Add more logging for zero byte reads

Azure · Oct 15, 2020 · 9339b26 · 9339b26
1 parent c5557ec
commit 9339b26
Show file tree

Hide file tree

Showing 3 changed files with 15 additions and 5 deletions.
diff --git a/HISTORY.rst b/HISTORY.rst
@@ -3,6 +3,10 @@
 Release History
 ===============
 
+0.0.51 (2020-10-15)
++++++++++++++++++++
+* Add more logging for zero byte reads to investigate root cause.
+
 0.0.50 (2020-09-10)
 +++++++++++++++++++
 * Fix bug with retrying for ADAL exception parsing.

diff --git a/azure/datalake/store/__init__.py b/azure/datalake/store/__init__.py
@@ -6,7 +6,7 @@
 # license information.
 # --------------------------------------------------------------------------
 
-__version__ = "0.0.50"
+__version__ = "0.0.51"
 
 from .core import AzureDLFileSystem
 from .multithread import ADLDownloader

diff --git a/azure/datalake/store/core.py b/azure/datalake/store/core.py
@@ -1135,7 +1135,9 @@ def _read_blocksize(self, offset=-1):
             self.end = self.size
             self.cache = b""
             return
-        if offset >= self.start and offset < self.end:
+        if self.start <= offset < self.end:
+            logger.info("Read offset {offset} is within cache {start}-{end}. "
+                        "Not going to server.".format(offset=offset, start=self.start, end=self.end))
             return
         if offset > self.size:
             raise ValueError('Read offset is outside the File')
@@ -1165,9 +1167,13 @@ def read(self, length=-1):
             if not data_read:  # Check to catch possible server errors. Ideally shouldn't happen.
                 flag += 1
                 if flag >= 5:
-                    raise DatalakeIncompleteTransferException('Could not read data: {}. '
-                                                              'Repeated zero byte reads. '
-                                                              'Possible file corruption'.format(self.path))
+                    exception_string = "Current Location:{loc}, " \
+                                       "File Size:{size}, Cache Start:{start}, " \
+                                       "Cache End:{end}".format(loc=self.loc, size=self.size,
+                                                                start=self.start, end=self.end)
+                    raise DatalakeIncompleteTransferException('Could not read data: {path}. '
+                                                              'Repeated zero byte reads. Possible file corruption. File Details'
+                                                              '{details}'.format(path=self.path, details=exception_string))
             out += data_read
             self.loc += len(data_read)
             length -= len(data_read)