1 // SPDX-License-Identifier: GPL-2.0
2 #include <linux/ceph/ceph_debug.h>
4 #include <linux/module.h>
5 #include <linux/types.h>
6 #include <linux/slab.h>
7 #include <linux/random.h>
8 #include <linux/sched.h>
10 #include <linux/ceph/ceph_features.h>
11 #include <linux/ceph/mon_client.h>
12 #include <linux/ceph/libceph.h>
13 #include <linux/ceph/debugfs.h>
14 #include <linux/ceph/decode.h>
15 #include <linux/ceph/auth.h>
18 * Interact with Ceph monitor cluster. Handle requests for new map
19 * versions, and periodically resend as needed. Also implement
20 * statfs() and umount().
22 * A small cluster of Ceph "monitors" are responsible for managing critical
23 * cluster configuration and state information. An odd number (e.g., 3, 5)
24 * of cmon daemons use a modified version of the Paxos part-time parliament
25 * algorithm to manage the MDS map (mds cluster membership), OSD map, and
26 * list of clients who have mounted the file system.
28 * We maintain an open, active session with a monitor at all times in order to
29 * receive timely MDSMap updates. We periodically send a keepalive byte on the
30 * TCP socket to ensure we detect a failure. If the connection does break, we
31 * randomly hunt for a new monitor. Once the connection is reestablished, we
32 * resend any outstanding requests.
35 static const struct ceph_connection_operations mon_con_ops;
37 static int __validate_auth(struct ceph_mon_client *monc);
39 static int decode_mon_info(void **p, void *end, bool msgr2,
40 struct ceph_entity_addr *addr)
47 ret = ceph_start_decoding(p, end, 1, "mon_info_t", &struct_v,
52 mon_info_end = *p + struct_len;
53 ceph_decode_skip_string(p, end, e_inval); /* skip mon name */
54 ret = ceph_decode_entity_addrvec(p, end, msgr2, addr);
66 * Decode a monmap blob (e.g., during mount).
68 * Assume MonMap v3 (i.e. encoding with MONNAMES and MONENC).
70 static struct ceph_monmap *ceph_monmap_decode(void **p, void *end, bool msgr2)
72 struct ceph_monmap *monmap = NULL;
73 struct ceph_fsid fsid;
82 ceph_decode_32_safe(p, end, blob_len, e_inval);
83 ceph_decode_need(p, end, blob_len, e_inval);
85 ret = ceph_start_decoding(p, end, 6, "monmap", &struct_v, &struct_len);
89 dout("%s struct_v %d\n", __func__, struct_v);
90 ceph_decode_copy_safe(p, end, &fsid, sizeof(fsid), e_inval);
91 ceph_decode_32_safe(p, end, epoch, e_inval);
96 *p += sizeof(struct ceph_timespec); /* skip last_changed */
97 *p += sizeof(struct ceph_timespec); /* skip created */
99 ret = ceph_start_decoding(p, end, 1, "mon_feature_t",
100 &feat_struct_v, &feat_struct_len);
104 *p += feat_struct_len; /* skip persistent_features */
106 ret = ceph_start_decoding(p, end, 1, "mon_feature_t",
107 &feat_struct_v, &feat_struct_len);
111 *p += feat_struct_len; /* skip optional_features */
113 ceph_decode_32_safe(p, end, num_mon, e_inval);
115 dout("%s fsid %pU epoch %u num_mon %d\n", __func__, &fsid, epoch,
117 if (num_mon > CEPH_MAX_MON)
120 monmap = kmalloc(struct_size(monmap, mon_inst, num_mon), GFP_NOIO);
126 monmap->epoch = epoch;
127 monmap->num_mon = num_mon;
129 /* legacy_mon_addr map or mon_info map */
130 for (i = 0; i < num_mon; i++) {
131 struct ceph_entity_inst *inst = &monmap->mon_inst[i];
133 ceph_decode_skip_string(p, end, e_inval); /* skip mon name */
134 inst->name.type = CEPH_ENTITY_TYPE_MON;
135 inst->name.num = cpu_to_le64(i);
138 ret = decode_mon_info(p, end, msgr2, &inst->addr);
140 ret = ceph_decode_entity_addr(p, end, &inst->addr);
144 dout("%s mon%d addr %s\n", __func__, i,
145 ceph_pr_addr(&inst->addr));
158 * return true if *addr is included in the monmap.
