fix for zsh compinit insecure directories error

2020-03-28

 | 

~6 min read

 | 

1013 words

With macOS Catalina, Apple made zsh the default shell. That’s all well and good since I have been using zsh for a while now.

However, in getting a new machine setup, I started getting an error every time I opened a new terminal:

zsh compinit: insecure directories, run compaudit for list.
Ignore insecure directories and continue [y] or abort compinit [n]?

Pressing y isn’t hard, but it certainly didn’t feel right that I should have to. So, I started looking around the internet to find an answer.

Background

Plenty of others have had this issue before, though the reason for it surprised me: write privileges for the directories in question.

Wesley Moore beat me to the investigation and has a good write up of the on his blog.

Searching for compinit, I found it in the compsys part of the manual. As the name suggests (in retrospect, this is clearer), compinit is the standard way that the shell is initialized.

The error has to do with a security check (emphasis mine):

For security reasons compinit also checks if the completion system would use files not owned by root or by the current user, or files in directories that are world- or group-writable or that are not owned by root or by the current user. If such files or directories are found, compinit will ask if the completion system should really be used. To avoid these tests and make all files found be used without asking, use the option -u, and to make compinit silently ignore all insecure files and directories use the option -i. This security check is skipped entirely when the -C option is given.

The security check can be retried at any time by running the function compaudit. This is the same check used by compinit, but when it is executed directly any changes to fpath are made local to the function so they do not persist. The directories to be checked may be passed as arguments; if none are given, compaudit uses fpath and _compdir to find completion system directories, adding missing ones to fpath as necessary. To force a check of exactly the directories currently named in fpath, set _compdir to an empty string before calling compaudit or compinit.

That seems to describe the error I’m getting, which begs the question - which directories are owned by the root / current user but are world- or group-writable?

Investigating My Case

Running the audit:

$ compaudit
There are insecure directories:
/usr/local/share/zsh/site-functions
/usr/local/share/zsh

Investigating those directories in particular:

$ ls -la /usr/local/share
total 0
drwxrwxr-x   9 stephen  admin  288 Mar 12 09:31 .
drwxr-xr-x  14 root     wheel  448 Mar 11 17:40 ..
-rw-r--r--   1 stephen  admin    0 Mar 12 09:27 .keepme
drwxr-xr-x   5 stephen  admin  160 Mar 12 09:31 doc
drwxr-xr-x   3 stephen  admin   96 Mar 11 17:43 fish
drwxr-xr-x   6 stephen  admin  192 Mar 12 09:31 man
drwxr-xr-x  19 stephen  admin  608 Mar 12 09:28 postgresql
lrwxr-xr-x   1 stephen  admin   38 Mar 12 09:31 systemtap -> ../Cellar/node/13.10.1/share/systemtap
drwxrwxr-x   3 stephen  admin   96 Mar 11 17:40 zsh

$ ls -la /usr/local/share/zsh
total 0
drwxrwxr-x  3 stephen  admin   96 Mar 11 17:40 .
drwxrwxr-x  9 stephen  admin  288 Mar 12 09:31 ..
drwxrwxr-x  4 stephen  admin  128 Mar 11 17:43 site-functions

Only the directories listed in the audit were writable in the user group. (For a quick refresher on file permissions in Unix systems, check out my previous post.)

Pulling Together The Tools

At this point, we know that the permissions of the directories are problematic - a perfect fit for the chmod utility.

the manual entry for chmod states the following in the Modes section:

Modes may be absolute or symbolic.
[…]
The symbolic mode is described by the following grammar:

           mode         ::= clause [, clause ...]
           clause       ::= [who ...] [action ...] action
           action       ::= op [perm ...]
           who          ::= a | u | g | o
           op           ::= + | - | =
           perm         ::= r | s | t | w | x | X | u | g | o

     The who symbols ‘’u’’, ‘’g’’, and ‘’o’’ specify the user, group, and
     other parts of the mode bits, respectively.  The who symbol ‘’a’’ is
     equivalent to ‘’ugo''.

     The perm symbols represent the portions of the mode bits as follows:

           r       The read bits.
           s       The set-user-ID-on-execution and set-group-ID-on-execution
                   bits.
           t       The sticky bit.
           w       The write bits.
           x       The execute/search bits.
           X       The execute/search bits if the file is a directory or any
                   of the execute/search bits are set in the original (unmodi-
                   fied) mode.  Operations with the perm symbol ``X'' are only
                   meaningful in conjunction with the op symbol ``+'', and are
                   ignored in all other cases.
           u       The user permission bits in the original mode of the file.
           g       The group permission bits in the original mode of the file.
           o       The other permission bits in the original mode of the file.

If you’ve ever seen chmod 755, that’s using the absolute to change file modes and access.

Wesley’s ultimate suggestion was a more expressive approach - describing intent instead of relying on magic numbers. He used the symbolic mode to describe his desired output:

chmod g-w <target file or directory> […other targets]

That is, for the group (g) section of the permissions, use the remove operation (-) for the write (w) permission.

This was further simplified into a one line command by piping the output into an xargs command.

As a reminder (from the manual):

The xargs utility works by reading “space, tab, newline and end-of-file delimited strings from the standard input” and executing a utility with the strings as arguments.

compaudit | xargs chmod g-w

Conclusion

Minor annoyance resolved and I learned more about the chmod utility in the process!



Hi there and thanks for reading! My name's Stephen. I live in Chicago with my wife, Kate, and dog, Finn. Want more? See about and get in touch!