lanscan command error

[MikeL_4]

I have two servers who make_net_recovery failed last night with:
ERROR: "lanscan" command did not return lanic id's for the system.

When I checked for some reason the lanscan command does not return and lan information:

root:=>/usr/sbin/lanscan

The lanscan executable is the same on all servers, the ones working and the 2 not working:
root:=>ls -al /usr/sbin/lanscan
-r-xr-sr-x 1 root sys 36864 Jun 12 2003 /usr/sbin/lanscan

any ideas???

[Raj D. · Accepted solution]

Hi Mike ,

lanscan command looks good .
You can also try to check the checksum of the file :
Here is the ckecksum:

[ 826859478 36864 /usr/sbin/lanscan ]

Also Check your lancards status with :
# ioscan -fnC lan



Cheers,
Raj.

[MikeL_4]

Is checksum not normally on a server?? I try to execute it and get not found?

All the LAN interfaces are there, the two servers are in a Service guard cluster and the packages are up and running normally on the correct server and lan:
root:=>ioscan -fnC lan
Class I H/W Path Driver S/W State H/W Type Description
========================================================================
lan 0 0/1/2/0 igelan CLAIMED INTERFACE HP A6825-60101 PCI 1000Base-T Adapter
lan 1 0/2/1/0/4/0 btlan CLAIMED INTERFACE HP A5506B PCI 10/100Base-TX 4 Port
/dev/diag/lan1 /dev/ether1 /dev/lan1
lan 2 0/2/1/0/5/0 btlan CLAIMED INTERFACE HP A5506B PCI 10/100Base-TX 4 Port
/dev/diag/lan2 /dev/ether2 /dev/lan2
lan 3 0/2/1/0/6/0 btlan CLAIMED INTERFACE HP A5506B PCI 10/100Base-TX 4 Port
/dev/diag/lan3 /dev/ether3 /dev/lan3
lan 4 0/2/1/0/7/0 btlan CLAIMED INTERFACE HP A5506B PCI 10/100Base-TX 4 Port
/dev/diag/lan4 /dev/ether4 /dev/lan4
lan 5 0/6/1/0/4/0 btlan CLAIMED INTERFACE HP A5506B PCI 10/100Base-TX 4 Port
/dev/diag/lan5 /dev/ether5 /dev/lan5
lan 6 0/6/1/0/5/0 btlan CLAIMED INTERFACE HP A5506B PCI 10/100Base-TX 4 Port
/dev/diag/lan6 /dev/ether6 /dev/lan6
lan 7 0/6/1/0/6/0 btlan CLAIMED INTERFACE HP A5506B PCI 10/100Base-TX 4 Port
/dev/diag/lan7 /dev/ether7 /dev/lan7
lan 8 0/6/1/0/7/0 btlan CLAIMED INTERFACE HP A5506B PCI 10/100Base-TX 4 Port
/dev/diag/lan8 /dev/ether8 /dev/lan8
root:=>

[Raj D.]

Hi Mike ,

Checksum can be cheked with :
# cksum /usr/sbin/lanscan


ioscan output shows all the Lan cards ok.

Are you getting same result in both the node.?

Also check with cstm , if the cards status are ok.

# echo "selclass qualifier lan;info;wait;infolog " | /usr/sbin/cstm

Cheers,
Raj.

[MikeL_4]

cksum output:

root:=>cksum /usr/sbin/lanscan
826859478 36864 /usr/sbin/lanscan

cstm out put is in attached file, I don't see where there is a problem there.

Both the servers where the lanscan does not work are in the same SG cluster and I have been doing make_net_recoveries for over a year without a problem. No maintenance has been done on these servers either for over 3 months.

[Raj D.]

Mike,

It seems that make_net_recovery is not able to get the lanic id by itself , and also as lanscan not working.


You can try running make_tape_recovery with -l lanic_id , option giving the lanic id manually. ( Lanic id of the system being archived.)

Lanic id can be found by # arp hostname,
and also to use make_net_recovery with -v option to get the error and details .


hth,
Raj.

[Raj D.]

Mike ,
Also you can test the Ignite client configuration with following command:


# cd /opt/ignite/bin/
# instl_adm instl_adm -T -D /var/opt/ignite/clients/0x080009455565

where 0x080009455565
is the lanic ID.

hth,
Raj.

[MikeL_4]

Using the -v option does not give any more details...

also tried running with -l lan_id on the make recovery and it gives the same error. We need to resolove my the lanscan command stopped working all of a sudden ???

[Sameer_Nirmal]

Hi,

The "lanscan" command queries currently running kernel memory ( default to /stand/vmunix and /dev/kmem )to provide it's usual output.
It is peculier problem as both the nodes of the SG are unable to run lanscan.

Are you using APA?
Do you see any core file being generated for "lanscan" command in /usr/sbin
Any error logged in syslog , dmesg?
"lanadmin" runs?

In the worst scenario,if nothing could be figured out then I guess rebooting the nodes is the only option.

[MikeL_4]

there is no core file generated, but a dmesg returns:

root:[/usr/sbin]:=>dmesg
Dec 19 07:43
Magic number wrong (namelist mismatch?)

[Robert-Jan Goossens_1]

Mike,

Check this doc.

Title: lanscan & dmesg error w/ hwiftp: empty list & Magic number wrong
Document ID: KBRC00002773
Last Modified Date: 8/4/05

US
http://www2.itrc.hp.com/service/cki/docDisplay.do?docLocale=en_US&docId=200000079992601

Europe
http://www4.itrc.hp.com/service/cki/docDisplay.do?docLocale=en_US&docId=200000079992601

Regards,
Robert-Jan

[MikeL_4]

Thanks all, The problem is related to the kernel that is running on the server is not the same as in /stand, but the previous kernel.
We are trying to reshedule a reboot of the server with the clients.