160 int ceph_monmap_contains(struct ceph_monmap *m, struct ceph_entity_addr *addr)
164 for (i = 0; i < m->num_mon; i++) {
165 if (ceph_addr_equal_no_type(addr, &m->mon_inst[i].addr))
173 * Send an auth request.
175 static void __send_prepared_auth_request(struct ceph_mon_client *monc, int len)
177 monc->pending_auth = 1;
178 monc->m_auth->front.iov_len = len;
179 monc->m_auth->hdr.front_len = cpu_to_le32(len);
180 ceph_msg_revoke(monc->m_auth);
181 ceph_msg_get(monc->m_auth); /* keep our ref */
182 ceph_con_send(&monc->con, monc->m_auth);
186 * Close monitor session, if any.
188 static void __close_session(struct ceph_mon_client *monc)
190 dout("__close_session closing mon%d\n", monc->cur_mon);
191 ceph_msg_revoke(monc->m_auth);
192 ceph_msg_revoke_incoming(monc->m_auth_reply);
193 ceph_msg_revoke(monc->m_subscribe);
194 ceph_msg_revoke_incoming(monc->m_subscribe_ack);
195 ceph_con_close(&monc->con);
197 monc->pending_auth = 0;
198 ceph_auth_reset(monc->auth);
202 * Pick a new monitor at random and set cur_mon. If we are repicking
203 * (i.e. cur_mon is already set), be sure to pick a different one.
205 static void pick_new_mon(struct ceph_mon_client *monc)
207 int old_mon = monc->cur_mon;
209 BUG_ON(monc->monmap->num_mon < 1);
211 if (monc->monmap->num_mon == 1) {
214 int max = monc->monmap->num_mon;
218 if (monc->cur_mon >= 0) {
219 if (monc->cur_mon < monc->monmap->num_mon)
225 n = get_random_u32_below(max);
226 if (o >= 0 && n >= o)
232 dout("%s mon%d -> mon%d out of %d mons\n", __func__, old_mon,
233 monc->cur_mon, monc->monmap->num_mon);
237 * Open a session with a new monitor.
239 static void __open_session(struct ceph_mon_client *monc)
245 monc->hunting = true;
246 if (monc->had_a_connection) {
247 monc->hunt_mult *= CEPH_MONC_HUNT_BACKOFF;
248 if (monc->hunt_mult > CEPH_MONC_HUNT_MAX_MULT)
249 monc->hunt_mult = CEPH_MONC_HUNT_MAX_MULT;
252 monc->sub_renew_after = jiffies; /* i.e., expired */
253 monc->sub_renew_sent = 0;
255 dout("%s opening mon%d\n", __func__, monc->cur_mon);
256 ceph_con_open(&monc->con, CEPH_ENTITY_TYPE_MON, monc->cur_mon,
257 &monc->monmap->mon_inst[monc->cur_mon].addr);
260 * Queue a keepalive to ensure that in case of an early fault
261 * the messenger doesn't put us into STANDBY state and instead
262 * retries. This also ensures that our timestamp is valid by
263 * the time we finish hunting and delayed_work() checks it.
265 ceph_con_keepalive(&monc->con);
266 if (ceph_msgr2(monc->client)) {
267 monc->pending_auth = 1;
271 /* initiate authentication handshake */
272 ret = ceph_auth_build_hello(monc->auth,
273 monc->m_auth->front.iov_base,
274 monc->m_auth->front_alloc_len);
276 __send_prepared_auth_request(monc, ret);
279 static void reopen_session(struct ceph_mon_client *monc)
282 pr_info("mon%d %s session lost, hunting for new mon\n",
283 monc->cur_mon, ceph_pr_addr(&monc->con.peer_addr));
285 __close_session(monc);
286 __open_session(monc);
289 void ceph_monc_reopen_session(struct ceph_mon_client *monc)
291 mutex_lock(&monc->mutex);
292 reopen_session(monc);
293 mutex_unlock(&monc->mutex);
296 static void un_backoff(struct ceph_mon_client *monc)
298 monc->hunt_mult /= 2; /* reduce by 50% */
299 if (monc->hunt_mult < 1)
301 dout("%s hunt_mult now %d\n", __func__, monc->hunt_mult);
305 * Reschedule delayed work timer.
