X-Git-Url: https://git.arvados.org/arvados.git/blobdiff_plain/5c4d9d38dcee73a7ffb6221c80f707c3924da64f..c56d04266de4e97b1b861c068309476ca562d72f:/sdk/python/arvados/keep.py

diff --git a/sdk/python/arvados/keep.py b/sdk/python/arvados/keep.py
index 71e101cf4c..86a28f54c4 100644
--- a/sdk/python/arvados/keep.py
+++ b/sdk/python/arvados/keep.py
@@ -377,7 +377,7 @@ class KeepClient(object):
                         curl.setopt(pycurl.SSL_VERIFYPEER, 0)
                     if method == "HEAD":
                         curl.setopt(pycurl.NOBODY, True)
-                    self._setcurltimeouts(curl, timeout)
+                    self._setcurltimeouts(curl, timeout, method=="HEAD")
 
                     try:
                         curl.perform()
@@ -421,6 +421,10 @@ class KeepClient(object):
                 _logger.info("HEAD %s: %s bytes",
                          self._result['status_code'],
                          self._result.get('content-length'))
+                if self._result['headers'].get('x-keep-locator'):
+                    # This is a response to a remote block copy request, return
+                    # the local copy block locator.
+                    return self._result['headers'].get('x-keep-locator')
                 return True
 
             _logger.info("GET %s: %s bytes in %s msec (%.3f MiB/sec)",
@@ -512,7 +516,7 @@ class KeepClient(object):
                 self.upload_counter.add(len(body))
             return True
 
-        def _setcurltimeouts(self, curl, timeouts):
+        def _setcurltimeouts(self, curl, timeouts, ignore_bandwidth=False):
             if not timeouts:
                 return
             elif isinstance(timeouts, tuple):
@@ -525,8 +529,9 @@ class KeepClient(object):
                 conn_t, xfer_t = (timeouts, timeouts)
                 bandwidth_bps = KeepClient.DEFAULT_TIMEOUT[2]
             curl.setopt(pycurl.CONNECTTIMEOUT_MS, int(conn_t*1000))
-            curl.setopt(pycurl.LOW_SPEED_TIME, int(math.ceil(xfer_t)))
-            curl.setopt(pycurl.LOW_SPEED_LIMIT, int(math.ceil(bandwidth_bps)))
+            if not ignore_bandwidth:
+                curl.setopt(pycurl.LOW_SPEED_TIME, int(math.ceil(xfer_t)))
+                curl.setopt(pycurl.LOW_SPEED_LIMIT, int(math.ceil(bandwidth_bps)))
 
         def _headerfunction(self, header_line):
             if isinstance(header_line, bytes):
@@ -786,6 +791,7 @@ class KeepClient(object):
 
         if local_store:
             self.local_store = local_store
+            self.head = self.local_store_head
             self.get = self.local_store_get
             self.put = self.local_store_put
         else:
@@ -975,6 +981,11 @@ class KeepClient(object):
         else:
             return None
 
+    def refresh_signature(self, loc):
+        """Ask Keep to get the remote block and return its local signature"""
+        now = datetime.datetime.utcnow().isoformat("T") + 'Z'
+        return self.head(loc, headers={'X-Keep-Signature': 'local, {}'.format(now)})
+
     @retry.retry_method
     def head(self, loc_s, **kwargs):
         return self._get_or_head(loc_s, method="HEAD", **kwargs)
@@ -983,7 +994,7 @@ class KeepClient(object):
     def get(self, loc_s, **kwargs):
         return self._get_or_head(loc_s, method="GET", **kwargs)
 
-    def _get_or_head(self, loc_s, method="GET", num_retries=None, request_id=None):
+    def _get_or_head(self, loc_s, method="GET", num_retries=None, request_id=None, headers=None):
         """Get data from Keep.
 
         This method fetches one or more blocks of data from Keep.  It
@@ -1024,11 +1035,11 @@ class KeepClient(object):
 
             self.misses_counter.add(1)
 
-            headers = {
-                'X-Request-Id': (request_id or
-                                 (hasattr(self, 'api_client') and self.api_client.request_id) or
-                                 arvados.util.new_request_id()),
-            }
+            if headers is None:
+                headers = {}
+            headers['X-Request-Id'] = (request_id or
+                                        (hasattr(self, 'api_client') and self.api_client.request_id) or
+                                        arvados.util.new_request_id())
 
             # If the locator has hints specifying a prefix (indicating a
             # remote keepproxy) or the UUID of a local gateway service,
@@ -1085,10 +1096,7 @@ class KeepClient(object):
 
             # Always cache the result, then return it if we succeeded.
             if loop.success():
-                if method == "HEAD":
-                    return True
-                else:
-                    return blob
+                return blob
         finally:
             if slot is not None:
                 slot.set(blob)
@@ -1109,7 +1117,7 @@ class KeepClient(object):
                 "{} not found".format(loc_s), service_errors)
         else:
             raise arvados.errors.KeepReadError(
-                "failed to read {}".format(loc_s), service_errors, label="service")
+                "failed to read {} after {}".format(loc_s, loop.attempts_str()), service_errors, label="service")
 
     @retry.retry_method
     def put(self, data, copies=2, num_retries=None, request_id=None):
@@ -1188,8 +1196,8 @@ class KeepClient(object):
                               for key in sorted_roots
                               if roots_map[key].last_result()['error'])
             raise arvados.errors.KeepWriteError(
-                "failed to write {} (wanted {} copies but wrote {})".format(
-                    data_hash, copies, writer_pool.done()), service_errors, label="service")
+                "failed to write {} after {} (wanted {} copies but wrote {})".format(
+                    data_hash, loop.attempts_str(), copies, writer_pool.done()), service_errors, label="service")
 
     def local_store_put(self, data, copies=1, num_retries=None):
         """A stub for put().
@@ -1223,5 +1231,17 @@ class KeepClient(object):
         with open(os.path.join(self.local_store, locator.md5sum), 'rb') as f:
             return f.read()
 
+    def local_store_head(self, loc_s, num_retries=None):
+        """Companion to local_store_put()."""
+        try:
+            locator = KeepLocator(loc_s)
+        except ValueError:
+            raise arvados.errors.NotFoundError(
+                "Invalid data locator: '%s'" % loc_s)
+        if locator.md5sum == config.EMPTY_BLOCK_LOCATOR.split('+')[0]:
+            return True
+        if os.path.exists(os.path.join(self.local_store, locator.md5sum)):
+            return True
+
     def is_cached(self, locator):
         return self.block_cache.reserve_cache(expect_hash)