Agregando una nueva interfaz de red a la red privada Cluster Interconnect (HAIP)
La mayoría de los problemas que suceden en un Oracle Rac están
relacionados con problemas de desconexión en la red privada.
Como sabemos, el software de Oracle Clusterware funciona con
dos redes diferentes.
Una red pública: utilizada para conexiones de usuarios
finales hacia los recursos soportados en el clúster.
Una red privada: necesaria para comunicación entre nodos que
integran el clúster. Mediante esta red
cada uno de los nodos conoce el estado de los demás y pueden mantener la integridad
de este tipo de infraetructura.
En versiones anteriores de Oracle Clusterware era necesario utilizar
tecnologías de terceros para ofrecer alta disponibilidad de interfaces de red
en la red privada. Por ejemplo conceptos como ‘bonding’ en Linux y Windows o
grupos IPMP’s en sistemas Unix. A partir de la versión 11.2 de Oracle Grid
Infraestructure, Clusterware es capaz de crear redundancia de interfaces de red
para asegurar la comunicación dentro de la red privada.
A continuación el procedimiento para agregar una segunda
interfaz de red:
El escenario inicial es el siguiente, tenemos una instalación
de Oracle Clusterware 12c con dos nodos:
oel-nodo1
oel-nodo2
En cada uno de los nodos únicamente existen dos
interfaces de red: una para la red pública y otra para la red privada, ubicadas
en las subredes 192.168.1.0 y 172.17.1.0 respectivamente.
[grid@oel-nodo1 ~]$ ifconfig -a
eth0
Link encap:Ethernet HWaddr
08:00:27:39:E5:98
inet addr:192.168.1.151
Bcast:192.168.1.255
Mask:255.255.255.0
inet6 addr: fe80::a00:27ff:fe39:e598/64 Scope:Link
UP
BROADCAST RUNNING MULTICAST
MTU:1500 Metric:1
RX
packets:0 errors:0 dropped:0 overruns:0 frame:0
TX
packets:68 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX
bytes:0 (0.0 b) TX bytes:11344 (11.0
KiB)
eth1
Link encap:Ethernet HWaddr
08:00:27:12:23:0D
inet addr:172.17.1.1
Bcast:172.17.1.255
Mask:255.255.255.0
inet6 addr: fe80::a00:27ff:fe12:230d/64 Scope:Link
UP
BROADCAST RUNNING MULTICAST
MTU:1500 Metric:1
RX
packets:0 errors:0 dropped:0 overruns:0 frame:0
TX
packets:63 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX
bytes:0 (0.0 b) TX bytes:11025 (10.7
KiB)
|
Para verificar la asignación a nivel de Clusterware
utilizamos:
[grid@oel-nodo1 ~]$ oifcfg getif
eth0
192.168.1.0 global public
eth1
172.17.1.0 global cluster_interconnect
|
El primer paso es agregar las interfaces de red físicamente.
En cada uno de los nodos se ha agregado la interfaz eth0 y se le asigno una ip
dentro del mismo segmento a la que pertenece la red privada:
eth2
Link encap:Ethernet HWaddr
08:00:27:CD:F4:AF
inet addr:172.17.1.3
Bcast:172.17.1.255
Mask:255.255.255.0
inet6 addr: fe80::a00:27ff:fecd:f4af/64 Scope:Link
UP
BROADCAST RUNNING MULTICAST
MTU:1500 Metric:1
RX
packets:63 errors:0 dropped:0 overruns:0 frame:0
TX
packets:42 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX
bytes:21282 (20.7 KiB) TX bytes:11960
(11.6 KiB)
|
Verificamos que Oracle Clusterware sea capaz de ver la
nueva interfaz agregada:
[grid@oel-nodo1 ~]$ oifcfg iflist
eth0
192.168.1.0
eth1
172.17.1.0
eth2 172.17.1.0
|
El siguiente paso será agregar la interfaz nueva para
crear la redundancia en la red privada:
[root@oel-nodo1 ~]# oifcfg setif -global
eht2/172.17.1.0:cluster_interconnect
[root@oel-nodo1 ~]#
|
Después de agregar la nueva interfaz de red, Clusterware
necesita un reinicio en todos los nodos para que los cambios en las nuevas
interfaces tomen efecto.