307 static void __schedule_delayed(struct ceph_mon_client *monc)
312 delay = CEPH_MONC_HUNT_INTERVAL * monc->hunt_mult;
314 delay = CEPH_MONC_PING_INTERVAL;
316 dout("__schedule_delayed after %lu\n", delay);
317 mod_delayed_work(system_wq, &monc->delayed_work,
318 round_jiffies_relative(delay));
321 const char *ceph_sub_str[] = {
322 [CEPH_SUB_MONMAP] = "monmap",
323 [CEPH_SUB_OSDMAP] = "osdmap",
324 [CEPH_SUB_FSMAP] = "fsmap.user",
325 [CEPH_SUB_MDSMAP] = "mdsmap",
329 * Send subscribe request for one or more maps, according to
332 static void __send_subscribe(struct ceph_mon_client *monc)
334 struct ceph_msg *msg = monc->m_subscribe;
335 void *p = msg->front.iov_base;
336 void *const end = p + msg->front_alloc_len;
340 dout("%s sent %lu\n", __func__, monc->sub_renew_sent);
342 BUG_ON(monc->cur_mon < 0);
344 if (!monc->sub_renew_sent)
345 monc->sub_renew_sent = jiffies | 1; /* never 0 */
347 msg->hdr.version = cpu_to_le16(2);
349 for (i = 0; i < ARRAY_SIZE(monc->subs); i++) {
350 if (monc->subs[i].want)
353 BUG_ON(num < 1); /* monmap sub is always there */
354 ceph_encode_32(&p, num);
355 for (i = 0; i < ARRAY_SIZE(monc->subs); i++) {
359 if (!monc->subs[i].want)
362 len = sprintf(buf, "%s", ceph_sub_str[i]);
363 if (i == CEPH_SUB_MDSMAP &&
364 monc->fs_cluster_id != CEPH_FS_CLUSTER_ID_NONE)
365 len += sprintf(buf + len, ".%d", monc->fs_cluster_id);
367 dout("%s %s start %llu flags 0x%x\n", __func__, buf,
368 le64_to_cpu(monc->subs[i].item.start),
369 monc->subs[i].item.flags);
370 ceph_encode_string(&p, end, buf, len);
371 memcpy(p, &monc->subs[i].item, sizeof(monc->subs[i].item));
372 p += sizeof(monc->subs[i].item);
376 msg->front.iov_len = p - msg->front.iov_base;
377 msg->hdr.front_len = cpu_to_le32(msg->front.iov_len);
378 ceph_msg_revoke(msg);
379 ceph_con_send(&monc->con, ceph_msg_get(msg));
382 static void handle_subscribe_ack(struct ceph_mon_client *monc,
383 struct ceph_msg *msg)
385 unsigned int seconds;
386 struct ceph_mon_subscribe_ack *h = msg->front.iov_base;
388 if (msg->front.iov_len < sizeof(*h))
390 seconds = le32_to_cpu(h->duration);
392 mutex_lock(&monc->mutex);
393 if (monc->sub_renew_sent) {
395 * This is only needed for legacy (infernalis or older)
396 * MONs -- see delayed_work().
