Kprop Eating Disk Space?

[Monica Lau]

I am using Kerberos version 5, release 1.2.4.  The krb5kdc and kadmind processes are not running at the slave kdc.  Here is the output of the "ps ax | more" command (from the slave kdc) when I don't have the propagation going on: 
  PID TTY      STAT   TIME COMMAND
    1 ?        S      0:08 init
    2 ?        SW     0:00 [keventd]
    4 ?        SWN    0:00 [ksoftirqd_CPU0]
    5 ?        SW     0:00 [kswapd]
    6 ?        SW     0:00 [bdflush]
    7 ?        SW     0:22 [kupdated]
    8 ?        SW     0:00 [mtdblockd]
    9 ?        SWN    0:40 [jffs2_gcd_mtd3]
   35 ?        S      0:00 /usr/sbin/cm -d
   37 ?        S      0:00 syslogd
   39 ?        S      0:08 /bin/sh /etc/rc.d/dmesgd /kern.log 10
   48 ?        S      0:01 inetd /etc/inetd.conf
   65 ?        S      0:00 /sbin/cardmgr
   80 ttySA0   S      0:00 -bash
  126 ?        S      0:00 dhclient eth0
  542 ?        S      0:01 telnetd
  543 ttyp0    S      0:00 /bin/sh /usr/bin/login -h 157.235.91.7 -p
  544 ttyp0    S      0:00 -sh
  554 ttyp0    S      0:01 -bash
 3911 ?        S      0:00 sleep 10
 3914 ttyp0    R      0:00 ps ax
 3915 ttyp0    S      0:00 more  
Here is the output of the "ps ax | more" command (from the slave kdc) when the propagation is going on: 
    1 ?        S      0:08 init
    2 ?        SW     0:00 [keventd]
    4 ?        RWN    0:00 [ksoftirqd_CPU0]
    5 ?        SW     0:00 [kswapd]
    6 ?        SW     0:00 [bdflush]
    7 ?        RW     0:18 [kupdated]
    8 ?        SW     0:00 [mtdblockd]
    9 ?        RWN    0:36 [jffs2_gcd_mtd3]
   35 ?        S      0:00 /usr/sbin/cm -d
   37 ?        S      0:00 syslogd
   39 ?        S      0:07 /bin/sh /etc/rc.d/dmesgd /kern.log 10
   48 ?        S      0:01 inetd /etc/inetd.conf
   65 ?        S      0:00 /sbin/cardmgr
   80 ttySA0   S      0:00 -bash
  126 ?        S      0:00 dhclient eth0
  542 ?        S      0:01 telnetd
  543 ttyp0    S      0:00 /bin/sh /usr/bin/login -h 157.235.91.7 -p
  544 ttyp0    S      0:00 -sh
  554 ttyp0    S      0:01 -bash
 3867 ?        S      0:00 sleep 10
 3874 ttyp0    R      0:00 ps ax
 3875 ttyp0    S      0:00 more
 3876 ?        D      0:00 kpropd 
Does this problem exist in the 1.2.4 version?  Should I try the latest version and see what happens?  Thanks for your help. 
Monica 

 Ezra Peisach <epeisach at MIT.EDU> wrote:
I suspect that some process is keeping the old database open when you
replace it with the new one generated by kprop.... 

You did not indicate which version of kerberos you are using - so
we do not know what could be holding the database open. Does the
disk usage go down if you kill off the kdc? kadmind? Essentially - which
process is keeping the old database open - which when 
killed - releases the old on disk copy and the O/S cleans it up...

Ezra
_______________________________________________
krbdev mailing list krbdev at mit.edu
https://mailman.mit.edu/mailman/listinfo/krbdev


---------------------------------
Do you Yahoo!?
Yahoo! Web Hosting - establish your business online
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/pipermail/krbdev/attachments/20030311/3b7c0e9f/attachment.htm