From daab3274ba03f4df0160374fa9fb5993847f3153 Mon Sep 17 00:00:00 2001

From: Kotresh HR <khiremat@redhat.com>

Date: Wed, 9 May 2018 01:10:29 -0400

Subject: [PATCH 267/271] geo-rep/scheduler: Fix crash

Fix crash where session_name is referenced

before assignment. Well, this is a corner

case where the geo-rep session exists and

the status output doesn't show any rows.

This might happen when glusterd is down

or when the system is in inconsistent state

w.r.t glusterd.

Upstream reference:

 > fixes: bz#1576179

 > Change-Id: Iec1557e01b35068041b4b3c1aacee2bfa0e05873

 > Patch: https://review.gluster.org/19991

BUG: 1575877

Change-Id: Iec1557e01b35068041b4b3c1aacee2bfa0e05873

Signed-off-by: Kotresh HR <khiremat@redhat.com>

Reviewed-on: https://code.engineering.redhat.com/gerrit/138668

Tested-by: RHGS Build Bot <nigelb@redhat.com>

Reviewed-by: Sunil Kumar Heggodu Gopala Acharya <sheggodu@redhat.com>

---

 extras/geo-rep/schedule_georep.py.in | 70 ++++++++++++++++++------------------

 1 file changed, 35 insertions(+), 35 deletions(-)

diff --git a/extras/geo-rep/schedule_georep.py.in b/extras/geo-rep/schedule_georep.py.in

index 1887fe7..4e1a071 100644

--- a/extras/geo-rep/schedule_georep.py.in

+++ b/extras/geo-rep/schedule_georep.py.in

@@ -297,6 +297,7 @@ def get_summary(mastervol, slave_url):

     status_data = get(mastervol, slave_url)

     for session in status_data:

+        session_name = ""

         summary = {

             "active": 0,

             "passive": 0,

@@ -339,7 +340,8 @@ def get_summary(mastervol, slave_url):

         if summary["faulty"] == 0 and summary["offline"] == 0:

             summary["ok"] = True

-        out.append([session_name, summary, faulty_rows, down_rows])

+        if session_name != "":

+                out.append([session_name, summary, faulty_rows, down_rows])

     return out

@@ -397,41 +399,39 @@ def main(args):

             # or any other error. Gluster cmd still produces XML output

             # with different message

             output_warning("Unable to get Geo-replication Status")

-            time.sleep(1)

-            continue

-

-        session_name, summary, faulty_rows, down_rows = session_summary[0]

-        chkpt_status = "COMPLETE" if summary["checkpoints_ok"] else \

-                       "NOT COMPLETE"

-        ok_status = "OK" if summary["ok"] else "NOT OK"

-

-        if summary["ok"]:

-            output_ok("All Checkpoints {1}, "

-                      "All status {2} (Turns {0:>3})".format(

-                          turns, chkpt_status, ok_status))

         else:

-            output_warning("All Checkpoints {1}, "

-                           "All status {2} (Turns {0:>3})".format(

-                               turns, chkpt_status, ok_status))

-

-            output_warning("Geo-rep workers Faulty/Offline, "

-                           "Faulty: {0} Offline: {1}".format(

-                               repr(faulty_rows),

-                               repr(down_rows)))

-

-        if summary["checkpoints_ok"]:

-            output_ok("Stopping Geo-replication session now")

-            cmd = ["@SBIN_DIR@/gluster", "volume", "geo-replication",

-                   args.mastervol,

-                   "%s::%s" % (args.slave, args.slavevol), "stop"]

-            execute(cmd)

-            break

-        else:

-            # If Checkpoint is not complete after a iteration means brick

-            # was down and came online now. SETATTR on mount is not

-            # recorded, So again issue touch on mount root So that

-            # Stime will increase and Checkpoint will complete.

-            touch_mount_root(args.mastervol)

+            session_name, summary, faulty_rows, down_rows = session_summary[0]

+            chkpt_status = "COMPLETE" if summary["checkpoints_ok"] else \

+                           "NOT COMPLETE"

+            ok_status = "OK" if summary["ok"] else "NOT OK"

+

+            if summary["ok"]:

+                output_ok("All Checkpoints {1}, "

+                          "All status {2} (Turns {0:>3})".format(

+                              turns, chkpt_status, ok_status))

+            else:

+                output_warning("All Checkpoints {1}, "

+                               "All status {2} (Turns {0:>3})".format(

+                                   turns, chkpt_status, ok_status))

+

+                output_warning("Geo-rep workers Faulty/Offline, "

+                               "Faulty: {0} Offline: {1}".format(

+                                   repr(faulty_rows),

+                                   repr(down_rows)))

+

+            if summary["checkpoints_ok"]:

+                output_ok("Stopping Geo-replication session now")

+                cmd = ["@SBIN_DIR@/gluster", "volume", "geo-replication",

+                       args.mastervol,

+                       "%s::%s" % (args.slave, args.slavevol), "stop"]

+                execute(cmd)

+                break

+            else:

+                # If Checkpoint is not complete after a iteration means brick

+                # was down and came online now. SETATTR on mount is not

+                # recorded, So again issue touch on mount root So that

+                # Stime will increase and Checkpoint will complete.

+                touch_mount_root(args.mastervol)

         # Increment the turns and Sleep for 10 sec

         turns += 1

-- 

1.8.3.1