[root@oel-nodo1
~]# crsctl stop cluster -all
CRS-2673: Attempting to stop 'ora.crsd' on
'oel-nodo1'
CRS-2673: Attempting to stop 'ora.crsd' on
'oel-nodo2'
CRS-2790: Starting shutdown of Cluster Ready
Services-managed resources on 'oel-nodo1'
CRS-2673: Attempting to stop
'ora.LISTENER_SCAN3.lsnr' on 'oel-nodo1'
CRS-2673: Attempting to stop 'ora.oc4j' on
'oel-nodo1'
CRS-2673: Attempting to stop 'ora.LISTENER.lsnr'
on 'oel-nodo1'
CRS-2673: Attempting to stop 'ora.mgmtdb' on
'oel-nodo1'
CRS-2673: Attempting to stop
'ora.LISTENER_SCAN2.lsnr' on 'oel-nodo1'
CRS-2673: Attempting to stop 'ora.cvu' on
'oel-nodo1'
CRS-2790: Starting shutdown of Cluster Ready
Services-managed resources on 'oel-nodo2'
CRS-2673: Attempting to stop 'ora.LISTENER.lsnr'
on 'oel-nodo2'
CRS-2677: Stop of 'ora.cvu' on 'oel-nodo1'
succeeded
CRS-2677: Stop of 'ora.LISTENER.lsnr' on
'oel-nodo2' succeeded
CRS-2673: Attempting to stop 'ora.oel-nodo2.vip'
on 'oel-nodo2'
CRS-2677: Stop of 'ora.LISTENER.lsnr' on
'oel-nodo1' succeeded
CRS-2673: Attempting to stop 'ora.oel-nodo1.vip'
on 'oel-nodo1'
CRS-2677: Stop of 'ora.LISTENER_SCAN2.lsnr' on
'oel-nodo1' succeeded
CRS-2677: Stop of 'ora.LISTENER_SCAN3.lsnr' on
'oel-nodo1' succeeded
CRS-2673: Attempting to stop 'ora.scan2.vip' on
'oel-nodo1'
CRS-2673: Attempting to stop 'ora.scan3.vip' on
'oel-nodo1'
CRS-2677: Stop of 'ora.oel-nodo2.vip' on
'oel-nodo2' succeeded
CRS-2677: Stop of 'ora.scan2.vip' on 'oel-nodo1'
succeeded
CRS-2677: Stop of 'ora.oel-nodo1.vip' on
'oel-nodo1' succeeded
CRS-2677: Stop of 'ora.scan3.vip' on 'oel-nodo1'
succeeded
CRS-2677: Stop of 'ora.mgmtdb' on 'oel-nodo1'
succeeded
CRS-2673: Attempting to stop 'ora.DATA.dg' on
'oel-nodo1'
CRS-2673: Attempting to stop 'ora.MGMTLSNR' on
'oel-nodo1'
CRS-2677: Stop of 'ora.MGMTLSNR' on 'oel-nodo1'
succeeded
CRS-2677: Stop of 'ora.DATA.dg' on 'oel-nodo1'
succeeded
CRS-2673: Attempting to stop 'ora.asm' on
'oel-nodo1'
CRS-2677: Stop of 'ora.asm' on 'oel-nodo1'
succeeded
CRS-2673: Attempting to stop 'ora.DATA.dg' on
'oel-nodo2'
CRS-2673: Attempting to stop
'ora.LISTENER_SCAN1.lsnr' on 'oel-nodo2'
CRS-2677: Stop of 'ora.DATA.dg' on 'oel-nodo2'
succeeded
CRS-2673: Attempting to stop 'ora.asm' on
'oel-nodo2'
CRS-2677: Stop of 'ora.asm' on 'oel-nodo2'
succeeded
CRS-2677: Stop of 'ora.LISTENER_SCAN1.lsnr' on
'oel-nodo2' succeeded
CRS-2673: Attempting to stop 'ora.scan1.vip' on
'oel-nodo2'
CRS-2677: Stop of 'ora.scan1.vip' on 'oel-nodo2'
succeeded
CRS-2673: Attempting to stop 'ora.ons' on
'oel-nodo2'
CRS-2677: Stop of 'ora.ons' on 'oel-nodo2'
succeeded
CRS-2673: Attempting to stop 'ora.net1.network' on
'oel-nodo2'
CRS-2677: Stop of 'ora.net1.network' on
'oel-nodo2' succeeded
CRS-2792: Shutdown of Cluster Ready
Services-managed resources on 'oel-nodo2' has completed
CRS-2677: Stop of 'ora.crsd' on 'oel-nodo2'
succeeded
CRS-2673: Attempting to stop 'ora.ctssd' on
'oel-nodo2'
CRS-2673: Attempting to stop 'ora.evmd' on
'oel-nodo2'
CRS-2673: Attempting to stop 'ora.storage' on
'oel-nodo2'
CRS-2677: Stop of 'ora.storage' on 'oel-nodo2'
succeeded
CRS-2673: Attempting to stop 'ora.asm' on
'oel-nodo2'
CRS-2677: Stop of 'ora.ctssd' on 'oel-nodo2'
succeeded
CRS-2677: Stop of 'ora.evmd' on 'oel-nodo2'
succeeded
CRS-2677: Stop of 'ora.asm' on 'oel-nodo2'
succeeded
CRS-2673: Attempting to stop
'ora.cluster_interconnect.haip' on 'oel-nodo2'
CRS-2677: Stop of 'ora.oc4j' on 'oel-nodo1'
succeeded
CRS-2673: Attempting to stop 'ora.ons' on
'oel-nodo1'
CRS-2677: Stop of 'ora.ons' on 'oel-nodo1'
succeeded
CRS-2673: Attempting to stop 'ora.net1.network' on
'oel-nodo1'
CRS-2677: Stop of 'ora.net1.network' on
