[Bug 997096] Re: nscd segmentation fault

[vhp]

Faced a similar issue with the newest version of nscd.  Users in
netgroups would be able to use sudo one time.  Once authed the nscd
process would die for all users.  `/etc/init.d/nscd status` showed it
was not running

> /etc/init.d/nscd status
 * Status of Name Service Cache Daemon service:                                                                                                                                                * not running.

syslog showed the following

May 22 11:46:31 default kernel: [84228.494446] nscd[28565]: segfault at 0 ip 00007f3400ed03e1 sp 00007f33f7d2f108 error 4 in libc-2.15.so[7f3400d6f000+1b3000]
May 22 11:53:18 default nscd: 29514 cannot stat() file `/etc/netgroup': No such file or directory
May 22 11:53:29 default nscd: 29558 cannot stat() file `/etc/netgroup': No such file or directory
May 22 11:55:19 default nscd: 29991 cannot stat() file `/etc/netgroup': No such file or directory
May 22 12:01:38 default nscd: 30481 cannot stat() file `/etc/netgroup': No such file or directory
May 22 12:02:03 default kernel: [85160.598268] nscd[30489]: segfault at 0 ip 00007f794433a3e1 sp 00007f793af98108 error 4 in libc-2.15.so[7f79441d9000+1b3000]
May 22 12:09:42 default nscd: 31130 cannot stat() file `/etc/netgroup': No such file or directory
May 22 12:09:50 default kernel: [85627.475382] nscd[31140]: segfault at 0 ip 00007f57b81ef3e1 sp 00007f57aea4b108 error 4 in libc-2.15.so[7f57b808e000+1b3000]
May 22 12:17:01 default CRON[31512]: (root) CMD (   cd / && run-parts --report /etc/cron.hourly)
May 22 12:20:39 default kernel: [86276.573029] nscd[31569]: segfault at 0 ip 00007f0a8626b3e1 sp 00007f0a7cac7108 error 4 in libc-2.15.so[7f0a8610a000+1b3000]

To resolve I removed the following from the bottom on /etc/nscd.conf


       enable-cache            netgroup        yes
       positive-time-to-live   netgroup        28800
       negative-time-to-live   netgroup        20
       suggested-size          netgroup        211
       check-files             netgroup        yes
       persistent              netgroup        yes
       shared                  netgroup        yes
       max-db-size             netgroup        33554432

Just removing these stopped the slow authentication time and more
importantly stopped the segfaults.  Users in netgroups can now use sudo
and auth as many times as they like.  There is still a bug to fix here.
Hopefully this workaround can help some until patched.

-- 
You received this bug notification because you are a member of Ubuntu
Foundations Bugs, which is subscribed to eglibc in Ubuntu.
https://bugs.launchpad.net/bugs/997096

Title:
  nscd segmentation fault

Status in “eglibc” package in Ubuntu:
  Confirmed

Bug description:
  nscd segfaults a few seconds after every getent netgroup requests:

  root at cmo-cluster2:~# nscd -d 
  Wed 09 May 2012 12:59:54 PM UTC - 13953: register trace file /etc/passwd for database passwd
  Wed 09 May 2012 12:59:54 PM UTC - 13953: register trace file /etc/group for database group
  Wed 09 May 2012 12:59:54 PM UTC - 13953: register trace file /etc/services for database services
  Wed 09 May 2012 12:59:54 PM UTC - 13953: register trace file /etc/netgroup for database netgroup
  Wed 09 May 2012 12:59:54 PM UTC - 13953: cannot stat() file `/etc/netgroup': No such file or directory
  Wed 09 May 2012 12:59:56 PM UTC - 13953: handle_request: request received (Version = 2) from PID 13962
  Wed 09 May 2012 12:59:56 PM UTC - 13953: 	GETFDNETGR
  Wed 09 May 2012 12:59:56 PM UTC - 13953: handle_request: request received (Version = 2) from PID 13962
  Wed 09 May 2012 12:59:56 PM UTC - 13953: 	GETNETGRENT (cmo-nfs-netgroup)
  Wed 09 May 2012 12:59:56 PM UTC - 13953: Haven't found "cmo-nfs-netgroup" in netgroup cache!
  Wed 09 May 2012 12:59:56 PM UTC - 13953: provide access to FD 11, for netgroup
  Segmentation fault (core dumped)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/eglibc/+bug/997096/+subscriptions