Made caches thread safe. In theory, caches are protected by ZEO

clients, but ZEO clients haven't provided very good protection,
leading to cache corruption.  We'll hopefully fix these client issues,
which cause other problems beside cache corruption, but it seems
prudent to provide low-level cache protection.

src/ZEO/ClientStorage.py

@@ -341,8 +341,6 @@ class ClientStorage(object):
         else:
             cache_path = None
         self._cache = self.ClientCacheClass(cache_path, size=cache_size)
-        # TODO:  maybe there's a better time to open the cache?  Unclear.
-        self._cache.open()
 
         self._rpc_mgr = self.ConnectionManagerClass(addr, self,
                                                     tmin=min_disconnect_poll,

src/ZEO/cache.py

@@ -30,6 +30,7 @@ import BTrees.LOBTree
 import logging
 import os
 import tempfile
+import threading
 import time
 
 import ZODB.fsIndex
@@ -119,6 +120,21 @@ ZEC_HEADER_SIZE = 12
 # to the end of the file that the new object can't fit in one
 # contiguous chunk, currentofs is reset to ZEC_HEADER_SIZE first.
 
+class locked(object):
+
+    def __init__(self, func):
+        self.func = func
+
+    def __get__(self, inst, class_):
+        if inst is None:
+            return self
+        def call(*args, **kw):
+            inst._lock.acquire()
+            try:
+                return self.func(inst, *args, **kw)
+            finally:
+                inst._lock.release()
+        return call
 
 class ClientCache(object):
     """A simple in-memory cache."""
@@ -197,6 +213,10 @@ class ClientCache(object):
 
         self._setup_trace(path)
 
+        self.open()
+
+        self._lock = threading.RLock()
+
     # Backward compatibility. Client code used to have to use the fc
     # attr to get to the file cache to get cache stats.
     @property
@@ -351,6 +371,7 @@ class ClientCache(object):
     # instance, and also written out near the start of the cache file.  The
     # new tid must be strictly greater than our current idea of the most
     # recent tid.
+    @locked
     def setLastTid(self, tid):
         if self.tid is not None and tid <= self.tid:
             raise ValueError("new last tid (%s) must be greater than "
@@ -367,10 +388,11 @@ class ClientCache(object):
     # @return a transaction id
     # @defreturn string, or None if no transaction is yet known
     def getLastTid(self):
-        if self.tid == z64:
+        tid = self.tid
+        if tid == z64:
             return None
         else:
-            return self.tid
+            return tid
 
     ##
     # Return the current data record for oid.
@@ -379,6 +401,7 @@ class ClientCache(object):
     #         in the cache
     # @defreturn 3-tuple: (string, string, string)
 
+    @locked
     def load(self, oid):
         ofs = self.current.get(oid)
         if ofs is None:
@@ -406,6 +429,7 @@ class ClientCache(object):
     # @return data record, serial number, start tid, and end tid
     # @defreturn 4-tuple: (string, string, string, string)
 
+    @locked
     def loadBefore(self, oid, before_tid):
         noncurrent_for_oid = self.noncurrent.get(u64(oid))
         if noncurrent_for_oid is None:
@@ -447,6 +471,7 @@ class ClientCache(object):
     #                current.
     # @param data the actual data
 
+    @locked
     def store(self, oid, start_tid, end_tid, data):
         seek = self.f.seek
         if end_tid is None:
@@ -533,6 +558,7 @@ class ClientCache(object):
     # @param oid object id
     # @param tid the id of the transaction that wrote a new revision of oid,
     #        or None to forget all cached info about oid.
+    @locked
     def invalidate(self, oid, tid):
         if tid > self.tid and tid is not None:
             self.setLastTid(tid)
@@ -572,12 +598,18 @@ class ClientCache(object):
         seek = self.f.seek
         read = self.f.read
         for oid, ofs in self.current.iteritems():
+            self._lock.acquire()
+            try:
                 seek(ofs)
                 assert read(1) == 'a', (ofs, self.f.tell(), oid)
                 size, saved_oid, tid, end_tid = unpack(">I8s8s8s", read(28))
                 assert saved_oid == oid, (ofs, self.f.tell(), oid, saved_oid)
                 assert end_tid == z64, (ofs, self.f.tell(), oid)
-            yield oid, tid
+                result = oid, tid
+            finally:
+                self._lock.release()
+
+            yield result
 