398 monc->sub_renew_after = monc->sub_renew_sent +
399 (seconds >> 1) * HZ - 1;
400 dout("%s sent %lu duration %d renew after %lu\n", __func__,
401 monc->sub_renew_sent, seconds, monc->sub_renew_after);
402 monc->sub_renew_sent = 0;
404 dout("%s sent %lu renew after %lu, ignoring\n", __func__,
405 monc->sub_renew_sent, monc->sub_renew_after);
407 mutex_unlock(&monc->mutex);
410 pr_err("got corrupt subscribe-ack msg\n");
415 * Register interest in a map
417 * @sub: one of CEPH_SUB_*
418 * @epoch: X for "every map since X", or 0 for "just the latest"
420 static bool __ceph_monc_want_map(struct ceph_mon_client *monc, int sub,
421 u32 epoch, bool continuous)
423 __le64 start = cpu_to_le64(epoch);
424 u8 flags = !continuous ? CEPH_SUBSCRIBE_ONETIME : 0;
426 dout("%s %s epoch %u continuous %d\n", __func__, ceph_sub_str[sub],
429 if (monc->subs[sub].want &&
430 monc->subs[sub].item.start == start &&
431 monc->subs[sub].item.flags == flags)
434 monc->subs[sub].item.start = start;
435 monc->subs[sub].item.flags = flags;
436 monc->subs[sub].want = true;
441 bool ceph_monc_want_map(struct ceph_mon_client *monc, int sub, u32 epoch,
446 mutex_lock(&monc->mutex);
447 need_request = __ceph_monc_want_map(monc, sub, epoch, continuous);
448 mutex_unlock(&monc->mutex);
452 EXPORT_SYMBOL(ceph_monc_want_map);
455 * Keep track of which maps we have
457 * @sub: one of CEPH_SUB_*
459 static void __ceph_monc_got_map(struct ceph_mon_client *monc, int sub,
462 dout("%s %s epoch %u\n", __func__, ceph_sub_str[sub], epoch);
464 if (monc->subs[sub].want) {
465 if (monc->subs[sub].item.flags & CEPH_SUBSCRIBE_ONETIME)
466 monc->subs[sub].want = false;
468 monc->subs[sub].item.start = cpu_to_le64(epoch + 1);
471 monc->subs[sub].have = epoch;
474 void ceph_monc_got_map(struct ceph_mon_client *monc, int sub, u32 epoch)
476 mutex_lock(&monc->mutex);
477 __ceph_monc_got_map(monc, sub, epoch);
478 mutex_unlock(&monc->mutex);
480 EXPORT_SYMBOL(ceph_monc_got_map);
482 void ceph_monc_renew_subs(struct ceph_mon_client *monc)
484 mutex_lock(&monc->mutex);
485 __send_subscribe(monc);
486 mutex_unlock(&monc->mutex);
488 EXPORT_SYMBOL(ceph_monc_renew_subs);
491 * Wait for an osdmap with a given epoch.
493 * @epoch: epoch to wait for
494 * @timeout: in jiffies, 0 means "wait forever"
496 int ceph_monc_wait_osdmap(struct ceph_mon_client *monc, u32 epoch,
497 unsigned long timeout)
499 unsigned long started = jiffies;
502 mutex_lock(&monc->mutex);
503 while (monc->subs[CEPH_SUB_OSDMAP].have < epoch) {
504 mutex_unlock(&monc->mutex);
506 if (timeout && time_after_eq(jiffies, started + timeout))
509 ret = wait_event_interruptible_timeout(monc->client->auth_wq,
510 monc->subs[CEPH_SUB_OSDMAP].have >= epoch,
511 ceph_timeout_jiffies(timeout));
515 mutex_lock(&monc->mutex);
518 mutex_unlock(&monc->mutex);
521 EXPORT_SYMBOL(ceph_monc_wait_osdmap);
524 * Open a session with a random monitor. Request monmap and osdmap,
525 * which are waited upon in __ceph_open_session().
