{ "content": [ { "type": "text", "text": "# perlsec (man)\n\n## NAME\n\nperlsec - Perl security\n\n## DESCRIPTION\n\nPerl is designed to make it easy to program securely even when running with extra privileges,\nlike setuid or setgid programs. Unlike most command line shells, which are based on multiple\nsubstitution passes on each line of the script, Perl uses a more conventional evaluation\nscheme with fewer hidden snags. Additionally, because the language has more builtin\nfunctionality, it can rely less upon external (and possibly untrustworthy) programs to\naccomplish its purposes.\n\n## Sections\n\n- **NAME**\n- **DESCRIPTION**\n- **SECURITY VULNERABILITY CONTACT INFORMATION**\n- **SECURITY MECHANISMS AND CONCERNS** (10 subsections)\n- **SEE ALSO**\n\nUse structuredContent.sections for detailed options, examples, and full documentation.\n" } ], "structuredContent": { "command": "perlsec", "section": "", "mode": "man", "summary": "perlsec - Perl security", "synopsis": null, "tldr_summary": null, "tldr_examples": [], "tldr_source": null, "flags": [], "examples": [], "see_also": [], "section_outline": [ { "name": "NAME", "lines": 2, "subsections": [] }, { "name": "DESCRIPTION", "lines": 7, "subsections": [] }, { "name": "SECURITY VULNERABILITY CONTACT INFORMATION", "lines": 7, "subsections": [] }, { "name": "SECURITY MECHANISMS AND CONCERNS", "lines": 1, "subsections": [ { "name": "Taint mode", "lines": 116 }, { "name": "Laundering and Detecting Tainted Data", "lines": 49 }, { "name": "Switches On the \"#!\" Line", "lines": 7 }, { "name": "Taint mode and @INC", "lines": 21 }, { "name": "Cleaning Up Your Path", "lines": 87 }, { "name": "Shebang Race Condition", "lines": 55 }, { "name": "Protecting Your Programs", "lines": 29 }, { "name": "Unicode", "lines": 4 }, { "name": "Algorithmic Complexity Attacks", "lines": 84 }, { "name": "Using Sudo", "lines": 17 } ] }, { "name": "SEE ALSO", "lines": 5, "subsections": [] } ], "sections": { "NAME": { "content": "perlsec - Perl security\n", "subsections": [] }, "DESCRIPTION": { "content": "Perl is designed to make it easy to program securely even when running with extra privileges,\nlike setuid or setgid programs. Unlike most command line shells, which are based on multiple\nsubstitution passes on each line of the script, Perl uses a more conventional evaluation\nscheme with fewer hidden snags. Additionally, because the language has more builtin\nfunctionality, it can rely less upon external (and possibly untrustworthy) programs to\naccomplish its purposes.\n", "subsections": [] }, "SECURITY VULNERABILITY CONTACT INFORMATION": { "content": "If you believe you have found a security vulnerability in the Perl interpreter or modules\nmaintained in the core Perl codebase, email the details to perl-security@perl.org\n. This address is a closed membership mailing list monitored\nby the Perl security team.\n\nSee perlsecpolicy for additional information.\n", "subsections": [] }, "SECURITY MECHANISMS AND CONCERNS": { "content": "", "subsections": [ { "name": "Taint mode", "content": "Perl automatically enables a set of special security checks, called taint mode, when it\ndetects its program running with differing real and effective user or group IDs. The setuid\nbit in Unix permissions is mode 04000, the setgid bit mode 02000; either or both may be set.\nYou can also enable taint mode explicitly by using the -T command line flag. This flag is\nstrongly suggested for server programs and any program run on behalf of someone else, such as\na CGI script. Once taint mode is on, it's on for the remainder of your script.\n\nWhile in this mode, Perl takes special precautions called taint checks to prevent both\nobvious and subtle traps. Some of these checks are reasonably simple, such as verifying that\npath directories aren't writable by others; careful programmers have always used checks like\nthese. Other checks, however, are best supported by the language itself, and it is these\nchecks especially that contribute to making a set-id Perl program more secure than the\ncorresponding C program.