Last night we had a Control Station failover on VNX7500 and before we could fail it back (Change Control etc etc) I needed to create a new Service Offering in EMC UIM/P. Naturally, I logged into UIM/P, changed Control Station IP address in the Vblock configuration and ran Vblock discovery again. Unfortunately, it failed on Control Station discovery.
Here is the error message:
Details revealed the following:
Action completed successfully... 192.168.100.143: found 1 reachable IP address out of 1 ping request Discovering device type and communication protocols for device (192.168.100.143) +++ Device authenticated and credentials identified ---- Driver (#4002:EMC VNX File) could not discover 192.168.100.143 ----> Refer additional details below. ----> ----> ADDITIONAL DETAILS ----> Did not receive SMI-S Provider version information ----> Please run 'dv' command of /opt/emc/ECIM/ECOM/bin/TestSmiProvider and ensure command completes successfully ----> dv ----> ++++ Display version information ++++ ----> ----> CIM ObjectManager Name: EMC:VNX001B.*.vStrong.info ----> ----> CIMOM Version: EMC CIM Server Version 2.6.3.0.0.38D ----> Error: malformed object name: VNX001B.*.vStrong.info ----> ----> Retrieve and Display data - 1 Iteration(s) In 0.283212 Seconds ----> ----> Please press enter key to continue... . . . . . ----> Namespace: root/emc ----> repeat count: 1 ----> (192.168.100.143:5989) ? q ----> Disconnected from 192.168.100.143:5989 ----> +++++ /opt/emc/ECIM/ECOM/bin/TestSmiProvider Terminated Normally ----> ----> Could not verify SMI-S Provider access on port 5989 to address 192.168.100.143 ---- Could not discover driver using SSH ================================================= (192.168.100.143) Valid Credentials Found, but could not determine driver for device
What is this VNX001B.*.vStrong.info? Where the hell the * (asterisk) came from? I logged into Unisphere, checked Control Station Properties and voila:
Disclaimer: .vstrong.info is used to protect the guilty.
I re-configured DNS domain name, NTP server IP addresses and the timezone which were also wrong.
Before we discover Vblock/Control Station again we need to restart CIM service on the Control Station:
cd /celerra/wbem/bin/start_cim_server ./stop_cim_server ./start_cim_server /celerra/wbem
Discovery should now finish successfully!
PS You can manually test connectivity from UIM/P to the Control Station
SSH to UIM/P, run these commands:
cd /opt/emc/ECIM/ECOM/bin ./TestSmiProvider
PPS If you are really fast, you can ‘catch’ the moment UIM/P connects to the Control Station:
netstat -an | grep 5989 tcp 0 0 :::5989 :::* LISTEN tcp 0 0 ::ffff:<CS_IP_ADDRESS>:5989 ::ffff:<UIM/P_IP_ADDRESS>:57081 ESTABLISHED
Recent Comments