From 670299cf4fc90c1c61032057b6ff570fee7a448e Mon Sep 17 00:00:00 2001 From: Mohammed Rafi KC Date: Mon, 24 Jun 2019 12:00:20 +0530 Subject: glusterd/svc: update pid of mux volumes from the shd process For a normal volume, we are updating the pid from a the process while we do a daemonization or at the end of the init if it is no-daemon mode. Along with updating the pid we also lock the file, to make sure that the process is running fine. With brick mux, we were updating the pidfile from gluterd after an attach/detach request. There are two problems with this approach. 1) We are not holding a pidlock for any file other than parent process. 2) There is a chance for possible race conditions with attach/detach. For example, shd start and a volume stop could race. Let's say we are starting an shd and it is attached to a volume. While we trying to link the pid file to the running process, this would have deleted by the thread that doing a volume stop. Change-Id: I29a00352102877ce09ea3f376ca52affceb5cf1a Updates: bz#1722541 Signed-off-by: Mohammed Rafi KC --- xlators/mgmt/glusterd/src/glusterd-svc-helper.c | 57 +++++++++++++------------ 1 file changed, 30 insertions(+), 27 deletions(-) (limited to 'xlators/mgmt/glusterd/src/glusterd-svc-helper.c') diff --git a/xlators/mgmt/glusterd/src/glusterd-svc-helper.c b/xlators/mgmt/glusterd/src/glusterd-svc-helper.c index cde77e5f6af..b98892af61e 100644 --- a/xlators/mgmt/glusterd/src/glusterd-svc-helper.c +++ b/xlators/mgmt/glusterd/src/glusterd-svc-helper.c @@ -517,7 +517,7 @@ glusterd_shd_svc_mux_init(glusterd_volinfo_t *volinfo, glusterd_svc_t *svc) /* Take first entry from the process */ parent_svc = cds_list_entry(mux_proc->svcs.next, glusterd_svc_t, mux_svc); - sys_link(parent_svc->proc.pidfile, svc->proc.pidfile); + glusterd_copy_file(parent_svc->proc.pidfile, svc->proc.pidfile); mux_conn = &parent_svc->conn; if (volinfo) volinfo->shd.attached = _gf_true; @@ -621,12 +621,9 @@ glusterd_svc_attach_cbk(struct rpc_req *req, struct iovec *iov, int count, glusterd_volinfo_t *volinfo = NULL; glusterd_shdsvc_t *shd = NULL; glusterd_svc_t *svc = frame->cookie; - glusterd_svc_t *parent_svc = NULL; - glusterd_svc_proc_t *mux_proc = NULL; glusterd_conf_t *conf = NULL; int *flag = (int *)frame->local; xlator_t *this = THIS; - int pid = -1; int ret = -1; gf_getspec_rsp rsp = { 0, @@ -677,27 +674,7 @@ glusterd_svc_attach_cbk(struct rpc_req *req, struct iovec *iov, int count, } if (rsp.op_ret == 0) { - pthread_mutex_lock(&conf->attach_lock); - { - if (!strcmp(svc->name, "glustershd")) { - mux_proc = svc->svc_proc; - if (mux_proc && - !gf_is_service_running(svc->proc.pidfile, &pid)) { - /* - * When svc's are restarting, there is a chance that the - * attached svc might not have updated it's pid. Because - * it was at connection stage. So in that case, we need - * to retry the pid file copy. - */ - parent_svc = cds_list_entry(mux_proc->svcs.next, - glusterd_svc_t, mux_svc); - if (parent_svc) - sys_link(parent_svc->proc.pidfile, svc->proc.pidfile); - } - } - svc->online = _gf_true; - } - pthread_mutex_unlock(&conf->attach_lock); + svc->online = _gf_true; gf_msg(this->name, GF_LOG_INFO, 0, GD_MSG_SVC_ATTACH_FAIL, "svc %s of volume %s attached successfully to pid %d", svc->name, volinfo->volname, glusterd_proc_get_pid(&svc->proc)); @@ -724,7 +701,7 @@ out: extern size_t build_volfile_path(char *volume_id, char *path, size_t path_len, - char *trusted_str); + char *trusted_str, dict_t *dict); int __glusterd_send_svc_configure_req(glusterd_svc_t *svc, int flags, @@ -749,6 +726,7 @@ __glusterd_send_svc_configure_req(glusterd_svc_t *svc, int flags, ssize_t req_size = 0; call_frame_t *frame = NULL; gd1_mgmt_brick_op_req brick_req; + dict_t *dict = NULL; void *req = &brick_req; void *errlbl = &&err; struct rpc_clnt_connection *conn; @@ -774,6 +752,8 @@ __glusterd_send_svc_configure_req(glusterd_svc_t *svc, int flags, brick_req.name = volfile_id; brick_req.input.input_val = NULL; brick_req.input.input_len = 0; + brick_req.dict.dict_val = NULL; + brick_req.dict.dict_len = 0; frame = create_frame(this, this->ctx->pool); if (!frame) { @@ -781,7 +761,13 @@ __glusterd_send_svc_configure_req(glusterd_svc_t *svc, int flags, } if (op == GLUSTERD_SVC_ATTACH) { - (void)build_volfile_path(volfile_id, path, sizeof(path), NULL); + dict = dict_new(); + if (!dict) { + ret = -ENOMEM; + goto *errlbl; + } + + (void)build_volfile_path(volfile_id, path, sizeof(path), NULL, dict); ret = sys_stat(path, &stbuf); if (ret < 0) { @@ -816,6 +802,18 @@ __glusterd_send_svc_configure_req(glusterd_svc_t *svc, int flags, ret = -EIO; goto *errlbl; } + if (dict->count > 0) { + ret = dict_allocate_and_serialize(dict, &brick_req.dict.dict_val, + &brick_req.dict.dict_len); + if (ret) { + gf_msg(this->name, GF_LOG_ERROR, 0, + GD_MSG_DICT_SERL_LENGTH_GET_FAIL, + "Failed to serialize dict " + "to request buffer"); + goto *errlbl; + } + dict->extra_free = brick_req.dict.dict_val; + } frame->cookie = svc; frame->local = GF_CALLOC(1, sizeof(int), gf_gld_mt_int); @@ -860,6 +858,8 @@ __glusterd_send_svc_configure_req(glusterd_svc_t *svc, int flags, GF_ATOMIC_INC(conf->blockers); ret = rpc_clnt_submit(rpc, &gd_brick_prog, op, cbkfn, &iov, 1, NULL, 0, iobref, frame, NULL, 0, NULL, 0, NULL); + if (dict) + dict_unref(dict); GF_FREE(volfile_content); if (spec_fd >= 0) sys_close(spec_fd); @@ -872,6 +872,9 @@ maybe_free_iobuf: iobuf_unref(iobuf); } err: + if (dict) + dict_unref(dict); + GF_FREE(volfile_content); if (spec_fd >= 0) sys_close(spec_fd); -- cgit