* Fix a huuuuge security hole in the Nix daemon. It didn't check that

derivations added to the store by clients have "correct" output
  paths (meaning that the output paths are computed by hashing the
  derivation according to a certain algorithm).  This means that a
  malicious user could craft a special .drv file to build *any*
  desired path in the store with any desired contents (so long as the
  path doesn't already exist).  Then the attacker just needs to wait
  for a victim to come along and install the compromised path.

  For instance, if Alice (the attacker) knows that the latest Firefox
  derivation in Nixpkgs produces the path

    /nix/store/1a5nyfd4ajxbyy97r1fslhgrv70gj8a7-firefox-5.0.1

  then (provided this path doesn't already exist) she can craft a .drv
  file that creates that path (i.e., has it as one of its outputs),
  add it to the store using "nix-store --add", and build it with
  "nix-store -r".  So the fake .drv could write a Trojan to the
  Firefox path.  Then, if user Bob (the victim) comes along and does

    $ nix-env -i firefox
    $ firefox

  he executes the Trojan injected by Alice.

  The fix is to have the Nix daemon verify that derivation outputs are
  correct (in addValidPath()).  This required some refactoring to move
  the hash computation code to libstore.
This commit is contained in:
Eelco Dolstra 2011-07-20 18:10:47 +00:00
parent d2bfe1b071
commit b2027f70d9
9 changed files with 180 additions and 81 deletions

View file

@ -188,8 +188,6 @@ void mkPath(Value & v, const char * s);
void copyContext(const Value & v, PathSet & context); void copyContext(const Value & v, PathSet & context);
typedef std::map<Path, Hash> DrvHashes;
/* Cache for calls to addToStore(); maps source paths to the store /* Cache for calls to addToStore(); maps source paths to the store
paths. */ paths. */
typedef std::map<Path, Path> SrcToStore; typedef std::map<Path, Path> SrcToStore;
@ -203,8 +201,6 @@ std::ostream & operator << (std::ostream & str, const Value & v);
class EvalState class EvalState
{ {
public: public:
DrvHashes drvHashes; /* normalised derivation hashes */
SymbolTable symbols; SymbolTable symbols;
const Symbol sWith, sOutPath, sDrvPath, sType, sMeta, sName, const Symbol sWith, sOutPath, sDrvPath, sType, sMeta, sName,

View file

@ -246,63 +246,6 @@ static void prim_trace(EvalState & state, Value * * args, Value & v)
*************************************************************/ *************************************************************/
static bool isFixedOutput(const Derivation & drv)
{
return drv.outputs.size() == 1 &&
drv.outputs.begin()->first == "out" &&
drv.outputs.begin()->second.hash != "";
}
/* Returns the hash of a derivation modulo fixed-output
subderivations. A fixed-output derivation is a derivation with one
output (`out') for which an expected hash and hash algorithm are
specified (using the `outputHash' and `outputHashAlgo'
attributes). We don't want changes to such derivations to
propagate upwards through the dependency graph, changing output
paths everywhere.
For instance, if we change the url in a call to the `fetchurl'
function, we do not want to rebuild everything depending on it
(after all, (the hash of) the file being downloaded is unchanged).
So the *output paths* should not change. On the other hand, the
*derivation paths* should change to reflect the new dependency
graph.
That's what this function does: it returns a hash which is just the
hash of the derivation ATerm, except that any input derivation
paths have been replaced by the result of a recursive call to this
function, and that for fixed-output derivations we return a hash of
its output path. */
static Hash hashDerivationModulo(EvalState & state, Derivation drv)
{
/* Return a fixed hash for fixed-output derivations. */
if (isFixedOutput(drv)) {
DerivationOutputs::const_iterator i = drv.outputs.begin();
return hashString(htSHA256, "fixed:out:"
+ i->second.hashAlgo + ":"
+ i->second.hash + ":"
+ i->second.path);
}
/* For other derivations, replace the inputs paths with recursive
calls to this function.*/
DerivationInputs inputs2;
foreach (DerivationInputs::const_iterator, i, drv.inputDrvs) {
Hash h = state.drvHashes[i->first];
if (h.type == htUnknown) {
Derivation drv2 = derivationFromPath(i->first);
h = hashDerivationModulo(state, drv2);
state.drvHashes[i->first] = h;
}
inputs2[printHash(h)] = i->second;
}
drv.inputDrvs = inputs2;
return hashString(htSHA256, unparseDerivation(drv));
}
/* Construct (as a unobservable side effect) a Nix derivation /* Construct (as a unobservable side effect) a Nix derivation
expression that performs the derivation described by the argument expression that performs the derivation described by the argument
set. Returns the original set extended with the following set. Returns the original set extended with the following
@ -497,11 +440,10 @@ static void prim_derivationStrict(EvalState & state, Value * * args, Value & v)
(i->first == "out" ...) doesn't affect the hash of the (i->first == "out" ...) doesn't affect the hash of the
others. Is that exploitable? */ others. Is that exploitable? */
if (!fixedOnly) { if (!fixedOnly) {
Hash h = hashDerivationModulo(state, drv); Hash h = hashDerivationModulo(drv);
foreach (DerivationOutputs::iterator, i, drv.outputs) foreach (DerivationOutputs::iterator, i, drv.outputs)
if (i->second.path == "") { if (i->second.path == "") {
Path outPath = makeStorePath("output:" + i->first, h, Path outPath = makeOutputPath(i->first, h, drvName);
drvName + (i->first == "out" ? "" : "-" + i->first));
drv.env[i->first] = outPath; drv.env[i->first] = outPath;
i->second.path = outPath; i->second.path = outPath;
} }
@ -516,7 +458,7 @@ static void prim_derivationStrict(EvalState & state, Value * * args, Value & v)
/* Optimisation, but required in read-only mode! because in that /* Optimisation, but required in read-only mode! because in that
case we don't actually write store derivations, so we can't case we don't actually write store derivations, so we can't
read them later. */ read them later. */
state.drvHashes[drvPath] = hashDerivationModulo(state, drv); drvHashes[drvPath] = hashDerivationModulo(drv);
state.mkAttrs(v, 1 + drv.outputs.size()); state.mkAttrs(v, 1 + drv.outputs.size());
mkString(*state.allocAttr(v, state.sDrvPath), drvPath, singleton<PathSet>("=" + drvPath)); mkString(*state.allocAttr(v, state.sDrvPath), drvPath, singleton<PathSet>("=" + drvPath));

