lan standby issue

[bitou]

hi,

we have a problem in our cluster and i need your help please, the lan2 is down

  NODE         STATUS       STATE

  hf1      up           running

Network_Parameters:
INTERFACE    STATUS       PATH                NAME
    PRIMARY       up           1/0/8/2/0/4/0        lan0
    PRIMARY       up           1/0/9/2/0             lan1

    STANDBY      down         0/0/2/1/0           lan2
    STANDBY      up           0/0/1/1/0/4/0       lan3

and from node 2

Network_Parameters:

    INTERFACE    STATUS       PATH                NAME

    PRIMARY       up           1/0/8/2/0/4/0        lan0
    PRIMARY       up           1/0/9/2/0             lan1

i checked cables, and ports and it seems that are working fine.

and also /etc/netconfig---> no error message and lan 2 is claimed from ioscan.

how can we bring the lan2 to up? thanks for your replay.

[Naj]

Hi,

How about lanscan output?

[bitou]

from lanscan i see all lan is up.

[Turgay Cavdar]

Please send the output of the following :

# lanadmin -x 2

Regards,,,

[bitou]

hi,

the command lanadmin -x2 will display a hardware problem if exists?

another question plz how can i check if the network card is faulty or cable or the port of the switch

is there any  parameters to check from lanadmin

PPA Number                      = 2
Description                     = lan2 HP PCI Core I/O 1000Base-T Release B.11.23.0712.01
Type (value)                    = ethernet-csmacd(6)
MTU Size                        = 1500
Speed                           = 1000000000
Station Address                 = 0x14c2650091
Administration Status (value)   = up(1)
Operation Status (value)        = up(1)
Last Change                     = 419
Inbound Octets                  = 120454615
Inbound Unicast Packets         = 863761
Inbound Non-Unicast Packets     = 4327
Inbound Discards                = 0
Inbound Errors                  = 0
Inbound Unknown Protocols       = 12
Outbound Octets                 = 145033817
Outbound Unicast Packets        = 1285500
Outbound Non-Unicast Packets    = 221
Outbound Discards               = 0
Outbound Errors                 = 0
Outbound Queue Length           = 0
Specific                        = 655367

[Stephen Doud]

Though Serviceguard does not use linkloop to test NIC availability, it is useful when trying to identify bridged-net issues.

Try:
$  linkloop -i  2 and <MAC for lan1>

Example:

# linkloop -i 2 0x000E7F4FD095 
Link connectivity to LAN station: 0x000E7F4FD095
 -- OK

If linkloop fails, resolve why lan1 and lan2 are not on the same physical network.  Then  check cmviewcl output (and syslog.log)..

[Steven E. Protter]

Shalom,

I would examine the package logs as well.

Might be a good idea to use xstm or mstm or cstm and check the state of the hardware.

I'd like to see instead of menu output:

lanadmin -x 2

dmesg

SEP

[Stephen Doud]

Serviceguard marks a NIC down when it cannot complete a link-layer communication between the NIC and another NIC on the same server, same bridged network, or other node, same bridged network.

Use ' linkloop -i' between lan2 and each of the primary NICs. If both fail, lan2 may be detached from the common network. If it succeeds, determine whether it has been assigned an IP (it's a STBY LAN - it should not be) or bogus netmask.