HADOOP-7115. Add a cache for getpwuid_r and getpwgid_r calls (tucu)

git-svn-id: https://svn.apache.org/repos/asf/hadoop/common/branches/branch-2@1407580 13f79535-47bb-0310-9956-ffa450edef68

hadoop-common-project/hadoop-common/CHANGES.txt

@@ -134,6 +134,8 @@ Release 2.0.3-alpha - Unreleased
 
     HADOOP-9012. IPC Client sends wrong connection context (daryn via bobby)
 
+    HADOOP-7115. Add a cache for getpwuid_r and getpwgid_r calls (tucu)
+
 Release 2.0.2-alpha - 2012-09-07 
 
   INCOMPATIBLE CHANGES

hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/CommonConfigurationKeys.java

@@ -184,5 +184,11 @@ public class CommonConfigurationKeys extends CommonConfigurationKeysPublic {
    */
   public static final String KERBEROS_TICKET_CACHE_PATH =
       "hadoop.security.kerberos.ticket.cache.path";
-}
 
+  public static final String HADOOP_SECURITY_UID_NAME_CACHE_TIMEOUT_KEY =
+    "hadoop.security.uid.cache.secs";
+
+  public static final long HADOOP_SECURITY_UID_NAME_CACHE_TIMEOUT_DEFAULT =
+    4*60*60; // 4 hours
+
+}

hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/io/SecureIOUtils.java

@@ -120,7 +120,7 @@ public class SecureIOUtils {
     FileInputStream fis = new FileInputStream(f);
     boolean success = false;
     try {
-      Stat stat = NativeIO.fstat(fis.getFD());
+      Stat stat = NativeIO.getFstat(fis.getFD());
       checkStat(f, stat.getOwner(), stat.getGroup(), expectedOwner,
           expectedGroup);
       success = true;

hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/io/nativeio/NativeIO.java

@@ -19,8 +19,13 @@ package org.apache.hadoop.io.nativeio;
 
 import java.io.FileDescriptor;
 import java.io.IOException;
+import java.util.Map;
+import java.util.concurrent.ConcurrentHashMap;
 
+import org.apache.hadoop.classification.InterfaceAudience;
+import org.apache.hadoop.classification.InterfaceStability;
 import org.apache.hadoop.conf.Configuration;
+import org.apache.hadoop.fs.CommonConfigurationKeys;
 import org.apache.hadoop.util.NativeCodeLoader;
 
 import org.apache.commons.logging.Log;
@@ -30,6 +35,8 @@ import org.apache.commons.logging.LogFactory;
  * These functions should generally be used alongside a fallback to another
  * more portable mechanism.
  */
+@InterfaceAudience.Private
+@InterfaceStability.Unstable
 public class NativeIO {
   // Flags for open() call from bits/fcntl.h
   public static final int O_RDONLY   =    00;
@@ -86,6 +93,8 @@ public class NativeIO {
     "hadoop.workaround.non.threadsafe.getpwuid";
   static final boolean WORKAROUND_NON_THREADSAFE_CALLS_DEFAULT = false;
 
+  private static long cacheTimeout = -1;
+
   static {
     if (NativeCodeLoader.isNativeCodeLoaded()) {
       try {
@@ -96,6 +105,14 @@ public class NativeIO {
 
         initNative();
         nativeLoaded = true;
+
+        cacheTimeout = conf.getLong(
+          CommonConfigurationKeys.HADOOP_SECURITY_UID_NAME_CACHE_TIMEOUT_KEY,
+          CommonConfigurationKeys.HADOOP_SECURITY_UID_NAME_CACHE_TIMEOUT_DEFAULT) *
+          1000;
+        LOG.debug("Initialized cache for IDs to User/Group mapping with a" +
+          " cache timeout of " + cacheTimeout/1000 + " seconds.");
+
       } catch (Throwable t) {
         // This can happen if the user has an older version of libhadoop.so
         // installed - in this case we can continue without native IO
@@ -115,7 +132,7 @@ public class NativeIO {
   /** Wrapper around open(2) */
   public static native FileDescriptor open(String path, int flags, int mode) throws IOException;
   /** Wrapper around fstat(2) */
-  public static native Stat fstat(FileDescriptor fd) throws IOException;
+  private static native Stat fstat(FileDescriptor fd) throws IOException;
   /** Wrapper around chmod(2) */
   public static native void chmod(String path, int mode) throws IOException;
 