View file

@ -1916,14 +1916,9 @@ void DerivationGoal::computeClosure()
hash). */ hash). */
if (i->second.hash != "") { if (i->second.hash != "") {
bool recursive = false; bool recursive; HashType ht; Hash h;
string algo = i->second.hashAlgo; i->second.parseHashInfo(recursive, ht, h);
if (string(algo, 0, 2) == "r:") {
recursive = true;
algo = string(algo, 2);
}
if (!recursive) { if (!recursive) {
/* The output path should be a regular file without /* The output path should be a regular file without
execute permission. */ execute permission. */
@ -1934,15 +1929,11 @@ void DerivationGoal::computeClosure()
} }
/* Check the hash. */ /* Check the hash. */
HashType ht = parseHashType(algo);
if (ht == htUnknown)
throw BuildError(format("unknown hash algorithm `%1%'") % algo);
Hash h = parseHash(ht, i->second.hash);
Hash h2 = recursive ? hashPath(ht, path).first : hashFile(ht, path); Hash h2 = recursive ? hashPath(ht, path).first : hashFile(ht, path);
if (h != h2) if (h != h2)
throw BuildError( throw BuildError(
format("output path `%1%' should have %2% hash `%3%', instead has `%4%'") format("output path `%1%' should have %2% hash `%3%', instead has `%4%'")
% path % algo % printHash(h) % printHash(h2)); % path % i->second.hashAlgo % printHash(h) % printHash(h2));
} }
/* Get rid of all weird permissions. */ /* Get rid of all weird permissions. */

View file

@ -2,11 +2,30 @@
#include "store-api.hh" #include "store-api.hh"
#include "globals.hh" #include "globals.hh"
#include "util.hh" #include "util.hh"
#include "misc.hh"
namespace nix { namespace nix {
void DerivationOutput::parseHashInfo(bool & recursive, HashType & hashType, Hash & hash) const
{
recursive = false;
string algo = hashAlgo;
if (string(algo, 0, 2) == "r:") {
recursive = true;
algo = string(algo, 2);
}
hashType = parseHashType(algo);
if (hashType == htUnknown)
throw Error(format("unknown hash algorithm `%1%'") % algo);
hash = parseHash(hashType, this->hash);
}
Path writeDerivation(const Derivation & drv, const string & name) Path writeDerivation(const Derivation & drv, const string & name)
{ {
PathSet references; PathSet references;
@ -171,4 +190,64 @@ bool isDerivation(const string & fileName)
} }
bool isFixedOutputDrv(const Derivation & drv)
{
return drv.outputs.size() == 1 &&
drv.outputs.begin()->first == "out" &&
drv.outputs.begin()->second.hash != "";
}
DrvHashes drvHashes;
/* Returns the hash of a derivation modulo fixed-output
subderivations. A fixed-output derivation is a derivation with one
output (`out') for which an expected hash and hash algorithm are
specified (using the `outputHash' and `outputHashAlgo'
attributes). We don't want changes to such derivations to
propagate upwards through the dependency graph, changing output
paths everywhere.
For instance, if we change the url in a call to the `fetchurl'
function, we do not want to rebuild everything depending on it
(after all, (the hash of) the file being downloaded is unchanged).
So the *output paths* should not change. On the other hand, the
*derivation paths* should change to reflect the new dependency
graph.
That's what this function does: it returns a hash which is just the
hash of the derivation ATerm, except that any input derivation
paths have been replaced by the result of a recursive call to this
function, and that for fixed-output derivations we return a hash of
its output path. */
Hash hashDerivationModulo(Derivation drv)
{
/* Return a fixed hash for fixed-output derivations. */
if (isFixedOutputDrv(drv)) {
DerivationOutputs::const_iterator i = drv.outputs.begin();
return hashString(htSHA256, "fixed:out:"
+ i->second.hashAlgo + ":"
+ i->second.hash + ":"
+ i->second.path);
}
/* For other derivations, replace the inputs paths with recursive
calls to this function.*/
DerivationInputs inputs2;
foreach (DerivationInputs::const_iterator, i, drv.inputDrvs) {
Hash h = drvHashes[i->first];
if (h.type == htUnknown) {
Derivation drv2 = derivationFromPath(i->first);
h = hashDerivationModulo(drv2);
drvHashes[i->first] = h;
}
inputs2[printHash(h)] = i->second;
}
drv.inputDrvs = inputs2;
return hashString(htSHA256, unparseDerivation(drv));
}
} }

