glusterd: allow multiple instances of glusterd on one machine

This is needed to support automated testing of cluster-communication
features such as probing and quorum.  In order to use this, you need to
do the following preparatory steps.

* Copy /var/lib/glusterd to another directory for each virtual host

* Ensure that each virtual host has a different UUID in its glusterd.info

Now you can start each copy of glusterd with the following xlator-options.

* management.transport.socket.bind-address=$ip_address

* management.working-directory=$unique_working_directory

You can use 127.x.y.z addresses for binding without needing to assign
them to interfaces explicitly.  Note that you must use addresses, not
names, because of some stuff in the socket code that's not worth fixing
just for this usage, but after that you can use names in /etc/hosts
instead.

At this point you can issue CLI commands to a specific glusterd using
the --remote-host option.  So far probe, volume create/start/stop,
mount, and basic I/O all seem to work as expected with multiple
instances.

Change-Id: I1beabb44cff8763d2774bc208b2ffcda27c1a550
BUG: 913555
Signed-off-by: Jeff Darcy <jdarcy@redhat.com>
Reviewed-on: http://review.gluster.org/4556
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Anand Avati <avati@redhat.com>

1 files changed, 25 insertions, 1 deletions

diff --git a/xlators/mgmt/glusterd/src/glusterd-handler.c b/xlators/mgmt/glusterd/src/glusterd-handler.c
index 7d39c1e0785..04e5833b732 100644
--- a/xlators/mgmt/glusterd/src/glusterd-handler.c
+++ b/xlators/mgmt/glusterd/src/glusterd-handler.c
@@ -707,6 +707,7 @@ glusterd_handle_cli_probe (rpcsvc_request_t *req)
         glusterd_peerinfo_t             *peerinfo = NULL;
         gf_boolean_t                    run_fsm = _gf_true;
         xlator_t                        *this = NULL;
+        char                            *bind_name = NULL;
 
         GF_ASSERT (req);
         this = THIS;
@@ -736,7 +737,16 @@ glusterd_handle_cli_probe (rpcsvc_request_t *req)
         gf_log ("glusterd", GF_LOG_INFO, "Received CLI probe req %s %d",
                 cli_req.hostname, cli_req.port);
 
-        if (glusterd_is_local_addr(cli_req.hostname)) {
+        if (dict_get_str(this->options,"transport.socket.bind-address",
+                         &bind_name) == 0) {
+                gf_log ("glusterd", GF_LOG_DEBUG,
+                        "only checking probe address vs. bind address");
+                ret = glusterd_is_same_address(bind_name,cli_req.hostname);
+        }
+        else {
+                ret = glusterd_is_local_addr(cli_req.hostname);
+        }
+        if (ret) {
                 glusterd_xfer_cli_probe_resp (req, 0, GF_PROBE_LOCALHOST, NULL,
                                               cli_req.hostname, cli_req.port);
                 ret = 0;
@@ -2434,6 +2444,7 @@ glusterd_friend_rpc_create (xlator_t *this, glusterd_peerinfo_t *peerinfo,
         dict_t                 *options = NULL;
         int                    ret = -1;
         glusterd_peerctx_t     *peerctx = NULL;
+        data_t                 *data = NULL;
 
         peerctx = GF_CALLOC (1, sizeof (*peerctx), gf_gld_mt_peerctx_t);
         if (!peerctx)
@@ -2450,6 +2461,19 @@ glusterd_friend_rpc_create (xlator_t *this, glusterd_peerinfo_t *peerinfo,
         if (ret)
                 goto out;
 
+        /*
+         * For simulated multi-node testing, we need to make sure that we
+         * create our RPC endpoint with the same address that the peer would
+         * use to reach us.
+         */
+        if (this->options) {
+                data = dict_get(this->options,"transport.socket.bind-address");
+                if (data) {
+                        ret = dict_set(options,
+                                       "transport.socket.source-addr",data);
+                }
+        }
+
         ret = glusterd_rpc_create (&peerinfo->rpc, options,
                                    glusterd_peer_rpc_notify, peerctx);
         if (ret) {