070's line is 719 chars
long.

It so happens this is the same underlying bug that was reported today in
a different context (it bites all sorts of get* functions with long results).
I had already fixed it before that evil bastard msw made me risk my immortal
soul by setting up NIS at home.  It didn't occur to me that it was probably the
same bug until after my fingers were fouled with that nasty yellow newsprint.
Now I'm screwed.

The fix is in glibc mainline already.  Jakub should catch it in his next
sweep of miscellaneous fixes into the RH rpm version.


Comment 4 Matt Wilson 2003-10-09 21:26:08 UTC
looks good now