[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

"arpresolve: can't allocate llinfo" message with alias IP address on June 20th snapshot



A ping at the console to an alias IP address of a local NIC, does not 
succeed.

# ping -c2 192.168.222.201
PING 192.168.222.201 (192.168.222.201): 56 data bytes
--- 192.168.222.201 ping statistics ---
2 packets transmitted, 0 packets received, 100.0% packet loss

It generates the following messages:

Jun 21 03:47:07 tantalus /bsd: arpresolve: can't allocate llinfo
Jun 21 03:48:35 tantalus last message repeated 4 times
Jun 21 03:54:26 tantalus /bsd: arp_rtrequest: bad gateway value
Jun 21 03:54:52 tantalus /bsd: arpresolve: can't allocate llinfo

Pings from other boxes in the lan to the alias IP address succeed,

The snapshot details:

OpenBSD 3.5-current (GENERIC) #165: Sun Jun 20 12:46:59 MDT 2004
    deraadt_(_at_)_i386_(_dot_)_openbsd_(_dot_)_org:/usr/src/sys/arch/i386/compile/GENERIC

An older snapshot on another box also shows the same behaviour: 

OpenBSD 3.5-current (GENERIC) #132: Wed Jun 9 14:45:57 MDT 2004
    deraadt_(_at_)_i386_(_dot_)_openbsd_(_dot_)_org:/usr/src/sys/arch/i386/compile/GENERIC

# cat /etc/hostname.fxp0
inet 192.168.222.200 255.255.255.0 NONE
inet alias 192.168.222.201 255.255.255.255

# ifconfig fxp0
fxp0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
        address: 00:08:c7:72:45:55
        media: Ethernet autoselect (100baseTX full-duplex)
        status: active
        inet 192.168.222.200 netmask 0xffffff00 broadcast 192.168.222.255
        inet6 fe80::208:c7ff:fe72:4555%fxp0 prefixlen 64 scopeid 0x1
        inet 192.168.222.201 netmask 0xffffffff broadcast 192.168.222.201

# netstat -rn -f inet
Routing tables

Internet:
Destination        Gateway            Flags     Refs     Use    Mtu  
Interface
default            192.168.222.10     UGS         0        0      -   fxp0
127/8              127.0.0.1          UGRS        0        0  33224   lo0
127.0.0.1          127.0.0.1          UH          2        0  33224   lo0
192.168.222/24     link#1             UC          3        0      -   fxp0
192.168.222.10     0:8:c7:89:10:c     UHLc        1        7      -   fxp0
192.168.222.111    0:8:c7:52:3e:78    UHLc        1     1679      -   fxp0
192.168.222.120    0:10:5a:b0:5b:4e   UHLc        0       25      -   fxp0
192.168.222.200    127.0.0.1          UGHS        0        0  33224   lo0
192.168.222.201    link#1             UC          0        4      -   fxp0
224/4              127.0.0.1          URS         0        0  33224   lo0

As a workaround I have added the following line to the 
"/etc/hostname.fxp0" file: 

        !route change 192.168.222.201 127.0.0.1

Pings at the console or via ssh, to the alias 192.168.222.201 succeed now. 
The box runs 'tinydns' configured to answer queries via the same alias IP.
Before the application of the workaround, tinydns did answer dns 
queries to remote client, although DNS queries at the console failed, 
just like the ping.

Netstat output after applying the "route change":

# netstat -rn -f inet                                                           
Routing tables

Internet:
Destination        Gateway            Flags     Refs     Use    Mtu  
Interface
default            192.168.222.10     UGS         0        0      -   fxp0
127/8              127.0.0.1          UGRS        0        0  33224   lo0
127.0.0.1          127.0.0.1          UH          2        4  33224   lo0
192.168.222/24     link#1             UC          2        0      -   fxp0
192.168.222.10     0:8:c7:89:10:c     UHLc        1        9      -   fxp0
192.168.222.111    0:8:c7:52:3e:78    UHLc        1       97      -   fxp0
192.168.222.200    127.0.0.1          UGHS        0        0  33224   lo0
192.168.222.201    127.0.0.1          UC          1       10  33224   lo0
224/4              127.0.0.1          URS         0        0  33224   lo0



Adriaan - Netherlands