\n\nYou may not use data derived from outside your program to affect something else outside your\nprogram--at least, not by accident. All command line arguments, environment variables,\nlocale information (see perllocale), results of certain system calls (\"readdir()\",\n\"readlink()\", the variable of \"shmread()\", the messages returned by \"msgrcv()\", the password,\ngcos and shell fields returned by the \"getpwxxx()\" calls), and all file input are marked as\n\"tainted\". Tainted data may not be used directly or indirectly in any command that invokes a\nsub-shell, nor in any command that modifies files, directories, or processes, with the\nfollowing exceptions:\n\n• Arguments to \"print\" and \"syswrite\" are not checked for taintedness.\n\n• Symbolic methods\n\n$obj->$method(@args);\n\nand symbolic sub references\n\n&{$foo}(@args);\n$foo->(@args);\n\nare not checked for taintedness. This requires extra carefulness unless you want\nexternal data to affect your control flow. Unless you carefully limit what these\nsymbolic values are, people are able to call functions outside your Perl code, such as\nPOSIX::system, in which case they are able to run arbitrary external code.\n\n• Hash keys are never tainted.\n\nFor efficiency reasons, Perl takes a conservative view of whether data is tainted. If an\nexpression contains tainted data, any subexpression may be considered tainted, even if the\nvalue of the subexpression is not itself affected by the tainted data.\n\nBecause taintedness is associated with each scalar value, some elements of an array or hash\ncan be tainted and others not. The keys of a hash are never tainted.\n\nFor example:\n\n$arg = shift; # $arg is tainted\n$hid = $arg . 'bar'; # $hid is also tainted\n$line = <>; # Tainted\n$line = ; # Also tainted\nopen FOO, \"/home/me/bar\" or die $!;\n$line = ; # Still tainted\n$path = $ENV{'PATH'}; # Tainted, but see below\n$data = 'abc'; # Not tainted\n\nsystem \"echo $arg\"; # Insecure\nsystem \"/bin/echo\", $arg; # Considered insecure\n# (Perl doesn't know about /bin/echo)\nsystem \"echo $hid\"; # Insecure\nsystem \"echo $data\"; # Insecure until PATH set\n\n$path = $ENV{'PATH'}; # $path now tainted\n\n$ENV{'PATH'} = '/bin:/usr/bin';\ndelete @ENV{'IFS', 'CDPATH', 'ENV', 'BASHENV'};\n\n$path = $ENV{'PATH'}; # $path now NOT tainted\nsystem \"echo $data\"; # Is secure now!\n\nopen(FOO, \"< $arg\"); # OK - read-only file\nopen(FOO, \"> $arg\"); # Not OK - trying to write\n\nopen(FOO,\"echo $arg|\"); # Not OK\nopen(FOO,\"-|\")\nor exec 'echo', $arg; # Also not OK\n\n$shout = `echo $arg`; # Insecure, $shout now tainted\n\nunlink $data, $arg; # Insecure\numask $arg; # Insecure\n\nexec \"echo $arg\"; # Insecure\nexec \"echo\", $arg; # Insecure\nexec \"sh\", '-c', $arg; # Very insecure!\n\n@files = <*.c>; # insecure (uses readdir() or similar)\n@files = glob('*.c'); # insecure (uses readdir() or similar)\n\n# In either case, the results of glob are tainted, since the list of\n# filenames comes from outside of the program.\n\n$bad = ($arg, 23); # $bad will be tainted\n$arg, `true`; # Insecure (although it isn't really)\n\nIf you try to do something insecure, you will get a fatal error saying something like\n\"Insecure dependency\" or \"Insecure $ENV{PATH}\".\n\nThe exception to the principle of \"one tainted value taints the whole expression\" is with the\nternary conditional operator \"?:\". Since code with a ternary conditional\n\n$result = $taintedvalue ? \"Untainted\" : \"Also untainted\";\n\nis effectively\n\nif ( $taintedvalue ) {\n$result = \"Untainted\";\n} else {\n$result = \"Also untainted\";\n}\n\nit doesn't make sense for $result to be tainted.\n" }, { "name": "Laundering and Detecting Tainted Data", "content": "To test whether a variable contains tainted data, and whose use would thus trigger an\n\"Insecure dependency\" message, you can use the \"tainted()\" function of the Scalar::Util\nmodule, available in your nearby CPAN mirror, and included in Perl starting from the release\n5.8.0. Or you may be able to use the following \"istainted()\" function.\n\nsub istainted {\nlocal $@; # Don't pollute caller's value.