1PERLSEC(1) Perl Programmers Reference Guide PERLSEC(1)
2
6 perlsec - Perl security
7
9 Perl is designed to make it easy to program securely even when running
10 with extra privileges, like setuid or setgid programs. Unlike most
11 command line shells, which are based on multiple substitution passes on
12 each line of the script, Perl uses a more conventional evaluation
13 scheme with fewer hidden snags. Additionally, because the language has
14 more builtin functionality, it can rely less upon external (and
15 possibly untrustworthy) programs to accomplish its purposes.
16
18 If you believe you have found a security vulnerability in the Perl
19 interpreter or modules maintained in the core Perl codebase, email the
20 details to perl-security@perl.org <mailto:perl-security@perl.org>.
21 This address is a closed membership mailing list monitored by the Perl
22 security team.
23 See perlsecpolicy for additional information.
25
27 Taint mode
28 By default, Perl automatically enables a set of special security
29 checks, called taint mode, when it detects its program running with
30 differing real and effective user or group IDs. The setuid bit in Unix
31 permissions is mode 04000, the setgid bit mode 02000; either or both
32 may be set. You can also enable taint mode explicitly by using the -T
33 command line flag. This flag is strongly suggested for server programs
34 and any program run on behalf of someone else, such as a CGI script.
35 Once taint mode is on, it's on for the remainder of your script.
36 While in this mode, Perl takes special precautions called taint checks
38 to prevent both obvious and subtle traps. Some of these checks are
39 reasonably simple, such as verifying that path directories aren't
40 writable by others; careful programmers have always used checks like
41 these. Other checks, however, are best supported by the language
42 itself, and it is these checks especially that contribute to making a
43 set-id Perl program more secure than the corresponding C program.
44 You may not use data derived from outside your program to affect
46 something else outside your program--at least, not by accident. All
47 command line arguments, environment variables, locale information (see
48 perllocale), results of certain system calls (readdir(), readlink(),
49 the variable of shmread(), the messages returned by msgrcv(), the
50 password, gcos and shell fields returned by the getpwxxx() calls), and
51 all file input are marked as "tainted". Tainted data may not be used
52 directly or indirectly in any command that invokes a sub-shell, nor in
53 any command that modifies files, directories, or processes, with the
54 following exceptions:
55 Support for taint checks adds an overhead to all Perl programs, whether
57 or not you're using the taint features. Perl 5.18 introduced C
58 preprocessor symbols that can be used to disable the taint features.
59 • Arguments to "print" and "syswrite" are not checked for
61 taintedness.
62 • Symbolic methods
64 $obj->$method(@args);
66 and symbolic sub references
68 &{$foo}(@args);
70 $foo->(@args);
71 are not checked for taintedness. This requires extra carefulness
73 unless you want external data to affect your control flow. Unless
74 you carefully limit what these symbolic values are, people are able
75 to call functions outside your Perl code, such as POSIX::system, in
76 which case they are able to run arbitrary external code.
77 • Hash keys are never tainted.
79 For efficiency reasons, Perl takes a conservative view of whether data
81 is tainted. If an expression contains tainted data, any subexpression
82 may be considered tainted, even if the value of the subexpression is
83 not itself affected by the tainted data.
84 Because taintedness is associated with each scalar value, some elements
86 of an array or hash can be tainted and others not. The keys of a hash
87 are never tainted.
88 For example:
90 $arg = shift; # $arg is tainted
92 $hid = $arg . 'bar'; # $hid is also tainted
93 $line = <>; # Tainted
94 $line = <STDIN>; # Also tainted
95 open FOO, "/home/me/bar" or die $!;
96 $line = <FOO>; # Still tainted
97 $path = $ENV{'PATH'}; # Tainted, but see below
98 $data = 'abc'; # Not tainted
99 system "echo $arg"; # Insecure
101 system "/bin/echo", $arg; # Considered insecure
102 # (Perl doesn't know about /bin/echo)
103 system "echo $hid"; # Insecure
104 system "echo $data"; # Insecure until PATH set
105 $path = $ENV{'PATH'}; # $path now tainted
107 $ENV{'PATH'} = '/bin:/usr/bin';
109 delete @ENV{'IFS', 'CDPATH', 'ENV', 'BASH_ENV'};
110 $path = $ENV{'PATH'}; # $path now NOT tainted
112 system "echo $data"; # Is secure now!
