I’m in the early process of planning a NAS data migration from VNX to Isilon, and one of the first steps I wanted to accomplish was to identify which of our VNX file systems are multiprotocol (with both CIFS shares and NFS exports from the same file system). In the environment I support, which has over 10,000 cifs shares, it’s not a trivial task to identify which shares are multiprotocol. After some research it doesn’t appear that there is a built in method from EMC for determining this information from within the Unisphere GUI. From the CLI, however, the server_export command can be used to view the shares and exports.
Here’s an example of listing shares and exports with the server_export command:
[nasadmin@VNX1 ~]$server_export ALL -Protocol cifs -list | grep filesystem01 share "share01$" "/filesystem01/data" umask=022 maxusr=4294967294 netbios=NASSERVER comment="Contact: John Doe" [nasadmin@VNX1 ~]$server_export ALL -Protocol nfs -list | grep filesystem01 export "/root_vdm_01/filesystem01/data01" rw=admins:powerusers:produsers:qausers root=storageadmins access=admins:powerusers:produsers:qausers:storageadmins export "/root_vdm_01/filesystem01/data02" rw=admins:powerusers:produsers:qausers root=storageadmins access=admins:powerusers:produsers:qausers:storageadmins
The output above shows me that the file system named “filesystem01” has one cifs share and two NFS exports. That’s a good start, but I want to get a count of the number of shares and exports rather than a detailed list of them. I can accomplish that by adding ‘wc’ [word count] to the command:
[nasadmin@VNX1 ~]$ server_export ALL -Protocol cifs -list | grep filesystem01 | wc 1 223 450 [nasadmin@VNX1 ~]$ server_export ALL -Protocol nfs -list | grep filesystem01 | wc 2 15 135
That’s closer to what I want. The output includes three numbers and the first number is the line count. I really only want that number, so I’ll just grab it with awk. Ultimately I want the output to go to a single file with each line containing the name of
the file system, the number of CIFS shares, and the number of NFS Exports. This line of code will give me what I want:
[nasadmin@VNX1 ~]$ echo -n "filesystem01",`server_export ALL -Protocol cifs -list | grep filesystem01 | wc | awk '{print $1}'`, `server_export ALL -Protocol nfs -list | grep filesystem01 | wc | awk '{print $1}'` >> multiprotocol.txt ; echo " " >> multiprotocol.txt
The output is below. It’s perfect, as it’s in the format of a comma delimited file and can be easily exported into Microsoft Excel for reporting purposes.
filesystem01, 1, 2
Here’s a more detailed explanation of the command:
echo -n “filesystem01”, : Echo will write the name of the file system to the screen or to a file if you’ve redirected it with “>” at the end of the command. Adding the “-n” supresses the “new line” that is automatically created after text is outputted, as I want each file system and it’s share & export count to be on the same line in the report.
`server_export ALL -Protocol cifs -list | grep filesystem01 | wc | awk ‘{print $1}’`,: The server_export command lists all of the cifs shares for the file system that you’re grepping for. The wc command is for the “word count”, we’re using it to count the number of output lines to verify how many exports exist for the specified file system. The awk ‘{print $1}’ command will output only the first item of data, when it hits a blank space it will stop. If the output is “1 23 34 32 43 1”, running ‘{print $1}’ will only output the 1.
`server_export ALL -Protocol nfs -list | grep filesystem01 | wc | awk ‘{print $1}’` >> multiprotocol.txt: This is the same command as above, but we’re now counting the number of NFS exports rather than CIFS shares.
; echo ” ” >> multiprotocol.txt: After the count is complete and the data has been outputted, I want to run an echo command without the “-n” option to force a line break to the next line, in preparation for the next line of the script. When exporting, using “>” will output the results to a file and overwrite the file if it already exists, if you use “>>”, it will append the results to the file if it already exists. In this case we want to append each line. In an actual script you’d want to create a blank file first with “echo > filename.ext”. Also, the “;” prior to the command instructs the interpreter to start a new command regardless of the success or failure of the prior command.
At this point, all that needs to be done is to create a script that includes the line above with every file system on the VNX. I copied the line of code above into excel into multiple columns, allowing me to copy and paste the file system list from Unisphere and then concatenate the results into a single script file. I’m including a screenshot of one of my script lines from Excel as an example. The final column (AG) has the following formula:
=CONCATENATE(A4,B4,C4,D4,E4,F4,G4,H4,I4,J4,K4,L4,M4,N4,O4,P4,Q4,R4,S4,T4,U4,V4,W4,X4,Y4,Z4,AA4,AB4,AC4,AD4,AE4)
Spreadsheet example:
