[Gluster-devel] Question on tests/basic/pump.t

[Emmanuel Dreyfus]

NetBSD failed tests/basic/pump.t but I have trouble to understand the reason.

It loops on
volume replace-brick: failed: Commit failed on localhost. Please check the log
file for more details.

The brick log loops on:

[2014-08-31 03:53:20.081416] I [server-handshake.c:585:server_setvolume]
0-patchy-server: accepted client from
netbsd0.cloud.gluster.org-2518-2014/08/31-03:53:20:64143-mnt-client-0-0
(version: 3.7dev)
[2014-08-31 03:53:20.173799] E [client-handshake.c:1544:client_query_portmap]
0-patchy-replace-brick: remote-subvolume not set in volfile
[2014-08-31 03:53:20.174099] I [client.c:2215:client_rpc_notify]
0-patchy-replace-brick: disconnected from patchy-replace-brick. Client process
will keep trying to connect to glusterd until brick's port is available

I patched to add the requested option:

--- a/xlators/mgmt/glusterd/src/glusterd-volgen.c
+++ b/xlators/mgmt/glusterd/src/glusterd-volgen.c
@@ -1659,6 +1659,13 @@ server_graph_builder (volgen_graph_t *graph, glusterd_vol
                 if (NULL == ptranst)
                         return -1;
 
+               ret = xlator_set_option (rbxl, "remote-subvolume", path);
+               if (ret) {
+                       gf_log ("glusterd", GF_LOG_WARNING,
+                               "failed to set remore-subvolume");
+                       return -1;
+               }
+
                 if (dict_get_str (set_dict, SSL_CERT_DEPTH_OPT, &value) == 0) {
                         ret = xlator_set_option (rbxl, "ssl-cert-depth", value)
                         if (ret) {


But now glusterd loops on:

[2014-08-31 04:53:18.958811] I
[glusterd-replace-brick.c:99:__glusterd_handle_replace_brick] 0-management:
Received replace brick req
[2014-08-31 04:53:18.958870] I
[glusterd-replace-brick.c:154:__glusterd_handle_replace_brick] 0-management:
Received replace brick status request
[2014-08-31 04:53:18.958920] W [glusterd-locks.c:547:glusterd_mgmt_v3_lock]
0-management: Lock for patchy held by 46ce553f-9746-4969-9dc7-b112caaeac01
[2014-08-31 04:53:16.509208] E [glusterd-handler.c:595:glusterd_op_txn_begin]
0-management: Unable to acquire lock for patchy

46ce553f-9746-4969-9dc7-b112caaeac01 is the brick. I assume it acquired a lock
it never released, but where? Any hint of how this is supposed to work?

-- 
Emmanuel Dreyfus
http://hcpnet.free.fr/pubz
manu at netbsd.org