113 open(FOO, "< $arg"); # OK - read-only file
115 open(FOO, "> $arg"); # Not OK - trying to write
116 open(FOO,"echo $arg|"); # Not OK
118 open(FOO,"-|")
119 or exec 'echo', $arg; # Also not OK
120 $shout = `echo $arg`; # Insecure, $shout now tainted
122 unlink $data, $arg; # Insecure
124 umask $arg; # Insecure
125 exec "echo $arg"; # Insecure
127 exec "echo", $arg; # Insecure
128 exec "sh", '-c', $arg; # Very insecure!
129 @files = <*.c>; # insecure (uses readdir() or similar)
131 @files = glob('*.c'); # insecure (uses readdir() or similar)
132 # In either case, the results of glob are tainted, since the list of
134 # filenames comes from outside of the program.
135 $bad = ($arg, 23); # $bad will be tainted
137 $arg, `true`; # Insecure (although it isn't really)
138 If you try to do something insecure, you will get a fatal error saying
140 something like "Insecure dependency" or "Insecure $ENV{PATH}".
141 The exception to the principle of "one tainted value taints the whole
143 expression" is with the ternary conditional operator "?:". Since code
144 with a ternary conditional
145 $result = $tainted_value ? "Untainted" : "Also untainted";
147 is effectively
149 if ( $tainted_value ) {
151 $result = "Untainted";
152 } else {
153 $result = "Also untainted";
154 }
155 it doesn't make sense for $result to be tainted.
157 Laundering and Detecting Tainted Data
159 To test whether a variable contains tainted data, and whose use would
160 thus trigger an "Insecure dependency" message, you can use the
161 tainted() function of the Scalar::Util module, available in your nearby
162 CPAN mirror, and included in Perl starting from the release 5.8.0. Or
163 you may be able to use the following is_tainted() function.
164 sub is_tainted {
166 local $@; # Don't pollute caller's value.
167 return ! eval { eval("#" . substr(join("", @_), 0, 0)); 1 };
168 }
169 This function makes use of the fact that the presence of tainted data
171 anywhere within an expression renders the entire expression tainted.
172 It would be inefficient for every operator to test every argument for
173 taintedness. Instead, the slightly more efficient and conservative
174 approach is used that if any tainted value has been accessed within the
175 same expression, the whole expression is considered tainted.
176 But testing for taintedness gets you only so far. Sometimes you have
178 just to clear your data's taintedness. Values may be untainted by
179 using them as keys in a hash; otherwise the only way to bypass the
180 tainting mechanism is by referencing subpatterns from a regular
181 expression match. Perl presumes that if you reference a substring
182 using $1, $2, etc. in a non-tainting pattern, that you knew what you
183 were doing when you wrote that pattern. That means using a bit of
184 thought--don't just blindly untaint anything, or you defeat the entire
185 mechanism. It's better to verify that the variable has only good
186 characters (for certain values of "good") rather than checking whether
187 it has any bad characters. That's because it's far too easy to miss
188 bad characters that you never thought of.
189 Here's a test to make sure that the data contains nothing but "word"
191 characters (alphabetics, numerics, and underscores), a hyphen, an at
192 sign, or a dot.
193 if ($data =~ /^([-\@\w.]+)$/) {
195 $data = $1; # $data now untainted
196 } else {
197 die "Bad data in '$data'"; # log this somewhere
198 }
199 This is fairly secure because "/\w+/" doesn't normally match shell
201 metacharacters, nor are dot, dash, or at going to mean something
202 special to the shell. Use of "/.+/" would have been insecure in theory
203 because it lets everything through, but Perl doesn't check for that.
204 The lesson is that when untainting, you must be exceedingly careful
205 with your patterns. Laundering data using regular expression is the
206 only mechanism for untainting dirty data, unless you use the strategy
207 detailed below to fork a child of lesser privilege.
208 The example does not untaint $data if "use locale" is in effect,
210 because the characters matched by "\w" are determined by the locale.
211 Perl considers that locale definitions are untrustworthy because they
212 contain data from outside the program. If you are writing a locale-
213 aware program, and want to launder data with a regular expression
214 containing "\w", put "no locale" ahead of the expression in the same
215 block. See "SECURITY" in perllocale for further discussion and
216 examples.
