diff options
author | Atin Mukherjee <amukherj@redhat.com> | 2017-10-17 21:32:44 +0530 |
---|---|---|
committer | Atin Mukherjee <amukherj@redhat.com> | 2017-10-31 04:36:44 +0000 |
commit | 30e0b86aae00430823f2523c6efa3c4ebbf0a478 (patch) | |
tree | 49920feaa971f799fc3eef7be31fc80743b11f47 /xlators/mgmt/glusterd/src/glusterd-pmap.c | |
parent | 15fe99995b8650a677b097028bc14d61a5dd5e1b (diff) |
glusterd: clean up portmap on brick disconnect
GlusterD's portmap entry for a brick is cleaned up when a PMAP_SIGNOUT event is
initiated by the brick process at the shutdown. But if the brick process crashes
or gets killed through SIGKILL then this event is not initiated and glusterd
ends up with a stale port. Since GlusterD's portmap traversal happens both ways,
forward for allocation and backward for registry search, there is a possibility
that glusterd might end up running with a stale port for a brick which
eventually will end up with clients to fail to connect to the bricks.
Solution is to clean up the port entry in case the process is down as
part of the brick disconnect event. Although with this the handling
PMAP_SIGNOUT event becomes redundant in most of the cases, but this is
the safeguard method to avoid glusterd getting into the stale port
issues.
Change-Id: I04c5be6d11e772ee4de16caf56dbb37d5c944303
BUG: 1503246
Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
Diffstat (limited to 'xlators/mgmt/glusterd/src/glusterd-pmap.c')
-rw-r--r-- | xlators/mgmt/glusterd/src/glusterd-pmap.c | 26 |
1 files changed, 17 insertions, 9 deletions
diff --git a/xlators/mgmt/glusterd/src/glusterd-pmap.c b/xlators/mgmt/glusterd/src/glusterd-pmap.c index 6d8cf3f894f..4f045ab17be 100644 --- a/xlators/mgmt/glusterd/src/glusterd-pmap.c +++ b/xlators/mgmt/glusterd/src/glusterd-pmap.c @@ -241,7 +241,8 @@ pmap_assign_port (xlator_t *this, int old_port, const char *path) if (old_port) { ret = pmap_registry_remove (this, 0, path, - GF_PMAP_PORT_BRICKSERVER, NULL); + GF_PMAP_PORT_BRICKSERVER, NULL, + _gf_false); if (ret) { gf_msg (this->name, GF_LOG_WARNING, GD_MSG_PMAP_REGISTRY_REMOVE_FAIL, 0, "Failed to" @@ -344,7 +345,8 @@ pmap_registry_extend (xlator_t *this, int port, const char *brickname) int pmap_registry_remove (xlator_t *this, int port, const char *brickname, - gf_pmap_port_type_t type, void *xprt) + gf_pmap_port_type_t type, void *xprt, + gf_boolean_t brick_disconnect) { struct pmap_registry *pmap = NULL; int p = 0; @@ -391,11 +393,16 @@ remove: * can delete the entire entry. */ if (!pmap->ports[p].xprt) { - brick_str = pmap->ports[p].brickname; - if (brick_str) { - while (*brick_str != '\0') { - if (*(brick_str++) != ' ') { - goto out; + /* If the signout call is being triggered by brick disconnect + * then clean up all the bricks (in case of brick mux) + */ + if (!brick_disconnect) { + brick_str = pmap->ports[p].brickname; + if (brick_str) { + while (*brick_str != '\0') { + if (*(brick_str++) != ' ') { + goto out; + } } } } @@ -550,14 +557,15 @@ __gluster_pmap_signout (rpcsvc_request_t *req) goto fail; } rsp.op_ret = pmap_registry_remove (THIS, args.port, args.brick, - GF_PMAP_PORT_BRICKSERVER, req->trans); + GF_PMAP_PORT_BRICKSERVER, req->trans, + _gf_false); ret = glusterd_get_brickinfo (THIS, args.brick, args.port, &brickinfo); if (args.rdma_port) { snprintf(brick_path, PATH_MAX, "%s.rdma", args.brick); rsp.op_ret = pmap_registry_remove (THIS, args.rdma_port, brick_path, GF_PMAP_PORT_BRICKSERVER, - req->trans); + req->trans, _gf_false); } /* Update portmap status on brickinfo */ if (brickinfo) |