Re: BUG #15120: use of getcwd(3)/chdir(2) during path resolution (exec.c)

From: Andrew Gierth <andrew(at)tao11(dot)riddles(dot)org(dot)uk>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: petar(at)smokva(dot)net, pgsql-bugs(at)lists(dot)postgresql(dot)org
Subject: Re: BUG #15120: use of getcwd(3)/chdir(2) during path resolution (exec.c)
Date: 2018-03-17 17:42:46
Message-ID: 87in9uhbco.fsf@news-spur.riddles.org.uk
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

>>>>> "Tom" == Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> writes:

Tom> ... I find that quite an unacceptable answer. We need to resolve
Tom> the symlink correctly, or we risk malfunctioning later, for the
Tom> reasons recited in the comment for find_my_exec().

On systems with openat(), would it not be possible to resolve symlinks
without ever needing chdir?

Tom> I think your example of "sudo /usr/bin/pg_ctl" is pretty
Tom> artificial; who'd bother spelling that out?

Not necessarily exactly like that, but we do occasionally get this issue
coming up on the IRC channel.

Tom> While I've not thought about it very hard, it might be possible to
Tom> rewrite find_my_exec() and resolve_symlinks() "from the ground up"
Tom> so that they don't do getcwd() except in cases where there's
Tom> really no alternative, such as the executable having been invoked
Tom> using a relative path.

I think getcwd is avoidable even then, if openat() etc. are available.

--
Andrew (irc:RhodiumToad)

In response to

Responses

Browse pgsql-bugs by date

  From Date Subject
Next Message Tom Lane 2018-03-17 18:06:14 Re: BUG #15120: use of getcwd(3)/chdir(2) during path resolution (exec.c)
Previous Message Magnus Hagander 2018-03-17 17:33:48 Re: BUG #15116: pg_recvlogical always fails