217 Switches On the "#!" Line
219 When you make a script executable, in order to make it usable as a
220 command, the system will pass switches to perl from the script's #!
221 line. Perl checks that any command line switches given to a setuid (or
222 setgid) script actually match the ones set on the #! line. Some Unix
223 and Unix-like environments impose a one-switch limit on the #! line,
224 so you may need to use something like "-wU" instead of "-w -U" under
225 such systems. (This issue should arise only in Unix or Unix-like
226 environments that support #! and setuid or setgid scripts.)
227 Taint mode and @INC
229 +When the taint mode ("-T") is in effect, the environment variables
230 +"PERL5LIB", "PERLLIB", and "PERL_USE_UNSAFE_INC" are ignored by Perl.
231 You can still adjust @INC from outside the program by using the "-I"
232 command line option as explained in perlrun. The two environment
233 variables are ignored because they are obscured, and a user running a
234 program could be unaware that they are set, whereas the "-I" option is
235 clearly visible and therefore permitted.
236 Another way to modify @INC without modifying the program, is to use the
238 "lib" pragma, e.g.:
239 perl -Mlib=/foo program
241 The benefit of using "-Mlib=/foo" over "-I/foo", is that the former
243 will automagically remove any duplicated directories, while the latter
244 will not.
245 Note that if a tainted string is added to @INC, the following problem
247 will be reported:
248 Insecure dependency in require while running with -T switch
250 On versions of Perl before 5.26, activating taint mode will also remove
252 the current directory (".") from the default value of @INC. Since
253 version 5.26, the current directory isn't included in @INC by default.
254 Cleaning Up Your Path
256 For "Insecure $ENV{PATH}" messages, you need to set $ENV{'PATH'} to a
257 known value, and each directory in the path must be absolute and non-
258 writable by others than its owner and group. You may be surprised to
259 get this message even if the pathname to your executable is fully
260 qualified. This is not generated because you didn't supply a full path
261 to the program; instead, it's generated because you never set your PATH
262 environment variable, or you didn't set it to something that was safe.
263 Because Perl can't guarantee that the executable in question isn't
264 itself going to turn around and execute some other program that is
265 dependent on your PATH, it makes sure you set the PATH.
266 The PATH isn't the only environment variable which can cause problems.
268 Because some shells may use the variables IFS, CDPATH, ENV, and
269 BASH_ENV, Perl checks that those are either empty or untainted when
270 starting subprocesses. You may wish to add something like this to your
271 setid and taint-checking scripts.
272 delete @ENV{qw(IFS CDPATH ENV BASH_ENV)}; # Make %ENV safer
274 It's also possible to get into trouble with other operations that don't
276 care whether they use tainted values. Make judicious use of the file
277 tests in dealing with any user-supplied filenames. When possible, do
278 opens and such after properly dropping any special user (or group!)
279 privileges. Perl doesn't prevent you from opening tainted filenames
280 for reading, so be careful what you print out. The tainting mechanism
281 is intended to prevent stupid mistakes, not to remove the need for
282 thought.
283 Perl does not call the shell to expand wild cards when you pass
285 "system" and "exec" explicit parameter lists instead of strings with
286 possible shell wildcards in them. Unfortunately, the "open", "glob",
287 and backtick functions provide no such alternate calling convention, so
288 more subterfuge will be required.
289 Perl provides a reasonably safe way to open a file or pipe from a
291 setuid or setgid program: just create a child process with reduced
292 privilege who does the dirty work for you. First, fork a child using
293 the special "open" syntax that connects the parent and child by a pipe.
294 Now the child resets its ID set and any other per-process attributes,
295 like environment variables, umasks, current working directories, back
296 to the originals or known safe values. Then the child process, which
297 no longer has any special permissions, does the "open" or other system
298 call. Finally, the child passes the data it managed to access back to
299 the parent. Because the file or pipe was opened in the child while
300 running under less privilege than the parent, it's not apt to be
301 tricked into doing something it shouldn't.
302 Here's a way to do backticks reasonably safely. Notice how the "exec"
304 is not called with a string that the shell could expand. This is by
305 far the best way to call something that might be subjected to shell
306 escapes: just never call the shell at all.