527 int ceph_monc_open_session(struct ceph_mon_client *monc)
529 mutex_lock(&monc->mutex);
530 __ceph_monc_want_map(monc, CEPH_SUB_MONMAP, 0, true);
531 __ceph_monc_want_map(monc, CEPH_SUB_OSDMAP, 0, false);
532 __open_session(monc);
533 __schedule_delayed(monc);
534 mutex_unlock(&monc->mutex);
537 EXPORT_SYMBOL(ceph_monc_open_session);
539 static void ceph_monc_handle_map(struct ceph_mon_client *monc,
540 struct ceph_msg *msg)
542 struct ceph_client *client = monc->client;
543 struct ceph_monmap *monmap;
546 mutex_lock(&monc->mutex);
548 dout("handle_monmap\n");
549 p = msg->front.iov_base;
550 end = p + msg->front.iov_len;
552 monmap = ceph_monmap_decode(&p, end, ceph_msgr2(client));
553 if (IS_ERR(monmap)) {
554 pr_err("problem decoding monmap, %d\n",
555 (int)PTR_ERR(monmap));
560 if (ceph_check_fsid(client, &monmap->fsid) < 0) {
566 monc->monmap = monmap;
568 __ceph_monc_got_map(monc, CEPH_SUB_MONMAP, monc->monmap->epoch);
569 client->have_fsid = true;
572 mutex_unlock(&monc->mutex);
573 wake_up_all(&client->auth_wq);
577 * generic requests (currently statfs, mon_get_version)
579 DEFINE_RB_FUNCS(generic_request, struct ceph_mon_generic_request, tid, node)
581 static void release_generic_request(struct kref *kref)
583 struct ceph_mon_generic_request *req =
584 container_of(kref, struct ceph_mon_generic_request, kref);
586 dout("%s greq %p request %p reply %p\n", __func__, req, req->request,
588 WARN_ON(!RB_EMPTY_NODE(&req->node));
591 ceph_msg_put(req->reply);
593 ceph_msg_put(req->request);
598 static void put_generic_request(struct ceph_mon_generic_request *req)
601 kref_put(&req->kref, release_generic_request);
604 static void get_generic_request(struct ceph_mon_generic_request *req)
606 kref_get(&req->kref);
609 static struct ceph_mon_generic_request *
610 alloc_generic_request(struct ceph_mon_client *monc, gfp_t gfp)
612 struct ceph_mon_generic_request *req;
614 req = kzalloc(sizeof(*req), gfp);
619 kref_init(&req->kref);
620 RB_CLEAR_NODE(&req->node);
621 init_completion(&req->completion);
623 dout("%s greq %p\n", __func__, req);
627 static void register_generic_request(struct ceph_mon_generic_request *req)
629 struct ceph_mon_client *monc = req->monc;
633 get_generic_request(req);
634 req->tid = ++monc->last_tid;
635 insert_generic_request(&monc->generic_request_tree, req);
638 static void send_generic_request(struct ceph_mon_client *monc,
639 struct ceph_mon_generic_request *req)
643 dout("%s greq %p tid %llu\n", __func__, req, req->tid);
644 req->request->hdr.tid = cpu_to_le64(req->tid);
645 ceph_con_send(&monc->con, ceph_msg_get(req->request));
648 static void __finish_generic_request(struct ceph_mon_generic_request *req)
650 struct ceph_mon_client *monc = req->monc;
652 dout("%s greq %p tid %llu\n", __func__, req, req->tid);
653 erase_generic_request(&monc->generic_request_tree, req);
655 ceph_msg_revoke(req->request);
656 ceph_msg_revoke_incoming(req->reply);
659 static void finish_generic_request(struct ceph_mon_generic_request *req)
661 __finish_generic_request(req);
662 put_generic_request(req);
665 static void complete_generic_request(struct ceph_mon_generic_request *req)
667 if (req->complete_cb)
668 req->complete_cb(req);
670 complete_all(&req->completion);
671 put_generic_request(req);
674 static void cancel_generic_request(struct ceph_mon_generic_request *req)
676 struct ceph_mon_client *monc = req->monc;
677 struct ceph_mon_generic_request *lookup_req;
679 dout("%s greq %p tid %llu\n", __func__, req, req->tid);
