------- You are receiving this mail because: -------
You are on the CC list for the bug.
http://bugs.exim.org/show_bug.cgi?id=722
Roman Odaisky <to.roma.from.eximbug@???> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|RESOLVED |REOPENED
Resolution|WONTFIX |
--- Comment #2 from Roman Odaisky <to.roma.from.eximbug@???> 2008-06-20 09:04:02 ---
That does not solve the problem. ‘Some sensible data’ is still predictable.
Messages arriving over the same connection are likely to have the same values
of $tod_epoch, $pid and other variables, which is hardly good for load
balancing. Also, how would one use ${md5} to choose one of three items from a
list?
Implementing $rand seems very simple, just cloning the code for $pid:
$ fgrep -C 1 pid expand.c
vtype_reply, /* value not used; get reply from headers */
vtype_pid, /* value not used; result is pid */
vtype_host_lookup, /* value not used; get host name */
--
{ "parent_local_part", vtype_stringptr, &deliver_localpart_parent },
{ "pid", vtype_pid, NULL },
{ "primary_hostname", vtype_stringptr, &primary_hostname },
--
case vtype_pid:
sprintf(CS var_buffer, "%d", (int)getpid()); /* pid */
return var_buffer;
and calling srand() somewhere.
WBR, Roman.
--
Configure bugmail:
http://bugs.exim.org/userprefs.cgi?tab=email
