From 941241b33f424a6c4b27883482e0c7b101f712c5 Mon Sep 17 00:00:00 2001 From: Jeff Darcy Date: Wed, 1 Feb 2017 21:54:30 -0500 Subject: [PATCH 321/361] glusterd: double-check whether brick is alive for stats With multiplexing, our tests detach bricks from their host processes without glusterd being involved. Thus, when we ask glusterd to fetch profile info, it will try to fetch from a brick that's actually not present any more. While it can handle the process being dead and its RPC connection being closed, it barfs if it gets a negative response from a live brick process. This is not a problem in normal use, because the brick can't disappear without glusterd seeing it. The fix is to double check that the brick is actually running, by looking for its pidfile which the tests *do* clean up as part of killing a brick. mainline: > BUG: 1385758 > Reviewed-on: https://review.gluster.org/16509 > Smoke: Gluster Build System > NetBSD-regression: NetBSD Build System > CentOS-regression: Gluster Build System > Reviewed-by: Atin Mukherjee (cherry picked from commit f4b94ddd3034f2ac27890f75ec28aa75b4fc18eb) BUG: 1417815 Change-Id: I098465b175ecf23538bd7207357c752a2bba8f4e Signed-off-by: Jeff Darcy Reviewed-on: https://code.engineering.redhat.com/gerrit/101302 Tested-by: Milind Changire Reviewed-by: Atin Mukherjee --- xlators/mgmt/glusterd/src/glusterd-op-sm.c | 17 ++++++++++++++--- 1 file changed, 14 insertions(+), 3 deletions(-) diff --git a/xlators/mgmt/glusterd/src/glusterd-op-sm.c b/xlators/mgmt/glusterd/src/glusterd-op-sm.c index a3a0462..ef31cdb 100644 --- a/xlators/mgmt/glusterd/src/glusterd-op-sm.c +++ b/xlators/mgmt/glusterd/src/glusterd-op-sm.c @@ -6247,15 +6247,14 @@ glusterd_bricks_select_profile_volume (dict_t *dict, char **op_errstr, glusterd_brickinfo_t *brickinfo = NULL; glusterd_pending_node_t *pending_node = NULL; char *brick = NULL; - - + int32_t pid = -1; + char pidfile[PATH_MAX] = {0}; this = THIS; GF_ASSERT (this); priv = this->private; GF_ASSERT (priv); - ret = dict_get_str (dict, "volname", &volname); if (ret) { gf_msg ("glusterd", GF_LOG_ERROR, 0, @@ -6386,6 +6385,18 @@ glusterd_bricks_select_profile_volume (dict_t *dict, char **op_errstr, cds_list_for_each_entry (brickinfo, &volinfo->bricks, brick_list) { if (glusterd_is_brick_started (brickinfo)) { + /* + * In normal use, glusterd_is_brick_started + * will give us the answer we need. However, + * in our tests the brick gets detached behind + * our back, so we need to double-check this + * way. + */ + GLUSTERD_GET_BRICK_PIDFILE (pidfile, volinfo, + brickinfo, priv); + if (!gf_is_service_running (pidfile, &pid)) { + continue; + } pending_node = GF_CALLOC (1, sizeof (*pending_node), gf_gld_mt_pending_node_t); if (!pending_node) { -- 1.8.3.1