[hacknews] overload response improved

henry@utzoo.UUCP (Henry Spencer) (06/26/85)

Now running a new /unix.  This one attempts to give a better (in some sense)
response to inode/file-table overflow, by waiting for empty slots rather
than failing immediately.  For our current environment (occasional transient
overflows), this is probably the preferred strategy.

As side issues, some minor internal improvements have been made, and
the core-dumper routine has been modified to make core dumps private
(i.e. rw------- rather than rw-rw-r-- or whatever).
-- 
				Henry Spencer @ U of Toronto Zoology
				{allegra,ihnp4,linus,decvax}!utzoo!henry