681 mutex_lock(&monc->mutex);
682 lookup_req = lookup_generic_request(&monc->generic_request_tree,
685 WARN_ON(lookup_req != req);
686 finish_generic_request(req);
689 mutex_unlock(&monc->mutex);
692 static int wait_generic_request(struct ceph_mon_generic_request *req)
696 dout("%s greq %p tid %llu\n", __func__, req, req->tid);
697 ret = wait_for_completion_interruptible(&req->completion);
699 cancel_generic_request(req);
701 ret = req->result; /* completed */
706 static struct ceph_msg *get_generic_reply(struct ceph_connection *con,
707 struct ceph_msg_header *hdr,
710 struct ceph_mon_client *monc = con->private;
711 struct ceph_mon_generic_request *req;
712 u64 tid = le64_to_cpu(hdr->tid);
715 mutex_lock(&monc->mutex);
716 req = lookup_generic_request(&monc->generic_request_tree, tid);
718 dout("get_generic_reply %lld dne\n", tid);
722 dout("get_generic_reply %lld got %p\n", tid, req->reply);
724 m = ceph_msg_get(req->reply);
726 * we don't need to track the connection reading into
727 * this reply because we only have one open connection
731 mutex_unlock(&monc->mutex);
738 static void handle_statfs_reply(struct ceph_mon_client *monc,
739 struct ceph_msg *msg)
741 struct ceph_mon_generic_request *req;
742 struct ceph_mon_statfs_reply *reply = msg->front.iov_base;
743 u64 tid = le64_to_cpu(msg->hdr.tid);
745 dout("%s msg %p tid %llu\n", __func__, msg, tid);
747 if (msg->front.iov_len != sizeof(*reply))
750 mutex_lock(&monc->mutex);
751 req = lookup_generic_request(&monc->generic_request_tree, tid);
753 mutex_unlock(&monc->mutex);
758 *req->u.st = reply->st; /* struct */
759 __finish_generic_request(req);
760 mutex_unlock(&monc->mutex);
762 complete_generic_request(req);
766 pr_err("corrupt statfs reply, tid %llu\n", tid);
771 * Do a synchronous statfs().
773 int ceph_monc_do_statfs(struct ceph_mon_client *monc, u64 data_pool,
774 struct ceph_statfs *buf)
776 struct ceph_mon_generic_request *req;
777 struct ceph_mon_statfs *h;
780 req = alloc_generic_request(monc, GFP_NOFS);
784 req->request = ceph_msg_new(CEPH_MSG_STATFS, sizeof(*h), GFP_NOFS,
789 req->reply = ceph_msg_new(CEPH_MSG_STATFS_REPLY, 64, GFP_NOFS, true);
794 req->request->hdr.version = cpu_to_le16(2);
796 mutex_lock(&monc->mutex);
797 register_generic_request(req);
798 /* fill out request */
799 h = req->request->front.iov_base;
800 h->monhdr.have_version = 0;
801 h->monhdr.session_mon = cpu_to_le16(-1);
802 h->monhdr.session_mon_tid = 0;
803 h->fsid = monc->monmap->fsid;
804 h->contains_data_pool = (data_pool != CEPH_NOPOOL);
805 h->data_pool = cpu_to_le64(data_pool);
806 send_generic_request(monc, req);
807 mutex_unlock(&monc->mutex);
809 ret = wait_generic_request(req);
811 put_generic_request(req);
814 EXPORT_SYMBOL(ceph_monc_do_statfs);
816 static void handle_get_version_reply(struct ceph_mon_client *monc,
817 struct ceph_msg *msg)
819 struct ceph_mon_generic_request *req;
820 u64 tid = le64_to_cpu(msg->hdr.tid);
821 void *p = msg->front.iov_base;
822 void *end = p + msg->front_alloc_len;
825 dout("%s msg %p tid %llu\n", __func__, msg, tid);
827 ceph_decode_need(&p, end, 2*sizeof(u64), bad);
828 handle = ceph_decode_64(&p);
829 if (tid != 0 && tid != handle)
832 mutex_lock(&monc->mutex);
833 req = lookup_generic_request(&monc->generic_request_tree, handle);
835 mutex_unlock(&monc->mutex);
840 req->u.newest = ceph_decode_64(&p);
841 __finish_generic_request(req);
842 mutex_unlock(&monc->mutex);
844 complete_generic_request(req);