\nreturn ! eval { eval(\"#\" . substr(join(\"\", @), 0, 0)); 1 };\n}\n\nThis function makes use of the fact that the presence of tainted data anywhere within an\nexpression renders the entire expression tainted. It would be inefficient for every operator\nto test every argument for taintedness. Instead, the slightly more efficient and\nconservative approach is used that if any tainted value has been accessed within the same\nexpression, the whole expression is considered tainted.\n\nBut testing for taintedness gets you only so far. Sometimes you have just to clear your\ndata's taintedness. Values may be untainted by using them as keys in a hash; otherwise the\nonly way to bypass the tainting mechanism is by referencing subpatterns from a regular\nexpression match. Perl presumes that if you reference a substring using $1, $2, etc. in a\nnon-tainting pattern, that you knew what you were doing when you wrote that pattern. That\nmeans using a bit of thought--don't just blindly untaint anything, or you defeat the entire\nmechanism. It's better to verify that the variable has only good characters (for certain\nvalues of \"good\") rather than checking whether it has any bad characters. That's because\nit's far too easy to miss bad characters that you never thought of.\n\nHere's a test to make sure that the data contains nothing but \"word\" characters (alphabetics,\nnumerics, and underscores), a hyphen, an at sign, or a dot.\n\nif ($data =~ /^([-\\@\\w.]+)$/) {\n$data = $1; # $data now untainted\n} else {\ndie \"Bad data in '$data'\"; # log this somewhere\n}\n\nThis is fairly secure because \"/\\w+/\" doesn't normally match shell metacharacters, nor are\ndot, dash, or at going to mean something special to the shell. Use of \"/.+/\" would have been\ninsecure in theory because it lets everything through, but Perl doesn't check for that. The\nlesson is that when untainting, you must be exceedingly careful with your patterns.\nLaundering data using regular expression is the only mechanism for untainting dirty data,\nunless you use the strategy detailed below to fork a child of lesser privilege.\n\nThe example does not untaint $data if \"use locale\" is in effect, because the characters\nmatched by \"\\w\" are determined by the locale. Perl considers that locale definitions are\nuntrustworthy because they contain data from outside the program. If you are writing a\nlocale-aware program, and want to launder data with a regular expression containing \"\\w\", put\n\"no locale\" ahead of the expression in the same block. See \"SECURITY\" in perllocale for\nfurther discussion and examples.\n" }, { "name": "Switches On the \"#!\" Line", "content": "When you make a script executable, in order to make it usable as a command, the system will\npass switches to perl from the script's #! line. Perl checks that any command line switches\ngiven to a setuid (or setgid) script actually match the ones set on the #! line. Some Unix\nand Unix-like environments impose a one-switch limit on the #! line, so you may need to use\nsomething like \"-wU\" instead of \"-w -U\" under such systems. (This issue should arise only in\nUnix or Unix-like environments that support #! and setuid or setgid scripts.)\n" }, { "name": "Taint mode and @INC", "content": "When the taint mode (\"-T\") is in effect, the environment variables \"PERL5LIB\" and \"PERLLIB\"\nare ignored by Perl. You can still adjust @INC from outside the program by using the \"-I\"\ncommand line option as explained in perlrun. The two environment variables are ignored\nbecause they are obscured, and a user running a program could be unaware that they are set,\nwhereas the \"-I\" option is clearly visible and therefore permitted.\n\nAnother way to modify @INC without modifying the program, is to use the \"lib\" pragma, e.g.:\n\nperl -Mlib=/foo program\n\nThe benefit of using \"-Mlib=/foo\" over \"-I/foo\", is that the former will automagically remove\nany duplicated directories, while the latter will not.\n\nNote that if a tainted string is added to @INC, the following problem will be reported:\n\nInsecure dependency in require while running with -T switch\n\nOn versions of Perl before 5.26, activating taint mode will also remove the current directory\n(\".\") from the default value of @INC. Since version 5.26, the current directory isn't\nincluded in @INC by default.