Python access to NFS quota information?

Question

Are there any Python modules that will let me get access to NFS quota information without having to parse the output of the quota command? I am aware that if I were sufficiently motivated I could write my own ONC/RPC code using Python's xdrlib module, but I am hoping with all my heart that I do not have to get that intimately involved with the innards of NFS.

Ideally, I am hoping for an existing RPC implementation, or some higher-level module (like a wrapper for an existing C library or something).

An acceptable substitute would be something like the quota command that produces more machine-readable output.

Answer 1

This question was spawned originally by problems with version 3.13 of the quota command, which is essentially useless as a tool for the end user to determine quota information. For example, if alice runs quota, she might see output like this:

Disk quotas for user alice (uid 100): 
     Filesystem  blocks   quota   limit   grace   files   quota   limit   grace
fileserver.example.com:/staff/b/bob
                9254720  9728000 10240000          119768       0       0       

Alice would rightly wonder what this had to do with her account.

It turns out that the minds behind the quota-utils package at long last realized the problems inherent in the tool; with version 4.00 (available in Fedora 16) there are several new options available that turn the quota command into something that actually produces useful information.

Of particular interest is the -f option:

-f, --filesystem-list     display quota information only for given filesystems

So now, a user can run:

quota -f ~

And get the quota information for their home directory. Additionally, the -s option will display "human readable" numbers instead of displaying everything as blocks.

For Alice, this might look like:

$ quota -s -f ~
Disk quotas for user alice (uid 100): 
     Filesystem  blocks   quota   limit   grace   files   quota   limit   grace
fileserver.example.com:/staff/a/alice
              9038M   9500M  10000M            120k       0       0        

Additionally, there are two options that make it much easier to process the output from quota in a script:

• The -w option inhibits the line wrapping when the "filesystem" name is too long.
• The -p option displays 0 for the grace time if the user is not in an over-quota situation. This means that there are always the same number of fields (whereas with the previous version of quota the number of fields could vary depending on the situation).

Combining all of the above, we get something like this:

$ quota -wp -f ~
Disk quotas for user alice (uid 100): 
     Filesystem   space   quota   limit   grace   files   quota   limit   grace
fileserver.example.com:/staff/a/alice 9254720  9728000 10240000       0  119768 0

Which is much more useful for automation.

Answer 2

command_array = ['quota', '-wp', '-f', '/home/nfs/user1']
output = subprocess.check_output(command_array)
# use only the third line, this is util info
beauty_output = output.split('\n')[2].split()
remote_fs = beauty_output[0]
blocks = beauty_output[1]
quota = beauty_output[2]
limit = beauty_output[3]
...

Consider that /home/nfs/user1 is the nfs mount point. I have done this way.