@@ -176,6 +193,7 @@ public class NativeIO {
    * Result type of the fstat call
    */
   public static class Stat {
+    private int ownerId, groupId;
     private String owner, group;
     private int mode;
 
@@ -196,9 +214,9 @@ public class NativeIO {
     public static final int S_IWUSR = 0000200;  /* write permission, owner */
     public static final int S_IXUSR = 0000100;  /* execute/search permission, owner */
 
-    Stat(String owner, String group, int mode) {
+    Stat(int ownerId, int groupId, int mode) {
-      this.owner = owner;
+      this.ownerId = ownerId;
-      this.group = group;
+      this.groupId = groupId;
       this.mode = mode;
     }
 
@@ -218,4 +236,61 @@ public class NativeIO {
       return mode;
     }
   }
+
+  static native String getUserName(int uid) throws IOException;
+
+  static native String getGroupName(int uid) throws IOException;
+
+  private static class CachedName {
+    final long timestamp;
+    final String name;
+
+    public CachedName(String name, long timestamp) {
+      this.name = name;
+      this.timestamp = timestamp;
+    }
+  }
+
+  private static final Map<Integer, CachedName> USER_ID_NAME_CACHE =
+    new ConcurrentHashMap<Integer, CachedName>();
+
+  private static final Map<Integer, CachedName> GROUP_ID_NAME_CACHE =
+    new ConcurrentHashMap<Integer, CachedName>();
+
+  private enum IdCache { USER, GROUP }
+
+  private static String getName(IdCache domain, int id) throws IOException {
+    Map<Integer, CachedName> idNameCache = (domain == IdCache.USER)
+      ? USER_ID_NAME_CACHE : GROUP_ID_NAME_CACHE;
+    String name;
+    CachedName cachedName = idNameCache.get(id);
+    long now = System.currentTimeMillis();
+    if (cachedName != null && (cachedName.timestamp + cacheTimeout) > now) {
+      name = cachedName.name;
+    } else {
+      name = (domain == IdCache.USER) ? getUserName(id) : getGroupName(id);
+      if (LOG.isDebugEnabled()) {
+        String type = (domain == IdCache.USER) ? "UserName" : "GroupName";
+        LOG.debug("Got " + type + " " + name + " for ID " + id +
+          " from the native implementation");
+      }
+      cachedName = new CachedName(name, now);
+      idNameCache.put(id, cachedName);
+    }
+    return name;
+  }
+
+  /**
+   * Returns the file stat for a file descriptor.
+   *
+   * @param fd file descriptor.
+   * @return the file descriptor file stat.
+   * @throws IOException thrown if there was an IO error while obtaining the file stat.
+   */
+  public static Stat getFstat(FileDescriptor fd) throws IOException {
+    Stat stat = fstat(fd);
+    stat.owner = getName(IdCache.USER, stat.ownerId);
+    stat.group = getName(IdCache.GROUP, stat.groupId);
+    return stat;
+  }
 }

hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/nativeio/NativeIO.c

@@ -72,16 +72,27 @@ static int workaround_non_threadsafe_calls(JNIEnv *env, jclass clazz) {
 static void stat_init(JNIEnv *env, jclass nativeio_class) {
   // Init Stat
   jclass clazz = (*env)->FindClass(env, "org/apache/hadoop/io/nativeio/NativeIO$Stat");
-  PASS_EXCEPTIONS(env);
+  if (!clazz) {
+    return; // exception has been raised
+  }
   stat_clazz = (*env)->NewGlobalRef(env, clazz);
+  if (!stat_clazz) {
+    return; // exception has been raised
+  }
   stat_ctor = (*env)->GetMethodID(env, stat_clazz, "<init>",
-    "(Ljava/lang/String;Ljava/lang/String;I)V");
+    "(III)V");
-  
+  if (!stat_ctor) {
+    return; // exception has been raised
+  }
   jclass obj_class = (*env)->FindClass(env, "java/lang/Object");
-  assert(obj_class != NULL);
+  if (!obj_class) {
+    return; // exception has been raised
+  }
   jmethodID  obj_ctor = (*env)->GetMethodID(env, obj_class,
     "<init>", "()V");
-  assert(obj_ctor != NULL);
+  if (!obj_ctor) {
+    return; // exception has been raised
+  }
 
   if (workaround_non_threadsafe_calls(env, nativeio_class)) {
     pw_lock_object = (*env)->NewObject(env, obj_class, obj_ctor);
@@ -158,8 +169,6 @@ Java_org_apache_hadoop_io_nativeio_NativeIO_fstat(
   JNIEnv *env, jclass clazz, jobject fd_object)
 {
   jobject ret = NULL;
-  char *pw_buf = NULL;
-  int pw_lock_locked = 0;
 
   int fd = fd_get(env, fd_object);
   PASS_EXCEPTIONS_GOTO(env, cleanup);
@@ -171,71 +180,14 @@ Java_org_apache_hadoop_io_nativeio_NativeIO_fstat(
     goto cleanup;
   }
 
-  size_t pw_buflen = get_pw_buflen();
-  if ((pw_buf = malloc(pw_buflen)) == NULL) {
-    THROW(env, "java/lang/OutOfMemoryError", "Couldn't allocate memory for pw buffer");
-    goto cleanup;
-  }
-
-  if (pw_lock_object != NULL) {
-    if ((*env)->MonitorEnter(env, pw_lock_object) != JNI_OK) {
-      goto cleanup;
-    }
-    pw_lock_locked = 1;
-  }
-
-  // Grab username
-  struct passwd pwd, *pwdp;
-  while ((rc = getpwuid_r(s.st_uid, &pwd, pw_buf, pw_buflen, &pwdp)) != 0) {
-    if (rc != ERANGE) {
-      throw_ioe(env, rc);
-      goto cleanup;
-    }
-    free(pw_buf);
-    pw_buflen *= 2;
-    if ((pw_buf = malloc(pw_buflen)) == NULL) {
-      THROW(env, "java/lang/OutOfMemoryError", "Couldn't allocate memory for pw buffer");
-      goto cleanup;
-    }
-  }
-  assert(pwdp == &pwd);
-
-  jstring jstr_username = (*env)->NewStringUTF(env, pwd.pw_name);
-  if (jstr_username == NULL) goto cleanup;
-
-  // Grab group
-  struct group grp, *grpp;
-  while ((rc = getgrgid_r(s.st_gid, &grp, pw_buf, pw_buflen, &grpp)) != 0) {
-    if (rc != ERANGE) {
-      throw_ioe(env, rc);
-      goto cleanup;
-    }
-    free(pw_buf);
-    pw_buflen *= 2;
-    if ((pw_buf = malloc(pw_buflen)) == NULL) {
-      THROW(env, "java/lang/OutOfMemoryError", "Couldn't allocate memory for pw buffer");
-      goto cleanup;
-    }
-  }
-  assert(grpp == &grp);
-
-  jstring jstr_groupname = (*env)->NewStringUTF(env, grp.gr_name);
-  PASS_EXCEPTIONS_GOTO(env, cleanup);
-
   // Construct result
   ret = (*env)->NewObject(env, stat_clazz, stat_ctor,
-    jstr_username, jstr_groupname, s.st_mode);
+    (jint)s.st_uid, (jint)s.st_gid, (jint)s.st_mode);
 
 cleanup:
-  if (pw_buf != NULL) free(pw_buf);
-  if (pw_lock_locked) {
-    (*env)->MonitorExit(env, pw_lock_object);
-  }
   return ret;
 }
 