\n" }, { "name": "Cleaning Up Your Path", "content": "For \"Insecure $ENV{PATH}\" messages, you need to set $ENV{'PATH'} to a known value, and each\ndirectory in the path must be absolute and non-writable by others than its owner and group.\nYou may be surprised to get this message even if the pathname to your executable is fully\nqualified. This is not generated because you didn't supply a full path to the program;\ninstead, it's generated because you never set your PATH environment variable, or you didn't\nset it to something that was safe. Because Perl can't guarantee that the executable in\nquestion isn't itself going to turn around and execute some other program that is dependent\non your PATH, it makes sure you set the PATH.\n\nThe PATH isn't the only environment variable which can cause problems. Because some shells\nmay use the variables IFS, CDPATH, ENV, and BASHENV, Perl checks that those are either empty\nor untainted when starting subprocesses. You may wish to add something like this to your\nsetid and taint-checking scripts.\n\ndelete @ENV{qw(IFS CDPATH ENV BASHENV)}; # Make %ENV safer\n\nIt's also possible to get into trouble with other operations that don't care whether they use\ntainted values. Make judicious use of the file tests in dealing with any user-supplied\nfilenames. When possible, do opens and such after properly dropping any special user (or\ngroup!) privileges. Perl doesn't prevent you from opening tainted filenames for reading, so\nbe careful what you print out. The tainting mechanism is intended to prevent stupid\nmistakes, not to remove the need for thought.\n\nPerl does not call the shell to expand wild cards when you pass \"system\" and \"exec\" explicit\nparameter lists instead of strings with possible shell wildcards in them. Unfortunately, the\n\"open\", \"glob\", and backtick functions provide no such alternate calling convention, so more\nsubterfuge will be required.\n\nPerl provides a reasonably safe way to open a file or pipe from a setuid or setgid program:\njust create a child process with reduced privilege who does the dirty work for you. First,\nfork a child using the special \"open\" syntax that connects the parent and child by a pipe.\nNow the child resets its ID set and any other per-process attributes, like environment\nvariables, umasks, current working directories, back to the originals or known safe values.\nThen the child process, which no longer has any special permissions, does the \"open\" or other\nsystem call. Finally, the child passes the data it managed to access back to the parent.\nBecause the file or pipe was opened in the child while running under less privilege than the\nparent, it's not apt to be tricked into doing something it shouldn't.\n\nHere's a way to do backticks reasonably safely. Notice how the \"exec\" is not called with a\nstring that the shell could expand. This is by far the best way to call something that might\nbe subjected to shell escapes: just never call the shell at all.\n\nuse English;\ndie \"Can't fork: $!\" unless defined($pid = open(KID, \"-|\"));\nif ($pid) { # parent\nwhile () {\n# do something\n}\nclose KID;\n} else {\nmy @temp = ($EUID, $EGID);\nmy $origuid = $UID;\nmy $origgid = $GID;\n$EUID = $UID;\n$EGID = $GID;\n# Drop privileges\n$UID = $origuid;\n$GID = $origgid;\n# Make sure privs are really gone\n($EUID, $EGID) = @temp;\ndie \"Can't drop privileges\"\nunless $UID == $EUID && $GID eq $EGID;\n$ENV{PATH} = \"/bin:/usr/bin\"; # Minimal PATH.\n# Consider sanitizing the environment even more.\nexec 'myprog', 'arg1', 'arg2'\nor die \"can't exec myprog: $!\";\n}\n\nA similar strategy would work for wildcard expansion via \"glob\", although you can use\n\"readdir\" instead.\n\nTaint checking is most useful when although you trust yourself not to have written a program\nto give away the farm, you don't necessarily trust those who end up using it not to try to\ntrick it into doing something bad. This is the kind of security checking that's useful for\nset-id programs and programs launched on someone else's behalf, like CGI programs.