A night for HATL

Testing on NIC failure. Malam ni aku ingat bleh la tido awal, baru nak tido Rosdi call dari Lumut, dia kata nak wat test NIC kat HAT96AP1. Hmmmmm...sikit preview mengenai case yang berlaku adalah seperti berikut:

#ifconfig -a

lo0: flags=1000849 mtu 8232 index 1
inet 127.0.0.1 netmask ff000000


ce0: flags=19040843 mtu 1500 index 2
inet 20.93.101.53 netmask ffffffc0 broadcast 20.93.101.63
groupname MNIC_Public
ether 0:3:ba:e:b5:3e


ce2: flags=19040843 mtu 1500 index 3
inet 20.93.101.143 netmask ffffffe0 broadcast 20.93.101.159
groupname dbipmp
ether 0:3:ba:e:b6:74


ce4: flags=9040843 mtu 1500 index 4
inet 20.93.101.144 netmask ffffffe0 broadcast 20.93.101.159
groupname dbipmp
ether 0:3:ba:9b:18:b8


ce4:1: flags=1000843 mtu 1500 index 4
inet 20.93.101.131 netmask ffffffe0 broadcast 20.93.101.159


ce5: flags=9040843 mtu 1500 index 5
inet 20.93.101.54 netmask ffffffc0 broadcast 20.93.101.63
groupname MNIC_Public
ether 0:3:ba:9b:18:b7


ce5:1: flags=1000843 mtu 1500 index 5
inet 20.93.101.43 netmask ffffffc0 broadcast 20.93.101.63


ce5:2: flags=1000843 mtu 1500 index 5
inet 20.93.101.50 netmask ffffffc0 broadcast 20.93.101.63


ce5:3: flags=1000843 mtu 1500 index 5
inet 20.93.101.31 netmask ffffffc0 broadcast 20.93.101.63


Kat sini kita boleh nampak yang interface ce0 & ce2 failed. Means that IP tu tak bleh guna. So apa aku buat, utk mengelakkan application ni terganggu, aku :

- Failover application & cluster service ke HAT96FO2
# hagrp -switch AppService -to HAT96FO2
# hagrp -switch ClusterService -to HAT96FO2

- Dapatkan ifconfig -a utk HAT96AP1

- Stop cluster kat HAT96AP1 utk elakkan gangguan pada cluter
# hastop -local

- Dapatkan value utk /etc/netmask & /etc/defaultrouter
# pg /etc/netmasks
# pg /etc/defautrouter

- Unplumb ce0 & ce2
# ifconfig ce0 unplumb
# ifconfig ce2 unplumb

- Ping mana-mana ip dalam range vlan server utk pastikan ada guna tak

- Configure ce0 & ce2 dgn mana2 IP yang tak digunakan lagi tu
# ifconfig ce0 plumb 20.93.101.1 netmask 255.255.255.192 broadcast + up
# ifconfig ce2 plumb 20.93.101.2 netmask 255.255.255.192 broadcast + up

- Test ping dari luar bleh masuk ke tak, NIC ce0 fail tapi ce2 bleh plak

- Reboot server utk set default configuration, tapi dua2 NIC still fail.
# init 6

- Start semula VCS kat HAT96AP1 then switch balik application & cluster service ke HAT96AP1
# hastart

- Test application bleh running ke tak.

- Remove deployment.cache
# rm -rf /app/eHIS/deployment/ejb/deployment.cache

Lepas boboi dah kata ok, aku pun siap2la nak masuk tido.....ngantuk beb....