View file

@ -4,6 +4,7 @@
#include <map> #include <map>
#include "types.hh" #include "types.hh"
#include "hash.hh"
namespace nix { namespace nix {
@ -29,6 +30,7 @@ struct DerivationOutput
this->hashAlgo = hashAlgo; this->hashAlgo = hashAlgo;
this->hash = hash; this->hash = hash;
} }
void parseHashInfo(bool & recursive, HashType & hashType, Hash & hash) const;
}; };
typedef std::map<string, DerivationOutput> DerivationOutputs; typedef std::map<string, DerivationOutput> DerivationOutputs;
@ -64,7 +66,16 @@ string unparseDerivation(const Derivation & drv);
derivations. */ derivations. */
bool isDerivation(const string & fileName); bool isDerivation(const string & fileName);
/* Return true iff this is a fixed-output derivation. */
bool isFixedOutputDrv(const Derivation & drv);
Hash hashDerivationModulo(Derivation drv);
/* Memoisation of hashDerivationModulo(). */
typedef std::map<Path, Hash> DrvHashes;
extern DrvHashes drvHashes;
} }

View file

@ -469,6 +469,47 @@ void canonicalisePathMetaData(const Path & path)
} }
void LocalStore::checkDerivationOutputs(const Path & drvPath, const Derivation & drv)
{
string drvName = storePathToName(drvPath);
assert(isDerivation(drvName));
drvName = string(drvName, 0, drvName.size() - drvExtension.size());
if (isFixedOutputDrv(drv)) {
DerivationOutputs::const_iterator out = drv.outputs.find("out");
if (out == drv.outputs.end())
throw Error(format("derivation `%1%' does not have an output named `out'") % drvPath);
bool recursive; HashType ht; Hash h;
out->second.parseHashInfo(recursive, ht, h);
Path outPath = makeFixedOutputPath(recursive, ht, h, drvName);
StringPairs::const_iterator j = drv.env.find("out");
if (out->second.path != outPath || j == drv.env.end() || j->second != outPath)
throw Error(format("derivation `%1%' has incorrect output `%2%', should be `%3%'")
% drvPath % out->second.path % outPath);
}
else {
Derivation drvCopy(drv);
foreach (DerivationOutputs::iterator, i, drvCopy.outputs) {
i->second.path = "";
drvCopy.env[i->first] = "";
}
Hash h = hashDerivationModulo(drvCopy);
foreach (DerivationOutputs::const_iterator, i, drv.outputs) {
Path outPath = makeOutputPath(i->first, h, drvName);
StringPairs::const_iterator j = drv.env.find(i->first);
if (i->second.path != outPath || j == drv.env.end() || j->second != outPath)
throw Error(format("derivation `%1%' has incorrect output `%2%', should be `%3%'")
% drvPath % i->second.path % outPath);
}
}
}
unsigned long long LocalStore::addValidPath(const ValidPathInfo & info) unsigned long long LocalStore::addValidPath(const ValidPathInfo & info)
{ {
SQLiteStmtUse use(stmtRegisterValidPath); SQLiteStmtUse use(stmtRegisterValidPath);
@ -493,6 +534,14 @@ unsigned long long LocalStore::addValidPath(const ValidPathInfo & info)
derivation. */ derivation. */
if (isDerivation(info.path)) { if (isDerivation(info.path)) {
Derivation drv = parseDerivation(readFile(info.path)); Derivation drv = parseDerivation(readFile(info.path));
/* Verify that the output paths in the derivation are correct
(i.e., follow the scheme for computing output paths from
derivations). Note that if this throws an error, then the
DB transaction is rolled back, so the path validity
registration above is undone. */
checkDerivationOutputs(info.path, drv);
foreach (DerivationOutputs::iterator, i, drv.outputs) { foreach (DerivationOutputs::iterator, i, drv.outputs) {
SQLiteStmtUse use(stmtAddDerivationOutput); SQLiteStmtUse use(stmtAddDerivationOutput);
stmtAddDerivationOutput.bind(id); stmtAddDerivationOutput.bind(id);
@ -925,6 +974,8 @@ void LocalStore::invalidatePath(const Path & path)
{ {
debug(format("invalidating path `%1%'") % path); debug(format("invalidating path `%1%'") % path);
drvHashes.erase(path);
SQLiteStmtUse use(stmtInvalidatePath); SQLiteStmtUse use(stmtInvalidatePath);
stmtInvalidatePath.bind(path); stmtInvalidatePath.bind(path);

