tools/glusterfind: ignoring deleted files

OSError and IOError exceptions were being thrown if files
were deleted after session was created and a subsequent
glusterfind pre was attmepted. glusterfind now detects
this scenario and safely ignores these changes to the
file-system. We also avoid recording deleted file paths
into database in the case where gfid to path resolution
cannot be performed for deleted files.

Also, we now turn on volume option to capture delete paths.

Change-Id: Iaf3883463f5e64643497bd77229ad80a9b2d5da5
BUG: 1228111
Signed-off-by: Milind Changire <mchangir@redhat.com>
Reviewed-on: http://review.gluster.org/11194
Reviewed-by: Aravinda VK <avishwan@redhat.com>
Tested-by: NetBSD Build System <jenkins@build.gluster.org>

2 files changed, 43 insertions, 24 deletions

diff --git a/tools/glusterfind/src/changelog.py b/tools/glusterfind/src/changelog.py
index fd2384beccf..5939d487a4d 100644
--- a/tools/glusterfind/src/changelog.py
+++ b/tools/glusterfind/src/changelog.py
@@ -61,10 +61,13 @@ def pgfid_to_path(brick, changelog_data):
         if row[0] == "":
             continue
 
-        path = symlink_gfid_to_path(brick, row[0])
-        path = output_path_prepare(path, args.output_prefix)
-
-        changelog_data.gfidpath_set_path1(path, row[0])
+        try:
+            path = symlink_gfid_to_path(brick, row[0])
+            path = output_path_prepare(path, args.output_prefix)
+            changelog_data.gfidpath_set_path1(path, row[0])
+        except (IOError, OSError) as e:
+            logger.warn("Error converting to path: %s" % e)
+            continue
 
     # pgfid2 to path2 in case of RENAME
     for row in changelog_data.gfidpath_get_distinct("pgfid2",
@@ -74,13 +77,14 @@ def pgfid_to_path(brick, changelog_data):
         if row[0] == "":
             continue
 
-        path = symlink_gfid_to_path(brick, row[0])
-        if path == "":
+        try:
+            path = symlink_gfid_to_path(brick, row[0])
+            path = output_path_prepare(path, args.output_prefix)
+            changelog_data.gfidpath_set_path2(path, row[0])
+        except (IOError, OSError) as e:
+            logger.warn("Error converting to path: %s" % e)
             continue
 
-        path = output_path_prepare(path, args.output_prefix)
-        changelog_data.gfidpath_set_path2(path, row[0])
-
 
 def populate_pgfid_and_inodegfid(brick, changelog_data):
     """
@@ -94,14 +98,14 @@ def populate_pgfid_and_inodegfid(brick, changelog_data):
         p = os.path.join(brick, ".glusterfs", gfid[0:2], gfid[2:4], gfid)
         if os.path.islink(p):
             # It is a Directory if GFID backend path is symlink
-            path = symlink_gfid_to_path(brick, gfid)
-            if path == "":
+            try:
+                path = symlink_gfid_to_path(brick, gfid)
+                path = output_path_prepare(path, args.output_prefix)
+                changelog_data.gfidpath_update({"path1": path},
+                                                {"gfid": gfid})
+            except (IOError, OSError) as e:
+                logger.warn("Error converting to path: %s" % e)
                 continue
-
-            path = output_path_prepare(path, args.output_prefix)
-
-            changelog_data.gfidpath_update({"path1": path},
-                                           {"gfid": gfid})
         else:
             try:
                 # INODE and GFID to inodegfid table
@@ -161,12 +165,16 @@ def gfid_to_path_using_pgfid(brick, changelog_data, args):
                    conf.get_opt("brick_ignore_dirs").split(",")]
 
     for row in changelog_data.pgfid_get():
-        path = symlink_gfid_to_path(brick, row[0])
-        find(os.path.join(brick, path),
-             callback_func=output_callback,
-             filter_func=inode_filter,
-             ignore_dirs=ignore_dirs,
-             subdirs_crawl=False)
+        try:
+            path = symlink_gfid_to_path(brick, row[0])
+            find(os.path.join(brick, path),
+                callback_func=output_callback,
+                filter_func=inode_filter,
+                ignore_dirs=ignore_dirs,
+                subdirs_crawl=False)
+        except (IOError, OSError) as e:
+            logger.warn("Error converting to path: %s" % e)
+            continue
 
 
 def gfid_to_path_using_batchfind(brick, changelog_data):
@@ -291,8 +299,12 @@ def get_changes(brick, hash_dir, log_file, start, end, args):
                 # again in list
                 if change.endswith(".%s" % start):
                     continue
-                parse_changelog_to_db(changelog_data, change)
-                libgfchangelog.cl_history_done(change)
+                try:
+                    parse_changelog_to_db(changelog_data, change)
+                    libgfchangelog.cl_history_done(change)
+                except IOError as e:
+                    logger.warn("Error parsing changelog file %s: %s" %
+                        (change, e))
 
             changelog_data.commit()
     except libgfchangelog.ChangelogException as e:
diff --git a/tools/glusterfind/src/main.py b/tools/glusterfind/src/main.py
index 850024be2fe..5d5058f761b 100644
--- a/tools/glusterfind/src/main.py
+++ b/tools/glusterfind/src/main.py
@@ -352,6 +352,13 @@ def mode_create(session_dir, args):
         logger.info("Volume option set %s, changelog.changelog on"
                     % args.volume)
 
+        execute(["gluster", "volume", "set",
+                 args.volume, "changelog.capture-del-path", "on"],
+                exit_msg="Failed to set volume option "
+                "changelog.capture-del-path on", logger=logger)
+        logger.info("Volume option set %s, changelog.capture-del-path on"
+                    % args.volume)
+
     # Add Rollover time to current time to make sure changelogs
     # will be available if we use this time as start time
     time_to_update = int(time.time()) + get_changelog_rollover_time(