\n\nThis is quite different, however, from not even trusting the writer of the code not to try to\ndo something evil. That's the kind of trust needed when someone hands you a program you've\nnever seen before and says, \"Here, run this.\" For that kind of safety, you might want to\ncheck out the Safe module, included standard in the Perl distribution. This module allows\nthe programmer to set up special compartments in which all system operations are trapped and\nnamespace access is carefully controlled. Safe should not be considered bullet-proof,\nthough: it will not prevent the foreign code to set up infinite loops, allocate gigabytes of\nmemory, or even abusing perl bugs to make the host interpreter crash or behave in\nunpredictable ways. In any case it's better avoided completely if you're really concerned\nabout security.\n" }, { "name": "Shebang Race Condition", "content": "Beyond the obvious problems that stem from giving special privileges to systems as flexible\nas scripts, on many versions of Unix, set-id scripts are inherently insecure right from the\nstart. The problem is a race condition in the kernel. Between the time the kernel opens the\nfile to see which interpreter to run and when the (now-set-id) interpreter turns around and\nreopens the file to interpret it, the file in question may have changed, especially if you\nhave symbolic links on your system.\n\nSome Unixes, especially more recent ones, are free of this inherent security bug. On such\nsystems, when the kernel passes the name of the set-id script to open to the interpreter,\nrather than using a pathname subject to meddling, it instead passes /dev/fd/3. This is a\nspecial file already opened on the script, so that there can be no race condition for evil\nscripts to exploit. On these systems, Perl should be compiled with\n\"-DSETUIDSCRIPTSARESECURENOW\". The Configure program that builds Perl tries to figure\nthis out for itself, so you should never have to specify this yourself. Most modern releases\nof SysVr4 and BSD 4.4 use this approach to avoid the kernel race condition.\n\nIf you don't have the safe version of set-id scripts, all is not lost. Sometimes this kernel\n\"feature\" can be disabled, so that the kernel either doesn't run set-id scripts with the set-\nid or doesn't run them at all. Either way avoids the exploitability of the race condition,\nbut doesn't help in actually running scripts set-id.\n\nIf the kernel set-id script feature isn't disabled, then any set-id script provides an\nexploitable vulnerability. Perl can't avoid being exploitable, but will point out vulnerable\nscripts where it can. If Perl detects that it is being applied to a set-id script then it\nwill complain loudly that your set-id script is insecure, and won't run it. When Perl\ncomplains, you need to remove the set-id bit from the script to eliminate the vulnerability.\nRefusing to run the script doesn't in itself close the vulnerability; it is just Perl's way\nof encouraging you to do this.\n\nTo actually run a script set-id, if you don't have the safe version of set-id scripts, you'll\nneed to put a C wrapper around the script. A C wrapper is just a compiled program that does\nnothing except call your Perl program. Compiled programs are not subject to the kernel bug\nthat plagues set-id scripts. Here's a simple wrapper, written in C:\n\n#include \n#include \n#include \n#include \n\n#define REALPATH \"/path/to/script\"\n\nint main(int argc, char argv)\n{\nexecv(REALPATH, argv);\nfprintf(stderr, \"%s: %s: %s\\n\",\nargv[0], REALPATH, strerror(errno));\nreturn 127;\n}\n\nCompile this wrapper into a binary executable and then make it rather than your script setuid\nor setgid. Note that this wrapper isn't doing anything to sanitise the execution environment\nother than ensuring that a safe path to the script is used. It only avoids the shebang race\ncondition. It relies on Perl's own features, and on the script itself being careful, to make\nit safe enough to run the script set-id.\n" }, { "name": "Protecting Your Programs", "content": "There are a number of ways to hide the source to your Perl programs, with varying levels of\n\"security\".