Re: rm/csup/svn/ldd make host unresponsive [WAS: Re: ldd leaves the machine unresponsive]

Marcel

On Sun, Mar 21, 2010 at 06:22:14PM +0000, Anton Shterenlikht wrote:
> 
> An update:
> 
> 1. reinstalled from 8.0-CURRENT-200906
> 
> 2. installed the ports tree via csup(1)
> 
> 3. installed svn(1) from ports
> 
> 4. updated src with svn.
> 	Both svn and csup worked fine here.
> 
> 5. rebuilt and reinstalled kernel and world as
>    usual to r205403.
> 
> 6. rebooted.
> The kernel config file:
> 	http://seis.bris.ac.uk/~mexas/freebsd/ia64/rx2600/uzi/UZI 
> 
> dmesg:
> 	http://seis.bris.ac.uk/~mexas/freebsd/ia64/rx2600/uzi/dmesg.boot
> 
> ifconfig -a:
> 	http://seis.bris.ac.uk/~mexas/freebsd/ia64/rx2600/uzi/ifconfig-a
> 
> 
> 7. tried to update the src again with svn and got stuck.
> 	All I can issue is CTRL/T, which shows for svn:
> 
> mech-as221# svn co svn://svn.freebsd.org/base/head/ /usr/src/
> 
> load: 0.00  cmd: svn 888 [biord] 8008.53r 0.09u 0.30s 0% 13992k
> load: 0.00  cmd: svn 888 [biord] 8009.53r 0.09u 0.30s 0% 13992k
> load: 0.00  cmd: svn 888 [biord] 8015.07r 0.09u 0.30s 0% 13992k
> 
> in another ssh session I was running gstat(8) which showed
> zero activity in the disk.
> 
> and in yet another ssh session I tried to launch top:
> 
> mech-as221# top
> load: 0.00  cmd: csh 915 [ufs] 6146.33r 0.00u 0.00s 0% 5008k
> load: 0.00  cmd: csh 915 [ufs] 6147.15r 0.00u 0.00s 0% 5008k
> 
> and on the serial console:
> 
> load: 0.00  cmd: getty 828 [ufs] 8129.90r 0.00u 0.00s 0% 2560k
> load: 0.00  cmd: getty 828 [ufs] 8130.70r 0.00u 0.00s 0% 2560k
> 
> but the shell prompt never appears.
> I've waited maybe 2-3 hours.

On reboot I did
# cd /usr/obj
# chflags -R noschg *
# rm -rf *

I monitor disk activity with iostat(8) and gstat(8),
and system activity with top(1) from 3 separate ssh sessions. 
For about 5-10 sec iostat(8) and gstat(8) show significant
disk activity. After that both iostat and gstat show
zero disk activity. top(1) shows:

[skip]
  PID    UID    THR PRI NICE   SIZE    RES STATE   C   TIME   WCPU COMMAND
   10      0      2 171 ki31     0K    64K RUN     0  28:40 198.00% idle
   11      0     17 -48    -     0K   544K WAIT    0   0:02  0.00% intr
  893      0      1  96    0 12800K  4008K CPU0    0   0:00  0.00% top
  918      0      1  -4    0 11592K  2424K getblk  0   0:00  0.00% rm
[skip]

rm never exits (well.. within 20 minutes).

kill -9 918 (issued from top(1)) makes no effect.
No new ssh logins are possible, and the existing
ssh sessions and the serial line don't show the
shell prompt.

It seems the problems I've had in the last
several days with ldd/csup/svn/rm have the
same root cause.

I'm just not sure if it's someting simple
that I've messed up, or something went wrong
in current..

many thanks for your help
anton

-- 
Anton Shterenlikht
Room 2.6, Queen's Building
Mech Eng Dept
Bristol University
University Walk, Bristol BS8 1TR, UK
Tel: +44 (0)117 331 5944
Fax: +44 (0)117 929 4423