ODA 12.2.1.4 patch drops IB connection on X6-2 X5-2 with Infiniband
(Doc ID 2446200.1)
Last updated on APRIL 17, 2023
Applies to:
Oracle Database Appliance X6-2 HA Hardware - Version All Versions to All Versions [Release All Releases]Oracle Database Appliance X5-2 - Version All Versions to All Versions [Release All Releases]
Linux x86-64
During the application of the patch 12.2.1.4 on an ODA X5-2 or X6-2 with an IB network may fail due to the CRS not coming up on either node. This note provides one reason for that failure to happen and its remediation. This is not the only issue that may affect the CRS coming up and is not that common. Other avenues should be explored prior to the steps in this document since the likelihood of this being an issue is very small.
Symptoms
Make sure the IB network is present in the system:
Example: (To verify you are using IB on your X5-x or X6-2 HA)
# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
link/loopback 00:xx:xx:xx:00:00 brd 00:xx:xx:xx:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP qlen 1000
link/ether 00:xx:xx:xx:2d:9b brd ff:ff:ff:ff:ff:ff
inet 10.xxx.132.224/24 brd 10.xxx.132.255 scope global eth0
valid_lft forever preferred_lft forever
inet 10.xxx.132.226/24 brd 10.xxx.132.255 scope global secondary eth0:1
valid_lft forever preferred_lft forever
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP qlen 1000
link/ether 00:xx:xx:xx:7a:1c brd ff:ff:ff:ff:ff:ff
4: veth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP qlen 1000
link/ether 00:xx:xx:xx:1c:8c brd ff:ff:ff:ff:ff:ff
inet 192.xxx.19.20/24 brd 192.xxx.19.255 scope global veth1
valid_lft forever preferred_lft forever
inet 192.xxx.19.21/24 brd 192.xxx.19.255 scope global secondary veth1:1
valid_lft forever preferred_lft forever
5: ib0: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 7000 qdisc pfifo_fast master ibbond0 state UP qlen 4096
link/infiniband 80:00:2c:25:fe:80:xx:xx:xx:00:00:00:0e:17:ad:72:da:a6:4c:a6 brd 00:ff:ff:ff:ff:12:40:1b:ff:ff:xx:xx:xx:00:00:00:ff:ff:ff:ff
6: ib1: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 7000 qdisc pfifo_fast master ibbond0 state UP qlen 4096
link/infiniband 80:00:2c:26:fe:80:xx:xx:xx:00:00:00:d6:75:87:fa:d7:8c:bf:53 brd 00:ff:ff:ff:ff:12:40:1b:ff:ff:xx:xx:xx:00:00:00:ff:ff:ff:ff
7: bond0: <BROADCAST,MULTICAST,MASTER> mtu 1500 qdisc noop state DOWN
link/ether b6:xx:xx:xx:50:ec brd ff:ff:ff:ff:ff:ff
8: ibbond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 7000 qdisc noqueue state UP
link/infiniband 80:00:2c:25:fe:80:xx:xx:xx:00:00:00:0e:17:ad:72:da:a6:4c:a6 brd 00:ff:ff:ff:ff:12:40:1b:ff:ff:xx:xx:xx:00:00:00:ff:ff:ff:ff
inet 192.xxx.16.28/24 brd 192.xxx.16.255 scope global ibbond0
valid_lft forever preferred_lft forever
inet 169.254.xxx.xxx/16 brd 169.254.xxx.xxx scope global ibbond0:1
valid_lft forever preferred_lft forever
Note: ib0 and ib1 should be present. If they are not, or they are down, proceed to the next steps.
If the IB network is not present or is down check to see that the modules are loaded.
This is what it should look like if it is healthy:
Changes
Upgrading an ODA X5-2 or X6-2 using infiniband for the interconnect to 12.2.1.4
Cause
To view full details, sign in with your My Oracle Support account. |
|
Don't have a My Oracle Support account? Click to get started! |
In this Document
| Symptoms |
| Make sure the IB network is present in the system: |
| If the IB network is not present or is down check to see that the modules are loaded. |
| Changes |
| Upgrading an ODA X5-2 or X6-2 using infiniband for the interconnect to 12.2.1.4 |
| Cause |
| What has happened is that the kernel has linked the weak module driver instead of the driver it should have loaded. |
| Solution |