307 use English;
309 die "Can't fork: $!" unless defined($pid = open(KID, "-|"));
310 if ($pid) { # parent
311 while (<KID>) {
312 # do something
313 }
314 close KID;
315 } else {
316 my @temp = ($EUID, $EGID);
317 my $orig_uid = $UID;
318 my $orig_gid = $GID;
319 $EUID = $UID;
320 $EGID = $GID;
321 # Drop privileges
322 $UID = $orig_uid;
323 $GID = $orig_gid;
324 # Make sure privs are really gone
325 ($EUID, $EGID) = @temp;
326 die "Can't drop privileges"
327 unless $UID == $EUID && $GID eq $EGID;
328 $ENV{PATH} = "/bin:/usr/bin"; # Minimal PATH.
329 # Consider sanitizing the environment even more.
330 exec 'myprog', 'arg1', 'arg2'
331 or die "can't exec myprog: $!";
332 }
333 A similar strategy would work for wildcard expansion via "glob",
335 although you can use "readdir" instead.
336 Taint checking is most useful when although you trust yourself not to
338 have written a program to give away the farm, you don't necessarily
339 trust those who end up using it not to try to trick it into doing
340 something bad. This is the kind of security checking that's useful for
341 set-id programs and programs launched on someone else's behalf, like
342 CGI programs.
343 This is quite different, however, from not even trusting the writer of
345 the code not to try to do something evil. That's the kind of trust
346 needed when someone hands you a program you've never seen before and
347 says, "Here, run this." For that kind of safety, you might want to
348 check out the Safe module, included standard in the Perl distribution.
349 This module allows the programmer to set up special compartments in
350 which all system operations are trapped and namespace access is
351 carefully controlled. Safe should not be considered bullet-proof,
352 though: it will not prevent the foreign code to set up infinite loops,
353 allocate gigabytes of memory, or even abusing perl bugs to make the
354 host interpreter crash or behave in unpredictable ways. In any case
355 it's better avoided completely if you're really concerned about
356 security.
357 Shebang Race Condition
359 Beyond the obvious problems that stem from giving special privileges to
360 systems as flexible as scripts, on many versions of Unix, set-id
361 scripts are inherently insecure right from the start. The problem is a
362 race condition in the kernel. Between the time the kernel opens the
363 file to see which interpreter to run and when the (now-set-id)
364 interpreter turns around and reopens the file to interpret it, the file
365 in question may have changed, especially if you have symbolic links on
366 your system.
367 Some Unixes, especially more recent ones, are free of this inherent
369 security bug. On such systems, when the kernel passes the name of the
370 set-id script to open to the interpreter, rather than using a pathname
371 subject to meddling, it instead passes /dev/fd/3. This is a special
372 file already opened on the script, so that there can be no race
373 condition for evil scripts to exploit. On these systems, Perl should
374 be compiled with "-DSETUID_SCRIPTS_ARE_SECURE_NOW". The Configure
375 program that builds Perl tries to figure this out for itself, so you
376 should never have to specify this yourself. Most modern releases of
377 SysVr4 and BSD 4.4 use this approach to avoid the kernel race
378 condition.
379 If you don't have the safe version of set-id scripts, all is not lost.
381 Sometimes this kernel "feature" can be disabled, so that the kernel
382 either doesn't run set-id scripts with the set-id or doesn't run them
383 at all. Either way avoids the exploitability of the race condition,
384 but doesn't help in actually running scripts set-id.
385 If the kernel set-id script feature isn't disabled, then any set-id
387 script provides an exploitable vulnerability. Perl can't avoid being
388 exploitable, but will point out vulnerable scripts where it can. If
389 Perl detects that it is being applied to a set-id script then it will
390 complain loudly that your set-id script is insecure, and won't run it.
391 When Perl complains, you need to remove the set-id bit from the script
392 to eliminate the vulnerability. Refusing to run the script doesn't in
393 itself close the vulnerability; it is just Perl's way of encouraging
394 you to do this.
