Daniel wrote:
The plot thickens

I've found a lot of good information touching on this subject at:

but nothing seems to work!

I am however closer, but I'm getting a very strange routing error I can't figure out. Although I have a default route setup that does not specify a device, apparently ping won't even try to route though it.

Last login: Sun Jan  7 21:59:18 on pts/4
Sun Microsystems Inc.   SunOS 5.10      Generic January 2005
-bash-3.00# netstat -rnv

IRE Table: IPv4
  Destination             Mask           Gateway          Device Mxfrg Rtt   
Ref Flg  Out  In/Fwd
-------------------- --------------- -------------------- ------ ----- ----- 
--- --- ----- ------
XXX.YYY.230.0 XXX.YYY.230.4         bge0:6  1500*    0  
 1 U        0     0           bge0:7  1500*    0   
1 U        0     0         XXX.YYY.230.4         bge0:6  1500*    0   
1 U        0     0
default             XXX.YYY.230.1                 1500*    0   
1 UG       0     0              lo0:2   8232*    0   
1 UH       0     0
-bash-3.00# ping XXX.YYY.226.208
no answer from XXX.YYY.226.208
-bash-3.00# ping XXX.YYY.226.209
ICMP Host Unreachable from gateway  (
 for icmp from  ( to XXX.YYY.226.209

Has anyone seen this before?

Not this particular instance, but I wouldn't be surprised if this is due to CR 6453732 "zones routing check for "next hop reachable from zone" for default routes is flawed".

That bug makes the existence of IRE_CACHE entries for some zone have an impact on routing out of other zones. Thus I wouldn't be surprised if there is some IRE_CACHE entry for XXX.YYY.226.208 which makes those packets be sent somewhere, but not for XXX.YYY.226.209.

FWIW ndd /dev/ip ipv4_ire_status
in the global zone will dump all of the IREs.


zones-discuss mailing list

Reply via email to