Blob Blame History Raw
From 029fbbdaa7c4ddcc2479f507345a5c3ab1035313 Mon Sep 17 00:00:00 2001
From: Ravishankar N <ravishankar@redhat.com>
Date: Mon, 2 Jul 2018 16:05:39 +0530
Subject: [PATCH 306/325] glusterfsd: Do not process GLUSTERD_BRICK_XLATOR_OP
 if graph is not ready

Patch in upstream master: https://review.gluster.org/#/c/20435/
Patch in release-3.12: https://review.gluster.org/#/c/20436/

Problem:
If glustershd gets restarted by glusterd due to node reboot/volume start force/
or any thing that changes shd graph (add/remove brick), and index heal
is launched via CLI, there can be a chance that shd receives this IPC
before the graph is fully active. Thus when it accesses
glusterfsd_ctx->active, it crashes.

Fix:
Since glusterd does not really wait for the daemons it spawned to be
fully initialized and can send the request as soon as rpc initialization has
succeeded, we just handle it at shd. If glusterfs_graph_activate() is
not yet done in shd but glusterd sends GD_OP_HEAL_VOLUME to shd,
we fail the request.

Change-Id: If6cc07bc5455c4ba03458a36c28b63664496b17d
BUG: 1593865
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/143097
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
---
 glusterfsd/src/glusterfsd-messages.h | 4 +++-
 glusterfsd/src/glusterfsd-mgmt.c     | 6 ++++++
 2 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/glusterfsd/src/glusterfsd-messages.h b/glusterfsd/src/glusterfsd-messages.h
index e9c28f7..e38a88b 100644
--- a/glusterfsd/src/glusterfsd-messages.h
+++ b/glusterfsd/src/glusterfsd-messages.h
@@ -36,7 +36,7 @@
  */
 
 #define GLFS_COMP_BASE          GLFS_MSGID_COMP_GLUSTERFSD
-#define GLFS_NUM_MESSAGES       37
+#define GLFS_NUM_MESSAGES       38
 #define GLFS_MSGID_END          (GLFS_COMP_BASE + GLFS_NUM_MESSAGES + 1)
 /* Messaged with message IDs */
 #define glfs_msg_start_x GLFS_COMP_BASE, "Invalid: Start of messages"
@@ -109,6 +109,8 @@
 #define glusterfsd_msg_36 (GLFS_COMP_BASE + 36), "problem in xlator " \
                         " loading."
 #define glusterfsd_msg_37 (GLFS_COMP_BASE + 37), "failed to get dict value"
+#define glusterfsd_msg_38 (GLFS_COMP_BASE + 38), "Not processing brick-op no."\
+                        " %d since volume graph is not yet active."
 
 /*------------*/
 #define glfs_msg_end_x GLFS_MSGID_END, "Invalid: End of messages"
diff --git a/glusterfsd/src/glusterfsd-mgmt.c b/glusterfsd/src/glusterfsd-mgmt.c
index 665b62c..2167241 100644
--- a/glusterfsd/src/glusterfsd-mgmt.c
+++ b/glusterfsd/src/glusterfsd-mgmt.c
@@ -790,6 +790,12 @@ glusterfs_handle_translator_op (rpcsvc_request_t *req)
 
         ctx = glusterfsd_ctx;
         active = ctx->active;
+        if (!active) {
+                ret = -1;
+                gf_msg (this->name, GF_LOG_ERROR, EAGAIN, glusterfsd_msg_38,
+                        xlator_req.op);
+                goto out;
+        }
         any = active->first;
         input = dict_new ();
         ret = dict_unserialize (xlator_req.input.input_val,
-- 
1.8.3.1