SUMMARY: init process looping / Sol 8

From: Geoff Lane <zzassgl_at_buffy.sighup.org.uk>
Date: Wed Oct 08 2003 - 03:50:09 EDT
As I did mention PAM, I should have pointed out that the PAM config on the
system in question is unchanged by us since installation.

A few useful answers received...

First, Casper tells us that the only solution for a looping init process is
a reboot.  That's disappointing but unexpected.  As we have a 24 CPU system
and it's currently not busy we can live with the situation for now

Christopher Wilkinson asked if we have SUN Resource Manager 1.2 installed? 
He'd had problems with patch 108993-18 which edits pam.conf. This is
interesting as we do have SRM installed but it's currently disabled. However
we only have up to patch 108993-13 installed.  I've been planning to remove
SRM, this is a good reason to do it as soon as possible.

S. Borkov suggested that an incorrectly edited /etc/inittab may be the cause
of the missing console login prompt.  Our /etc/inittab hasn't been altered
in over three years.

Finally, thanks for all the "I out of the office" notices and the invitation
to visit an anti-spam web site to confirm that I'm not a spammer.  The
_second_ posting of the question didn't come from me; from the headers it
looks like the original bounced around igatecorp.com and mascotsystems.com
and ended up being sent back to the list.

-------- Original question ---------------

I have a multi-cpu system where the init process is looping and taking 100%
of one CPU. We use NIS+.  pstack implies that init is looping following some
PAM operation.  Normal network login is OK but console login is not possible
- no prompt is provided. I've tried "telinit q" and "kill -HUP" but init
ignores both. Some additional info is provided below.

Is there anything other than a reboot that I can try to clear this problem?

Thanks.
----------------------------------------------------------
SunOS xxx 5.8 Generic_108528-15 sun4u sparc SUNW,Ultra-Enterprise

# pstack 1
1:      /etc/init -
 00058c14 ???????? (ff2d01c8, 4, ff2d0448, ff2d04cc, c, ff3a159c)
 ff3b55ec lookup_sym (ff3a0e0c, 2ba, ff3a0218, 0, 3ffba92, ff3a159c) + 1ac
 ff3c6b58 elf_bndr (2ba, 11d, ff3a1680, ff3e86b0, 0, d8c) + 140
 ff3b2c04 elf_rtbndr (ff14b408, 0, 0, 0, 0, 0) + 10
 ff19e630 ???????? (ff2be570, ff2be570, 0, ff19ce9c, 0, 0)
 ff14b408 _libnsl_fini (ff1a5b94, ff3e8124, 0, ff3e86b0, 2e6f8, ff3c0700) +
2c
 ff17f4e8 _fini    (0, ff3e86b0, ff3e80c8, f, 0, 0) + 4
 ff3bcda4 call_fini (ff3e8174, ff3e8154, ff3e86b0, ff2d0280, ff3a159c,
800000) + f0
 ff3c4388 remove_hdl (ff3e9d78, 0, ff2d04cc, ff2d0200, ff3a0218, ff2d0274) +
a44 ff3c48fc remove_hdl (ff3e9d78, 0, ff2d04cc, ff3e9d14, ff3a0218,
ff3a1d00) + fb8 ff3c14a8 dlclose_intn (ff2d08dc, 0, ff3e86b0, ff3e80c8,
ff3a0218, ff3e80c8) + 8c
 ff3c15b0 dlclose  (ff2d08dc, ff3e9d78, 0, ff3a0218, 1, 0) + 58
 00059868 pam_end  (0, 0, a61e0, 0, 13, 13) + b4
 00023dd4 notify_pam_dead (a0380, 0, ffffffff, 0, 0, 0) + a0
 00023f94 account  (8, a0380, 8, 8, 0, ffbefc79) + 1b0
 00021f40 remv     (a2a9c, 5, 9c218, a23ec, 9c294, 0) + 3bc
 0002150c main     (9c000, a2800, a2000, a2800, 1, 9c294) + 494
 00021060 _start   (0, 0, 0, 0, 0, 0) + b8
#
# truss -p 1 2>&1 | more
    Received signal #11, SIGSEGV [caught]
      siginfo: SIGSEGV SEGV_MAPERR addr=0x0000002C
setcontext(0xFFBEEF98)
    Incurred fault #6, FLTBOUNDS  %pc = 0xFF3B76B8
      siginfo: SIGSEGV SEGV_MAPERR addr=0x0000002C
    Received signal #11, SIGSEGV [caught]
      siginfo: SIGSEGV SEGV_MAPERR addr=0x0000002C
setcontext(0xFFBEEF98)
    Incurred fault #6, FLTBOUNDS  %pc = 0xFF3B76B8
      siginfo: SIGSEGV SEGV_MAPERR addr=0x0000002C
    Received signal #11, SIGSEGV [caught]
      siginfo: SIGSEGV SEGV_MAPERR addr=0x0000002C
setcontext(0xFFBEEF98)
    Incurred fault #6, FLTBOUNDS  %pc = 0xFF3B76B8
      siginfo: SIGSEGV SEGV_MAPERR addr=0x0000002C
    Received signal #11, SIGSEGV [caught]
      siginfo: SIGSEGV SEGV_MAPERR addr=0x0000002C
setcontext(0xFFBEEF98)
    Incurred fault #6, FLTBOUNDS  %pc = 0xFF3B76B8
      siginfo: SIGSEGV SEGV_MAPERR addr=0x0000002C
    Received signal #11, SIGSEGV [caught]
      siginfo: SIGSEGV SEGV_MAPERR addr=0x0000002C
[ad infinitum]

-- 
Geoff Lane
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
Received on Wed Oct 8 03:50:05 2003

This archive was generated by hypermail 2.1.8 : Thu Mar 03 2016 - 06:43:22 EST