Anyone seen this:
   
     PID USERNAME PRI NICE   SIZE   RES STATE   TIME   WCPU    CPU COMMAND
     19539 root       2    0  7540K 7728K sleep   2:50 18.41% 18.41% sendmail
        92 root      18    0    12K  112K sleep   1:12  0.06%  0.05% update
     19185 root      18    0    11M   12M sleep   4:29  0.00%  0.00% <sendmail>
     18913 root      18    0    11M   11M sleep   4:18  0.00%  0.00% <sendmail>
     18587 root      18    0    11M   11M sleep   4:12  0.00%  0.00% <sendmail>
     18166 root      18    0    11M  788K sleep   4:06  0.00%  0.00% <sendmail>
     14999 root      18    0    15M  464K sleep   6:30  0.00%  0.00% <sendmail>
     17692 root      18    0    11M  460K sleep   3:58  0.00%  0.00% <sendmail>
     17177 root      18    0    10M  460K sleep   3:44  0.00%  0.00% <sendmail>
     16751 root      18    0    10M  460K sleep   3:34  0.00%  0.00% <sendmail>
     16317 root      18    0    10M  460K sleep   3:09  0.00%  0.00% <sendmail>
     15425 root      18    0  6088K  456K sleep   1:43  0.00%  0.00% <sendmail>
     14624 root      18    0  1100K  452K sleep   0:07  0.00%  0.00% <sendmail>
     12496 root      18    0  1008K  452K sleep   0:05  0.00%  0.00% <sendmail>
   
   What on earth is going on?  How do I stop this?  This is happening under 1.2
   on i386.  I'm running out of RAM and swap.....!

i didn't see this under 1.2 but i did see it under -current from
about 2 months ago.  i had to reboot to stop sendmail from losing
like this -- every connection to sendmail would send it off into
a loop of some sort, where it would use as much cpu as possible.

never had a clue what the cause was.