[texhax] Umasks, Permissions, and All That

Thomas Schneider schneidt at mail.nih.gov
Fri Jul 12 19:19:38 CEST 2013


Justin:

> > I think the only reasonable permissions are drwxr-xr-x with ownership
> > being root.
> 
> This is certainly reasonable, but definitely not the only reasonable
> choice.  The most reasonable one, without information to the
> contrary, is the one that exists on the user's system.  It should
> stay that way.

No.  There is a mistake in this logic.  My operating system would
produce drwxr-xr-x BUT I am running the script as sudo and sudo
inherits MY permissions which are drwx------.  So the result is the
SYSTEM files in /usr/local get MY permissions - which is not
appropriate.

> >> Then the 2013 folder and its contents are created. If the
> >> installation is done from the DVD, the permissions of this folder
> >> and its contents are solely determined by the TeX Live install
> >> script, and not by MacTeX. Thus our umask problems this year are
> >> problems from that TeX Live script, and MacTeX is only in the
> >> picture because we decided it should never "fix" permissions.
> > 
> > I think that's not a good idea.
> 
> Unfortunately, it's probably the best idea.  "Fixing" things behind
> the user's back is generally a bad idea, because it leads to the
> kind of problems being discussed in these threads: mysterious
> behavior and failures, with calls to support groups for help with
> little to go on.

The WHOLE reason to put files into /usr/local is to allow all users of
the computer to access them.  Changing the /usr/local to be owned by a
particular user (as advocated by some using home brew apparently) is a
bad idea because they are not accessible then.  Files in /usr/local
ought to be readable and executable by everyone.  This is the normal
systems file setup as in /etc for example.

Tom

  Thomas D. Schneider, Ph.D.
  Senior Investigator
  National Institutes of Health
  National Cancer Institute
  Center for Cancer Research
  Gene Regulation and Chromosome Biology Laboratory
  Molecular Information Theory Group
  Frederick, Maryland  21702-1201
  schneidt at mail.nih.gov
  http://schneider.ncifcrf.gov/(current link)
  http://alum.mit.edu/www/toms (permanent link)


More information about the texhax mailing list