-
-
 /**
  * public static native void posix_fadvise(
  *   FileDescriptor fd, long offset, long len, int flags);
@@ -385,6 +337,128 @@ Java_org_apache_hadoop_io_nativeio_NativeIO_chmod(
   (*env)->ReleaseStringUTFChars(env, j_path, path);
 }
 
+/*
+ * static native String getUserName(int uid);
+ */
+JNIEXPORT jstring JNICALL 
+Java_org_apache_hadoop_io_nativeio_NativeIO_getUserName(JNIEnv *env, 
+jclass clazz, jint uid)
+{
+  int pw_lock_locked = 0;
+  if (pw_lock_object != NULL) {
+    if ((*env)->MonitorEnter(env, pw_lock_object) != JNI_OK) {
+      goto cleanup;
+    }
+    pw_lock_locked = 1;
+  }
+
+  char *pw_buf = NULL;
+  int rc;
+  size_t pw_buflen = get_pw_buflen();
+  if ((pw_buf = malloc(pw_buflen)) == NULL) {
+    THROW(env, "java/lang/OutOfMemoryError", "Couldn't allocate memory for pw buffer");
+    goto cleanup;
+  }
+
+  // Grab username
+  struct passwd pwd, *pwdp;
+  while ((rc = getpwuid_r((uid_t)uid, &pwd, pw_buf, pw_buflen, &pwdp)) != 0) {
+    if (rc != ERANGE) {
+      throw_ioe(env, rc);
+      goto cleanup;
+    }
+    free(pw_buf);
+    pw_buflen *= 2;
+    if ((pw_buf = malloc(pw_buflen)) == NULL) {
+      THROW(env, "java/lang/OutOfMemoryError", "Couldn't allocate memory for pw buffer");
+      goto cleanup;
+    }
+  }
+  if (pwdp == NULL) {
+    char msg[80];
+    snprintf(msg, sizeof(msg), "uid not found: %d", uid);
+    THROW(env, "java/io/IOException", msg);
+    goto cleanup;
+  }
+  if (pwdp != &pwd) {
+    char msg[80];
+    snprintf(msg, sizeof(msg), "pwd pointer inconsistent with reference. uid: %d", uid);
+    THROW(env, "java/lang/IllegalStateException", msg);
+    goto cleanup;
+  }
+
+  jstring jstr_username = (*env)->NewStringUTF(env, pwd.pw_name);
+
+cleanup:
+  if (pw_lock_locked) {
+    (*env)->MonitorExit(env, pw_lock_object);
+  }
+  if (pw_buf != NULL) free(pw_buf);
+  return jstr_username;
+}
+
+/*
+ * static native String getGroupName(int gid);
+ */
+JNIEXPORT jstring JNICALL 
+Java_org_apache_hadoop_io_nativeio_NativeIO_getGroupName(JNIEnv *env, 
+jclass clazz, jint gid)
+{
+  int pw_lock_locked = 0;
+ 
+  if (pw_lock_object != NULL) {
+    if ((*env)->MonitorEnter(env, pw_lock_object) != JNI_OK) {
+      goto cleanup;
+    }
+    pw_lock_locked = 1;
+  }
+  
+  char *pw_buf = NULL;
+  int rc;
+  size_t pw_buflen = get_pw_buflen();
+  if ((pw_buf = malloc(pw_buflen)) == NULL) {
+    THROW(env, "java/lang/OutOfMemoryError", "Couldn't allocate memory for pw buffer");
+    goto cleanup;
+  }
+  
+  // Grab group
+  struct group grp, *grpp;
+  while ((rc = getgrgid_r((uid_t)gid, &grp, pw_buf, pw_buflen, &grpp)) != 0) {
+    if (rc != ERANGE) {
+      throw_ioe(env, rc);
+      goto cleanup;
+    }
+    free(pw_buf);
+    pw_buflen *= 2;
+    if ((pw_buf = malloc(pw_buflen)) == NULL) {
+      THROW(env, "java/lang/OutOfMemoryError", "Couldn't allocate memory for pw buffer");
+      goto cleanup;
+    }
+  }
+  if (grpp == NULL) {
+    char msg[80];
+    snprintf(msg, sizeof(msg), "gid not found: %d", gid);
+    THROW(env, "java/io/IOException", msg);
+    goto cleanup;
+  }
+  if (grpp != &grp) {
+    char msg[80];
+    snprintf(msg, sizeof(msg), "pwd pointer inconsistent with reference. gid: %d", gid);
+    THROW(env, "java/lang/IllegalStateException", msg);
+    goto cleanup;
+  }
+
+  jstring jstr_groupname = (*env)->NewStringUTF(env, grp.gr_name);
+  PASS_EXCEPTIONS_GOTO(env, cleanup);
+  
+cleanup:
+  if (pw_lock_locked) {
+    (*env)->MonitorExit(env, pw_lock_object);
+  }
+  if (pw_buf != NULL) free(pw_buf);
+  return jstr_groupname;
+}
+
 