View file

@ -25,6 +25,9 @@ const int nixSchemaVersion = 6;
extern string drvsLogDir; extern string drvsLogDir;
struct Derivation;
struct OptimiseStats struct OptimiseStats
{ {
unsigned long totalFiles; unsigned long totalFiles;
@ -255,6 +258,8 @@ private:
RunningSubstituter & runningSubstituter); RunningSubstituter & runningSubstituter);
Path createTempDirInStore(); Path createTempDirInStore();
void checkDerivationOutputs(const Path & drvPath, const Derivation & drv);
}; };

View file

@ -72,6 +72,13 @@ Path followLinksToStorePath(const Path & path)
} }
string storePathToName(const Path & path)
{
assertStorePath(path);
return string(path, nixStore.size() + 34);
}
void checkStoreName(const string & name) void checkStoreName(const string & name)
{ {
string validChars = "+-._?="; string validChars = "+-._?=";
@ -101,7 +108,9 @@ void checkStoreName(const string & name)
<name> = a human readable name for the path, typically obtained <name> = a human readable name for the path, typically obtained
from the name attribute of the derivation, or the name of the from the name attribute of the derivation, or the name of the
source file from which the store path is created source file from which the store path is created. For derivation
outputs other than the default "out" output, the string "-<id>"
is suffixed to <name>.
<h> = base-32 representation of the first 160 bits of a SHA-256 <h> = base-32 representation of the first 160 bits of a SHA-256
hash of <s>; the hash part of the store name hash of <s>; the hash part of the store name
@ -120,11 +129,12 @@ void checkStoreName(const string & name)
"source" "source"
for paths copied to the store using addToStore() when recursive for paths copied to the store using addToStore() when recursive
= true and hashAlgo = "sha256" = true and hashAlgo = "sha256"
"output:out" "output:<id>"
for either the outputs created by derivations, OR paths copied for either the outputs created by derivations, OR paths copied
to the store using addToStore() with recursive != true or to the store using addToStore() with recursive != true or
hashAlgo != "sha256" (in that case "source" is used; it's hashAlgo != "sha256" (in that case "source" is used; it's
silly, but it's done that way for compatibility). silly, but it's done that way for compatibility). <id> is the
name of the output (usually, "out").
<h2> = base-16 representation of a SHA-256 hash of: <h2> = base-16 representation of a SHA-256 hash of:
if <type> = "text:...": if <type> = "text:...":
@ -174,6 +184,14 @@ Path makeStorePath(const string & type,
} }
Path makeOutputPath(const string & id,
const Hash & hash, const string & name)
{
return makeStorePath("output:" + id, hash,
name + (id == "out" ? "" : "-" + id));
}
Path makeFixedOutputPath(bool recursive, Path makeFixedOutputPath(bool recursive,
HashType hashAlgo, Hash hash, string name) HashType hashAlgo, Hash hash, string name)
{ {

View file

@ -250,6 +250,9 @@ void assertStorePath(const Path & path);
bool isInStore(const Path & path); bool isInStore(const Path & path);
bool isStorePath(const Path & path); bool isStorePath(const Path & path);
/* Extract the name part of the given store path. */
string storePathToName(const Path & path);
void checkStoreName(const string & name); void checkStoreName(const string & name);
@ -271,6 +274,9 @@ Path followLinksToStorePath(const Path & path);
Path makeStorePath(const string & type, Path makeStorePath(const string & type,
const Hash & hash, const string & name); const Hash & hash, const string & name);
Path makeOutputPath(const string & id,
const Hash & hash, const string & name);
Path makeFixedOutputPath(bool recursive, Path makeFixedOutputPath(bool recursive,
HashType hashAlgo, Hash hash, string name); HashType hashAlgo, Hash hash, string name);