MCSG Primary lan failed and never switched back.

[Global Service Desk]

HI All,

i dont know is this the way SG works,

Due to some reason the lan0 went down and SG failed over to standby lan in this case lan1,

PRIMARY down 0/1/2/0 lan0
STANDBY up 0/4/1/0/6/0 lan1
PRIMARY up 0/4/1/0/7/0 lan2
STANDBY up 0/6/1/0 lan3

i can see that !!

Name Mtu Network Address Ipkts Ierrs Opkts Oerrs Coll
lan3* 1500 none none 0 0 0 0 0
lan2 1500 193.164.192.0 193.164.192.46 10417710 0 18627260 0 0
lan1 1500 10.0.21.0 10.0.21.31 552366 0 794621 0 0
lan0 1500 none none 20458532 0 28499115 0 0
lo0 4136 127.0.0.0 127.0.0.1 69876336 0 69876336 0 0


i made the lan0 up manually.

psa1b:/root# ifconfig lan0
lan0: flags=1843
inet 0.0.0.0 netmask 0

but after this i still see that lan0 is down and SG never failed back to Primary though its up.

my question is !

1. Should SG automatically fail back to Primary once it is up and running?
2. or do we need to move manually? if so how can i do that?

Thanks in advance !!

REgards,
Govinda G.

[Chan 007]

Hi Govinda,

If you are talking about primary and secondary NIC, then after the Primary NIC comes UP, it will automatically switchover to Primary NIC.

If you are taking about Node, primary node is down and now started. In this case you have to manually shutodwn the package on secondary node and start on the primary node.

Please find the documents, Managing SG.

http://docs.hp.com/en/B3936-90079/index.html

Chan

[Global Service Desk]

Thanks for the reply,

your may be right !!

i am talking about

"primary and secondary NIC, then after the Primary NIC comes UP, it will automatically switchover to Primary NIC."

but in my case It never switch over ever though after the primary was made up!

please help !!


i still see that it is down in sg and the netstat shows that lan1 is still having the IP.


cmviewcl -v
-------
PRIMARY down 0/1/2/0 lan0
STANDBY up 0/4/1/0/6/0 lan1
PRIMARY up 0/4/1/0/7/0 lan2
STANDBY up 0/6/1/0 lan3

netstat -in
------------
Name Mtu Network Address Ipkts Ierrs Opkts Oerrs Coll
lan3* 1500 none none 0 0 0 0 0
lan2 1500 193.164.192.0 193.164.192.46 10417710 0 18627260 0 0
lan1 1500 10.0.21.0 10.0.21.31 552366 0 794621 0 0
lan0 1500 none none 20458532 0 28499115 0 0
lo0 4136 127.0.0.0 127.0.0.1 69876336 0 69876336 0 0


sa1b:/root# ifconfig lan0
lan0: flags=1843
inet 0.0.0.0 netmask 0

many thanks!!
govinda

[Global Service Desk]

from the document u mentioned i can see that

<<<<
Switching Back to Primary LAN Interfaces after Local Switching

Whenever a node is halted, the cluster daemon (cmcld) will always attempt to switch any Serviceguard-configured subnets running on standby interfaces back to the primary interfaces. This is done regardless of the link state of the primary interfaces. The intent of this switchback is to preserve the original network configuration as it was before the cluster started. Switching back occurs on the specified node if a cmhaltnode command is issued or on all nodes in the cluster if a cmhaltcl command is issued.
>>>

does this means that do i need to run cmhaltnode command on this node to bring it back to original configuration. ?

Regards,
govinda.

[melvyn burnard]

When the failed Primary LAN is restored (i.e. there ios traffic appearing on it) SG will switch back to this LAN.
You have made this impossible as you have done and ifconfig on the LAN, which makes it not possioble to switch back.
You have to fix the Primary LAN problem, and we have seen instances where NW siwtches do NOT resume traffic across the port even though the link may have been restored.
Check your link, NIC and port on teh switch.

[Enrico P.]

In my cluster the stby lan5 is show up in the cmviewcl -v command but not in the ifconfig command so I don' t think there is correlation.
ifconfig:
lan5: flags=842
inet 0.0.0.0 netmask 0

netstat:
lan5* 1500 none none

cmviewcl -v
STANDBY up 0/0/4/1/0/4/0 lan5

Have you try with the command linkloop to see if your lan0 is reachable from the lan of the other node?

Enrico

[Global Service Desk]

I can link loop from another node to this Lan0.

psa1b:/root# lanscan
Hardware Station Crd Hdw Net-Interface NM MAC HP-DLPI DLPI
Path Address In# State NamePPA ID Type Support Mjr#
0/1/2/0 0x00306E4B4968 0 UP lan0 snap0 1 ETHER Yes 119
0/4/1/0/6/0 0x00118572EBB0 1 UP lan1 snap1 2 ETHER Yes 119
0/4/1/0/7/0 0x00118572EBB1 2 UP lan2 snap2 3 ETHER Yes 119
0/6/1/0 0x00110A806192 3 UP lan3 snap3 4 ETHER Yes 119

i can link loop from another node!!

psa1a:/root# linkloop -i1 0x00306E4B4968
Link connectivity to LAN station: 0x00306E4B4968
-- OK


ANy more thoughts !!

Regards!!

GovindaG.

[Josiah Henline]

Try running a cmcheckconf on your cluster ascii file. It may give you more information as to why your IP did not fail back to the primary interface.

[Stephen Doud]

Your linkloop shows a test of lan1, not of lan0. (Not sure what the MAC address is of lan0. Try a linkloop test of lan0, and show us a lanscan output, and please let us know the version of SG installed, as well as the SG patch (if any) - use 'what /usr/lbin/cmcld | grep Date'.

[Prashant Zanwar_4]

try

linkloop -i 0
linkloop -i 1

It shud failover to primary and it does usually.. check for cluster config as recommended by other ITRC folks..

Thanks and regards
Prashant

[Marcel Boogert_1]

Hi there,

Is your lan0 also configured as a heartbeat interface?

Are there any syslog messages?

regards, MB.