 /*
  * Throw a java.IO.IOException, generating the message from errno.

hadoop-common-project/hadoop-common/src/main/resources/core-default.xml

@@ -214,6 +214,17 @@
   </description>
 </property>
 
+
+<property>
+    <name>hadoop.security.uid.cache.secs</name>
+    <value>14400</value>
+    <description>
+        This is the config controlling the validity of the entries in the cache
+        containing the userId to userName and groupId to groupName used by
+        NativeIO getFstat().
+    </description>
+</property>
+
 <property>
   <name>hadoop.rpc.protection</name>
   <value>authentication</value>

hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/io/nativeio/TestNativeIO.java

@@ -61,7 +61,7 @@ public class TestNativeIO {
   public void testFstat() throws Exception {
     FileOutputStream fos = new FileOutputStream(
       new File(TEST_DIR, "testfstat"));
-    NativeIO.Stat stat = NativeIO.fstat(fos.getFD());
+    NativeIO.Stat stat = NativeIO.getFstat(fos.getFD());
     fos.close();
     LOG.info("Stat: " + String.valueOf(stat));
 
@@ -93,7 +93,7 @@ public class TestNativeIO {
           long et = Time.now() + 5000;
           while (Time.now() < et) {
             try {
-              NativeIO.Stat stat = NativeIO.fstat(fos.getFD());
+              NativeIO.Stat stat = NativeIO.getFstat(fos.getFD());
               assertEquals(System.getProperty("user.name"), stat.getOwner());
               assertNotNull(stat.getGroup());
               assertTrue(!"".equals(stat.getGroup()));
@@ -125,7 +125,7 @@ public class TestNativeIO {
       new File(TEST_DIR, "testfstat2"));
     fos.close();
     try {
-      NativeIO.Stat stat = NativeIO.fstat(fos.getFD());
+      NativeIO.Stat stat = NativeIO.getFstat(fos.getFD());
     } catch (NativeIOException nioe) {
       LOG.info("Got expected exception", nioe);
       assertEquals(Errno.EBADF, nioe.getErrno());
@@ -283,4 +283,14 @@ public class TestNativeIO {
     assertEquals(expected, perms.toShort());
   }
 
+  @Test
+  public void testGetUserName() throws IOException {
+    assertFalse(NativeIO.getUserName(0).isEmpty());
+  }
+
+  @Test
+  public void testGetGroupName() throws IOException {
+    assertFalse(NativeIO.getGroupName(0).isEmpty());
+  }
+
 }