Hello,

This behavior is apparent with both Linux kernels 2.2.14 and
2.3.99-pre6.3 on the DEC (Compaq) Alpha.  This box is acting as a MASQ
gateway (ipchains; iptables -t nat) and our ISP is preparing to change
out subnet this weekend (or sometime thereabouts).  Not a problem says
I; I'll just add an alias and new default route, and let it failover.

All worked fine the remainder of the day, but today things were not all
well.  None of the masqueraded boxes were able to access beyond the
gateway; however, the gateway box itself had no problems going to either
side.  It is not an ISP problem, and I can narrow it down to this
scenario:

# route
Kernel IP routing table
Destination     Gateway         Genmask         Flags Metric Ref    Use
Iface
localnet        *               255.255.255.0   U     0      0        0
eth1
209.101.87.0    *               255.255.255.0   U     0      0        0
eth0
default         CISCO.SPACEY.NE 0.0.0.0         UG    0      0        0
eth0
default         CISCO.SPACEY.NE 0.0.0.0         UG    0      0        0
eth0

# ifconfig
eth0      Link encap:Ethernet  HWaddr 08:00:2B:E4:32:7E  
          inet addr:209.101.87.28  Bcast:209.101.87.255 
Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:190133 errors:0 dropped:0 overruns:0 frame:0
          TX packets:35699 errors:0 dropped:0 overruns:0 carrier:0
          collisions:46 txqueuelen:100 
          Interrupt:10 Base address:0x8800 

eth0:1    Link encap:Ethernet  HWaddr 08:00:2B:E4:32:7E  
          inet addr:207.168.50.156  Bcast:207.168.50.255 
Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          Interrupt:10 Base address:0x8800 

eth1      Link encap:Ethernet  HWaddr 00:50:04:A8:DB:9E  
          inet addr:192.168.68.252  Bcast:192.168.68.255 
Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:2180626 errors:2 dropped:0 overruns:0 frame:2
          TX packets:139768 errors:0 dropped:0 overruns:0 carrier:0
          collisions:33 txqueuelen:100 
          Interrupt:15 Base address:0x9000 

If I "ifconfig eth0:1 up", MASQ stops working; if I "ifconifg eth0:1
down", it begins working immediately.  I don't know if this is a bug
that needs to be reported, or if it is simply my own misunderstanding,
in which case enlightenment would be appreciated.

Thanks,
Christopher
-
To unsubscribe from this list: send the line "unsubscribe linux-net" in
the body of a message to [EMAIL PROTECTED]

Reply via email to