395 To actually run a script set-id, if you don't have the safe version of
397 set-id scripts, you'll need to put a C wrapper around the script. A C
398 wrapper is just a compiled program that does nothing except call your
399 Perl program. Compiled programs are not subject to the kernel bug
400 that plagues set-id scripts. Here's a simple wrapper, written in C:
401 #include <unistd.h>
403 #include <stdio.h>
404 #include <string.h>
405 #include <errno.h>
406 #define REAL_PATH "/path/to/script"
408 int main(int argc, char **argv)
410 {
411 execv(REAL_PATH, argv);
412 fprintf(stderr, "%s: %s: %s\n",
413 argv[0], REAL_PATH, strerror(errno));
414 return 127;
415 }
416 Compile this wrapper into a binary executable and then make it rather
418 than your script setuid or setgid. Note that this wrapper isn't doing
419 anything to sanitise the execution environment other than ensuring that
420 a safe path to the script is used. It only avoids the shebang race
421 condition. It relies on Perl's own features, and on the script itself
422 being careful, to make it safe enough to run the script set-id.
423 Protecting Your Programs
425 There are a number of ways to hide the source to your Perl programs,
426 with varying levels of "security".
427 First of all, however, you can't take away read permission, because the
429 source code has to be readable in order to be compiled and interpreted.
430 (That doesn't mean that a CGI script's source is readable by people on
431 the web, though.) So you have to leave the permissions at the socially
432 friendly 0755 level. This lets people on your local system only see
433 your source.
434 Some people mistakenly regard this as a security problem. If your
436 program does insecure things, and relies on people not knowing how to
437 exploit those insecurities, it is not secure. It is often possible for
438 someone to determine the insecure things and exploit them without
439 viewing the source. Security through obscurity, the name for hiding
440 your bugs instead of fixing them, is little security indeed.
441 You can try using encryption via source filters (Filter::* from CPAN,
443 or Filter::Util::Call and Filter::Simple since Perl 5.8). But crackers
444 might be able to decrypt it. You can try using the byte code compiler
445 and interpreter described below, but crackers might be able to de-
446 compile it. You can try using the native-code compiler described
447 below, but crackers might be able to disassemble it. These pose
448 varying degrees of difficulty to people wanting to get at your code,
449 but none can definitively conceal it (this is true of every language,
450 not just Perl).
451 If you're concerned about people profiting from your code, then the
453 bottom line is that nothing but a restrictive license will give you
454 legal security. License your software and pepper it with threatening
455 statements like "This is unpublished proprietary software of XYZ Corp.
456 Your access to it does not give you permission to use it blah blah
457 blah." You should see a lawyer to be sure your license's wording will
458 stand up in court.
459 Unicode
461 Unicode is a new and complex technology and one may easily overlook
462 certain security pitfalls. See perluniintro for an overview and
463 perlunicode for details, and "Security Implications of Unicode" in
464 perlunicode for security implications in particular.
465 Algorithmic Complexity Attacks
467 Certain internal algorithms used in the implementation of Perl can be
468 attacked by choosing the input carefully to consume large amounts of
469 either time or space or both. This can lead into the so-called Denial
470 of Service (DoS) attacks.
471 • Hash Algorithm - Hash algorithms like the one used in Perl are well
473 known to be vulnerable to collision attacks on their hash function.
474 Such attacks involve constructing a set of keys which collide into
475 the same bucket producing inefficient behavior. Such attacks often
476 depend on discovering the seed of the hash function used to map the
477 keys to buckets. That seed is then used to brute-force a key set
478 which can be used to mount a denial of service attack. In Perl
479 5.8.1 changes were introduced to harden Perl to such attacks, and
480 then later in Perl 5.18.0 these features were enhanced and
481 additional protections added.
482 At the time of this writing, Perl 5.18.0 is considered to be well-
484 hardened against algorithmic complexity attacks on its hash
485 implementation. This is largely owed to the following measures
486 mitigate attacks:
487 Hash Seed Randomization
489 In order to make it impossible to know what seed to generate an
490 attack key set for, this seed is randomly initialized at
491 process start. This may be overridden by using the
492 PERL_HASH_SEED environment variable, see "PERL_HASH_SEED" in
493 perlrun. This environment variable controls how items are
494 actually stored, not how they are presented via "keys",
495 "values" and "each".
496 Hash Traversal Randomization
498 Independent of which seed is used in the hash function, "keys",
499 "values", and "each" return items in a per-hash randomized
500 order. Modifying a hash by insertion will change the iteration
501 order of that hash. This behavior can be overridden by using
502 hash_traversal_mask() from Hash::Util or by using the
503 PERL_PERTURB_KEYS environment variable, see "PERL_PERTURB_KEYS"
504 in perlrun. Note that this feature controls the "visible"
505 order of the keys, and not the actual order they are stored in.