\n\nFirst of all, however, you can't take away read permission, because the source code has to be\nreadable in order to be compiled and interpreted. (That doesn't mean that a CGI script's\nsource is readable by people on the web, though.) So you have to leave the permissions at\nthe socially friendly 0755 level. This lets people on your local system only see your\nsource.\n\nSome people mistakenly regard this as a security problem. If your program does insecure\nthings, and relies on people not knowing how to exploit those insecurities, it is not secure.\nIt is often possible for someone to determine the insecure things and exploit them without\nviewing the source. Security through obscurity, the name for hiding your bugs instead of\nfixing them, is little security indeed.\n\nYou can try using encryption via source filters (Filter::* from CPAN, or Filter::Util::Call\nand Filter::Simple since Perl 5.8). But crackers might be able to decrypt it. You can try\nusing the byte code compiler and interpreter described below, but crackers might be able to\nde-compile it. You can try using the native-code compiler described below, but crackers\nmight be able to disassemble it. These pose varying degrees of difficulty to people wanting\nto get at your code, but none can definitively conceal it (this is true of every language,\nnot just Perl).\n\nIf you're concerned about people profiting from your code, then the bottom line is that\nnothing but a restrictive license will give you legal security. License your software and\npepper it with threatening statements like \"This is unpublished proprietary software of XYZ\nCorp. Your access to it does not give you permission to use it blah blah blah.\" You should\nsee a lawyer to be sure your license's wording will stand up in court.\n" }, { "name": "Unicode", "content": "Unicode is a new and complex technology and one may easily overlook certain security\npitfalls. See perluniintro for an overview and perlunicode for details, and \"Security\nImplications of Unicode\" in perlunicode for security implications in particular.\n" }, { "name": "Algorithmic Complexity Attacks", "content": "Certain internal algorithms used in the implementation of Perl can be attacked by choosing\nthe input carefully to consume large amounts of either time or space or both. This can lead\ninto the so-called Denial of Service (DoS) attacks.\n\n• Hash Algorithm - Hash algorithms like the one used in Perl are well known to be\nvulnerable to collision attacks on their hash function. Such attacks involve\nconstructing a set of keys which collide into the same bucket producing inefficient\nbehavior. Such attacks often depend on discovering the seed of the hash function used to\nmap the keys to buckets. That seed is then used to brute-force a key set which can be\nused to mount a denial of service attack. In Perl 5.8.1 changes were introduced to\nharden Perl to such attacks, and then later in Perl 5.18.0 these features were enhanced\nand additional protections added.\n\nAt the time of this writing, Perl 5.18.0 is considered to be well-hardened against\nalgorithmic complexity attacks on its hash implementation. This is largely owed to the\nfollowing measures mitigate attacks:\n\nHash Seed Randomization\nIn order to make it impossible to know what seed to generate an attack key set for,\nthis seed is randomly initialized at process start. This may be overridden by using\nthe PERLHASHSEED environment variable, see \"PERLHASHSEED\" in perlrun. This\nenvironment variable controls how items are actually stored, not how they are\npresented via \"keys\", \"values\" and \"each\".\n\nHash Traversal Randomization\nIndependent of which seed is used in the hash function, \"keys\", \"values\", and \"each\"\nreturn items in a per-hash randomized order. Modifying a hash by insertion will\nchange the iteration order of that hash. This behavior can be overridden by using\n\"hashtraversalmask()\" from Hash::Util or by using the PERLPERTURBKEYS environment\nvariable, see \"PERLPERTURBKEYS\" in perlrun. Note that this feature controls the\n\"visible\" order of the keys, and not the actual order they are stored in.