848 pr_err("corrupt mon_get_version reply, tid %llu\n", tid);
852 static struct ceph_mon_generic_request *
853 __ceph_monc_get_version(struct ceph_mon_client *monc, const char *what,
854 ceph_monc_callback_t cb, u64 private_data)
856 struct ceph_mon_generic_request *req;
858 req = alloc_generic_request(monc, GFP_NOIO);
862 req->request = ceph_msg_new(CEPH_MSG_MON_GET_VERSION,
863 sizeof(u64) + sizeof(u32) + strlen(what),
868 req->reply = ceph_msg_new(CEPH_MSG_MON_GET_VERSION_REPLY, 32, GFP_NOIO,
873 req->complete_cb = cb;
874 req->private_data = private_data;
876 mutex_lock(&monc->mutex);
877 register_generic_request(req);
879 void *p = req->request->front.iov_base;
880 void *const end = p + req->request->front_alloc_len;
882 ceph_encode_64(&p, req->tid); /* handle */
883 ceph_encode_string(&p, end, what, strlen(what));
886 send_generic_request(monc, req);
887 mutex_unlock(&monc->mutex);
892 put_generic_request(req);
893 return ERR_PTR(-ENOMEM);
897 * Send MMonGetVersion and wait for the reply.
899 * @what: one of "mdsmap", "osdmap" or "monmap"
901 int ceph_monc_get_version(struct ceph_mon_client *monc, const char *what,
904 struct ceph_mon_generic_request *req;
907 req = __ceph_monc_get_version(monc, what, NULL, 0);
911 ret = wait_generic_request(req);
913 *newest = req->u.newest;
915 put_generic_request(req);
918 EXPORT_SYMBOL(ceph_monc_get_version);
921 * Send MMonGetVersion,
923 * @what: one of "mdsmap", "osdmap" or "monmap"
925 int ceph_monc_get_version_async(struct ceph_mon_client *monc, const char *what,
926 ceph_monc_callback_t cb, u64 private_data)
928 struct ceph_mon_generic_request *req;
930 req = __ceph_monc_get_version(monc, what, cb, private_data);
934 put_generic_request(req);
937 EXPORT_SYMBOL(ceph_monc_get_version_async);
939 static void handle_command_ack(struct ceph_mon_client *monc,
940 struct ceph_msg *msg)
942 struct ceph_mon_generic_request *req;
943 void *p = msg->front.iov_base;
944 void *const end = p + msg->front_alloc_len;
945 u64 tid = le64_to_cpu(msg->hdr.tid);
947 dout("%s msg %p tid %llu\n", __func__, msg, tid);
949 ceph_decode_need(&p, end, sizeof(struct ceph_mon_request_header) +
951 p += sizeof(struct ceph_mon_request_header);
953 mutex_lock(&monc->mutex);
954 req = lookup_generic_request(&monc->generic_request_tree, tid);
956 mutex_unlock(&monc->mutex);
960 req->result = ceph_decode_32(&p);
961 __finish_generic_request(req);
962 mutex_unlock(&monc->mutex);
964 complete_generic_request(req);
968 pr_err("corrupt mon_command ack, tid %llu\n", tid);
972 static __printf(2, 0)
973 int do_mon_command_vargs(struct ceph_mon_client *monc, const char *fmt,
976 struct ceph_mon_generic_request *req;
977 struct ceph_mon_command *h;
981 req = alloc_generic_request(monc, GFP_NOIO);
985 req->request = ceph_msg_new(CEPH_MSG_MON_COMMAND, 256, GFP_NOIO, true);
989 req->reply = ceph_msg_new(CEPH_MSG_MON_COMMAND_ACK, 512, GFP_NOIO,
994 mutex_lock(&monc->mutex);
995 register_generic_request(req);
996 h = req->request->front.iov_base;
997 h->monhdr.have_version = 0;
998 h->monhdr.session_mon = cpu_to_le16(-1);
999 h->monhdr.session_mon_tid = 0;
1000 h->fsid = monc->monmap->fsid;
1001 h->num_strs = cpu_to_le32(1);
1002 len = vsprintf(h->str, fmt, ap);
1003 h->str_len = cpu_to_le32(len);
1004 send_generic_request(monc, req);
1005 mutex_unlock(&monc->mutex);
1007 ret = wait_generic_request(req);
1009 put_generic_request(req);
1013 static __printf(2, 3)
1014 int do_mon_command(struct ceph_mon_client *monc, const char *fmt, ...)