506 Bucket Order Perturbance
508 When items collide into a given hash bucket the order they are
509 stored in the chain is no longer predictable in Perl 5.18.
510 This has the intention to make it harder to observe a
511 collision. This behavior can be overridden by using the
512 PERL_PERTURB_KEYS environment variable, see "PERL_PERTURB_KEYS"
513 in perlrun.
514 New Default Hash Function
516 The default hash function has been modified with the intention
517 of making it harder to infer the hash seed.
518 Alternative Hash Functions
520 The source code includes multiple hash algorithms to choose
521 from. While we believe that the default perl hash is robust to
522 attack, we have included the hash function Siphash as a fall-
523 back option. At the time of release of Perl 5.18.0 Siphash is
524 believed to be of cryptographic strength. This is not the
525 default as it is much slower than the default hash.
526 Without compiling a special Perl, there is no way to get the exact
528 same behavior of any versions prior to Perl 5.18.0. The closest
529 one can get is by setting PERL_PERTURB_KEYS to 0 and setting the
530 PERL_HASH_SEED to a known value. We do not advise those settings
531 for production use due to the above security considerations.
532 Perl has never guaranteed any ordering of the hash keys, and the
534 ordering has already changed several times during the lifetime of
535 Perl 5. Also, the ordering of hash keys has always been, and
536 continues to be, affected by the insertion order and the history of
537 changes made to the hash over its lifetime.
538 Also note that while the order of the hash elements might be
540 randomized, this "pseudo-ordering" should not be used for
541 applications like shuffling a list randomly (use
542 List::Util::shuffle() for that, see List::Util, a standard core
543 module since Perl 5.8.0; or the CPAN module
544 "Algorithm::Numerical::Shuffle"), or for generating permutations
545 (use e.g. the CPAN modules "Algorithm::Permute" or
546 "Algorithm::FastPermute"), or for any cryptographic applications.
547 Tied hashes may have their own ordering and algorithmic complexity
549 attacks.
550 • Regular expressions - Perl's regular expression engine is so called
552 NFA (Non-deterministic Finite Automaton), which among other things
553 means that it can rather easily consume large amounts of both time
554 and space if the regular expression may match in several ways.
555 Careful crafting of the regular expressions can help but quite
556 often there really isn't much one can do (the book "Mastering
557 Regular Expressions" is required reading, see perlfaq2). Running
558 out of space manifests itself by Perl running out of memory.
559 • Sorting - the quicksort algorithm used in Perls before 5.8.0 to
561 implement the sort() function was very easy to trick into
562 misbehaving so that it consumes a lot of time. Starting from Perl
563 5.8.0 a different sorting algorithm, mergesort, is used by default.
564 Mergesort cannot misbehave on any input.
565 See
567 <https://www.usenix.org/legacy/events/sec03/tech/full_papers/crosby/crosby.pdf>
568 for more information, and any computer science textbook on algorithmic
569 complexity.
570 Using Sudo
572 The popular tool "sudo" provides a controlled way for users to be able
573 to run programs as other users. It sanitises the execution environment
574 to some extent, and will avoid the shebang race condition. If you
575 don't have the safe version of set-id scripts, then "sudo" may be a
576 more convenient way of executing a script as another user than writing
577 a C wrapper would be.
578 However, "sudo" sets the real user or group ID to that of the target
580 identity, not just the effective ID as set-id bits do. As a result,
581 Perl can't detect that it is running under "sudo", and so won't
582 automatically take its own security precautions such as turning on
583 taint mode. Where "sudo" configuration dictates exactly which command
584 can be run, the approved command may include a "-T" option to perl to
585 enable taint mode.
586 In general, it is necessary to evaluate the suitability of a script to
588 run under "sudo" specifically with that kind of execution environment
589 in mind. It is neither necessary nor sufficient for the same script to
590 be suitable to run in a traditional set-id arrangement, though many of
591 the issues overlap.
592
594 "ENVIRONMENT" in perlrun for its description of cleaning up environment
595 variables.
596perl v5.38.2 2023-11-30 PERLSEC(1)