     def dump(self):
         from ZODB.utils import oid_repr

src/ZEO/tests/test_cache.py

@@ -67,7 +67,6 @@ class CacheTests(unittest.TestCase):
         # testSerialization reads the entire file into a string, it's not
         # good to leave it that big.
         self.cache = ZEO.cache.ClientCache(size=1024**2)
-        self.cache.open()
 
     def tearDown(self):
         if self.cache.path:
@@ -154,7 +153,6 @@ class CacheTests(unittest.TestCase):
         dst.write(src.read(self.cache.maxsize))
         dst.close()
         copy = ZEO.cache.ClientCache(path)
-        copy.open()
 
         # Verify that internals of both objects are the same.
         # Could also test that external API produces the same results.
@@ -170,7 +168,6 @@ class CacheTests(unittest.TestCase):
         if self.cache.path:
             os.remove(self.cache.path)
         self.cache = ZEO.cache.ClientCache(size=50)
-        self.cache.open()
 
         # We store an object that is a bit larger than the cache can handle.
         self.cache.store(n1, n2, None, "x"*64)
@@ -186,7 +183,6 @@ class CacheTests(unittest.TestCase):
         if self.cache.path:
             os.remove(self.cache.path)
         cache = ZEO.cache.ClientCache(size=50)
-        cache.open()
 
         # We store an object that is a bit larger than the cache can handle.
         cache.store(n1, n2, n3, "x"*64)
@@ -231,7 +227,6 @@ __test__ = dict(
     ...     _ = os.spawnl(os.P_WAIT, sys.executable, sys.executable, 't')
     ...     if os.path.exists('cache'):
     ...         cache = ZEO.cache.ClientCache('cache')
-    ...         cache.open()
     ...         cache.close()
     ...         os.remove('cache')
     ...         os.remove('cache.lock')
@@ -251,7 +246,6 @@ __test__ = dict(
     >>> cache.store(p64(1), p64(1), None, data)
     >>> cache.close()
     >>> cache = ZEO.cache.ClientCache('cache', 1000)
-    >>> cache.open()
     >>> cache.store(p64(2), p64(2), None, 'XXX')
 
     >>> cache.close()
@@ -267,6 +261,56 @@ __test__ = dict(
     LockError: Couldn't lock 'cache.lock'
 
     >>> cache.close()
+    """,
+
+    thread_safe =
+    r"""
+
+    >>> import ZEO.cache, ZODB.utils
+    >>> cache = ZEO.cache.ClientCache('cache', 1000000)
+
+    >>> for i in range(100):
+    ...     cache.store(ZODB.utils.p64(i), ZODB.utils.p64(1), None, '0')
+
+    >>> import random, sys, threading
+    >>> random = random.Random(0)
+    >>> stop = False
+    >>> read_failure = None
+
+    >>> def read_thread():
+    ...     def pick_oid():
+    ...         return ZODB.utils.p64(random.randint(0,99))
+    ...
+    ...     try:
+    ...         while not stop:
+    ...             cache.load(pick_oid())
+    ...             cache.loadBefore(pick_oid(), ZODB.utils.p64(2))
+    ...     except:
+    ...         global read_failure
+    ...         read_failure = sys.exc_info()
+
+    >>> thread = threading.Thread(target=read_thread)
+    >>> thread.start()
+
+    >>> for tid in range(2,10):
+    ...     for oid in range(100):
+    ...         oid = ZODB.utils.p64(oid)
+    ...         cache.invalidate(oid, ZODB.utils.p64(tid))
+    ...         cache.store(oid, ZODB.utils.p64(tid), None, str(tid))
+
+    >>> stop = True
+    >>> thread.join()
+    >>> if read_failure:
+    ...    print 'Read failure:'
+    ...    import traceback
+    ...    traceback.print_exception(*read_failure)
+
+    >>> expected = '9', ZODB.utils.p64(9)
+    >>> for oid in range(100):
+    ...     loaded = cache.load(ZODB.utils.p64(oid))
+    ...     if loaded != expected:
+    ...         print oid, loaded
+    
     """,
     )