'oel-nodo1' succeeded
CRS-2792: Shutdown of Cluster Ready
Services-managed resources on 'oel-nodo1' has completed
CRS-2677: Stop of 'ora.cluster_interconnect.haip'
on 'oel-nodo2' succeeded
CRS-2673: Attempting to stop 'ora.cssd' on
'oel-nodo2'
CRS-2677: Stop of 'ora.cssd' on 'oel-nodo2'
succeeded
CRS-2677: Stop of 'ora.crsd' on 'oel-nodo1'
succeeded
CRS-2673: Attempting to stop 'ora.ctssd' on
'oel-nodo1'
CRS-2673: Attempting to stop 'ora.evmd' on
'oel-nodo1'
CRS-2673: Attempting to stop 'ora.storage' on
'oel-nodo1'
CRS-2677: Stop of 'ora.storage' on 'oel-nodo1'
succeeded
CRS-2673: Attempting to stop 'ora.asm' on
'oel-nodo1'
CRS-2677: Stop of 'ora.ctssd' on 'oel-nodo1'
succeeded
CRS-2677: Stop of 'ora.evmd' on 'oel-nodo1'
succeeded
CRS-2677: Stop of 'ora.asm' on 'oel-nodo1'
succeeded
CRS-2673: Attempting to stop
'ora.cluster_interconnect.haip' on 'oel-nodo1'
CRS-2677: Stop of 'ora.cluster_interconnect.haip'
on 'oel-nodo1' succeeded
CRS-2673: Attempting to stop 'ora.cssd' on
'oel-nodo1'
CRS-2677: Stop of 'ora.cssd' on 'oel-nodo1'
succeeded
[root@oel-nodo1 ~]#
|
[root@oel-nodo1
~]# crsctl stop crs
CRS-2791: Starting shutdown of Oracle High
Availability Services-managed resources on 'oel-nodo1'
CRS-2673: Attempting to stop 'ora.crf' on
'oel-nodo1'
CRS-2673: Attempting to stop 'ora.mdnsd' on
'oel-nodo1'
CRS-2673: Attempting to stop 'ora.gpnpd' on
'oel-nodo1'
CRS-2673: Attempting to stop 'ora.drivers.acfs' on
'oel-nodo1'
CRS-2677: Stop of 'ora.drivers.acfs' on
'oel-nodo1' succeeded
CRS-2677: Stop of 'ora.crf' on 'oel-nodo1'
succeeded
CRS-2673: Attempting to stop 'ora.gipcd' on
'oel-nodo1'
CRS-2677: Stop of 'ora.mdnsd' on 'oel-nodo1'
succeeded
CRS-2677: Stop of 'ora.gpnpd' on 'oel-nodo1'
succeeded
CRS-2677: Stop of 'ora.gipcd' on 'oel-nodo1'
succeeded
CRS-2793: Shutdown of Oracle High Availability
Services-managed resources on 'oel-nodo1' has completed
CRS-4133: Oracle High Availability Services has
been stopped.
[root@oel-nodo2
~]# crsctl stop crs
CRS-2791: Starting shutdown of Oracle High
Availability Services-managed resources on 'oel-nodo2'
CRS-2673: Attempting to stop 'ora.crf' on
'oel-nodo2'
CRS-2673: Attempting to stop 'ora.mdnsd' on
'oel-nodo2'
CRS-2673: Attempting to stop 'ora.gpnpd' on
'oel-nodo2'
CRS-2673: Attempting to stop 'ora.drivers.acfs' on
'oel-nodo2'
CRS-2677: Stop of 'ora.drivers.acfs' on 'oel-nodo2'
succeeded
CRS-2677: Stop of 'ora.crf' on 'oel-nodo2'
succeeded
CRS-2673: Attempting to stop 'ora.gipcd' on
'oel-nodo2'
CRS-2677: Stop of 'ora.mdnsd' on 'oel-nodo2'
succeeded
CRS-2677: Stop of 'ora.gpnpd' on 'oel-nodo2'
succeeded
CRS-2677: Stop of 'ora.gipcd' on 'oel-nodo2'
succeeded
CRS-2793: Shutdown of Oracle High Availability
Services-managed resources on 'oel-nodo2' has completed
CRS-4133: Oracle High Availability Services has
been stopped.
|
Y finalmente iniciamos de nuevo los servicios en cada
nodo:
[root@oel-nodo1 ~]# crsctl start crs
CRS-4123: Oracle High Availability Services has
been started.
[root@oel-nodo2 ~]# crsctl start crs
CRS-4123: Oracle High Availability Services has
been started.
|
Cuando verificamos las interfaces disponibles podremos
observar la nueva interfaz disponible para la red privada.
[root@oel-nodo1 ~]# oifcfg getif
eth0
192.168.1.0 global public
eth1
172.17.1.0 global cluster_interconnect
eht2 172.17.1.0
global cluster_interconnect
|
Esta característica nos permite aprovechar al máximo las
capacidades de Oracle Clusterware y no depender de tecnologías de terceros para
la redundancia de un elemento tan importante como lo es la red privada de
nuestro cluster.
Espero sea de utilidad,
Comments
Post a Comment