\n\nBucket Order Perturbance\nWhen items collide into a given hash bucket the order they are stored in the chain is\nno longer predictable in Perl 5.18. This has the intention to make it harder to\nobserve a collision. This behavior can be overridden by using the PERLPERTURBKEYS\nenvironment variable, see \"PERLPERTURBKEYS\" in perlrun.\n\nNew Default Hash Function\nThe default hash function has been modified with the intention of making it harder to\ninfer the hash seed.\n\nAlternative Hash Functions\nThe source code includes multiple hash algorithms to choose from. While we believe\nthat the default perl hash is robust to attack, we have included the hash function\nSiphash as a fall-back option. At the time of release of Perl 5.18.0 Siphash is\nbelieved to be of cryptographic strength. This is not the default as it is much\nslower than the default hash.\n\nWithout compiling a special Perl, there is no way to get the exact same behavior of any\nversions prior to Perl 5.18.0. The closest one can get is by setting PERLPERTURBKEYS\nto 0 and setting the PERLHASHSEED to a known value. We do not advise those settings\nfor production use due to the above security considerations.\n\nPerl has never guaranteed any ordering of the hash keys, and the ordering has already\nchanged several times during the lifetime of Perl 5. Also, the ordering of hash keys has\nalways been, and continues to be, affected by the insertion order and the history of\nchanges made to the hash over its lifetime.\n\nAlso note that while the order of the hash elements might be randomized, this \"pseudo-\nordering\" should not be used for applications like shuffling a list randomly (use\n\"List::Util::shuffle()\" for that, see List::Util, a standard core module since Perl\n5.8.0; or the CPAN module \"Algorithm::Numerical::Shuffle\"), or for generating\npermutations (use e.g. the CPAN modules \"Algorithm::Permute\" or\n\"Algorithm::FastPermute\"), or for any cryptographic applications.\n\nTied hashes may have their own ordering and algorithmic complexity attacks.\n\n• Regular expressions - Perl's regular expression engine is so called NFA (Non-\ndeterministic Finite Automaton), which among other things means that it can rather easily\nconsume large amounts of both time and space if the regular expression may match in\nseveral ways. Careful crafting of the regular expressions can help but quite often there\nreally isn't much one can do (the book \"Mastering Regular Expressions\" is required\nreading, see perlfaq2). Running out of space manifests itself by Perl running out of\nmemory.\n\n• Sorting - the quicksort algorithm used in Perls before 5.8.0 to implement the sort()\nfunction was very easy to trick into misbehaving so that it consumes a lot of time.\nStarting from Perl 5.8.0 a different sorting algorithm, mergesort, is used by default.\nMergesort cannot misbehave on any input.\n\nSee for more\ninformation, and any computer science textbook on algorithmic complexity.\n" }, { "name": "Using Sudo", "content": "The popular tool \"sudo\" provides a controlled way for users to be able to run programs as\nother users. It sanitises the execution environment to some extent, and will avoid the\nshebang race condition. If you don't have the safe version of set-id scripts, then \"sudo\"\nmay be a more convenient way of executing a script as another user than writing a C wrapper\nwould be.\n\nHowever, \"sudo\" sets the real user or group ID to that of the target identity, not just the\neffective ID as set-id bits do. As a result, Perl can't detect that it is running under\n\"sudo\", and so won't automatically take its own security precautions such as turning on taint\nmode. Where \"sudo\" configuration dictates exactly which command can be run, the approved\ncommand may include a \"-T\" option to perl to enable taint mode.\n\nIn general, it is necessary to evaluate the suitability of a script to run under \"sudo\"\nspecifically with that kind of execution environment in mind. It is neither necessary nor\nsufficient for the same script to be suitable to run in a traditional set-id arrangement,\nthough many of the issues overlap.\n" } ] }, "SEE ALSO": { "content": "\"ENVIRONMENT\" in perlrun for its description of cleaning up environment variables.\n\n\n\nperl v5.34.0 2025-07-25 PERLSEC(1)", "subsections": [] } } } }