1020 ret = do_mon_command_vargs(monc, fmt, ap);
1025 int ceph_monc_blocklist_add(struct ceph_mon_client *monc,
1026 struct ceph_entity_addr *client_addr)
1030 ret = do_mon_command(monc,
1031 "{ \"prefix\": \"osd blocklist\", \
1032 \"blocklistop\": \"add\", \
1033 \"addr\": \"%pISpc/%u\" }",
1034 &client_addr->in_addr,
1035 le32_to_cpu(client_addr->nonce));
1036 if (ret == -EINVAL) {
1038 * The monitor returns EINVAL on an unrecognized command.
1039 * Try the legacy command -- it is exactly the same except
1042 ret = do_mon_command(monc,
1043 "{ \"prefix\": \"osd blacklist\", \
1044 \"blacklistop\": \"add\", \
1045 \"addr\": \"%pISpc/%u\" }",
1046 &client_addr->in_addr,
1047 le32_to_cpu(client_addr->nonce));
1053 * Make sure we have the osdmap that includes the blocklist
1054 * entry. This is needed to ensure that the OSDs pick up the
1055 * new blocklist before processing any future requests from
1058 return ceph_wait_for_latest_osdmap(monc->client, 0);
1060 EXPORT_SYMBOL(ceph_monc_blocklist_add);
1063 * Resend pending generic requests.
1065 static void __resend_generic_request(struct ceph_mon_client *monc)
1067 struct ceph_mon_generic_request *req;
1070 for (p = rb_first(&monc->generic_request_tree); p; p = rb_next(p)) {
1071 req = rb_entry(p, struct ceph_mon_generic_request, node);
1072 ceph_msg_revoke(req->request);
1073 ceph_msg_revoke_incoming(req->reply);
1074 ceph_con_send(&monc->con, ceph_msg_get(req->request));
1079 * Delayed work. If we haven't mounted yet, retry. Otherwise,
1080 * renew/retry subscription as needed (in case it is timing out, or we
1081 * got an ENOMEM). And keep the monitor connection alive.
1083 static void delayed_work(struct work_struct *work)
1085 struct ceph_mon_client *monc =
1086 container_of(work, struct ceph_mon_client, delayed_work.work);
1088 dout("monc delayed_work\n");
1089 mutex_lock(&monc->mutex);
1090 if (monc->hunting) {
1091 dout("%s continuing hunt\n", __func__);
1092 reopen_session(monc);
1094 int is_auth = ceph_auth_is_authenticated(monc->auth);
1095 if (ceph_con_keepalive_expired(&monc->con,
1096 CEPH_MONC_PING_TIMEOUT)) {
1097 dout("monc keepalive timeout\n");
1099 reopen_session(monc);
1102 if (!monc->hunting) {
1103 ceph_con_keepalive(&monc->con);
1104 __validate_auth(monc);
1109 !(monc->con.peer_features & CEPH_FEATURE_MON_STATEFUL_SUB)) {
1110 unsigned long now = jiffies;
1112 dout("%s renew subs? now %lu renew after %lu\n",
1113 __func__, now, monc->sub_renew_after);
1114 if (time_after_eq(now, monc->sub_renew_after))
1115 __send_subscribe(monc);
1118 __schedule_delayed(monc);
1119 mutex_unlock(&monc->mutex);
1123 * On startup, we build a temporary monmap populated with the IPs
1124 * provided by mount(2).
1126 static int build_initial_monmap(struct ceph_mon_client *monc)
1128 __le32 my_type = ceph_msgr2(monc->client) ?
1129 CEPH_ENTITY_ADDR_TYPE_MSGR2 : CEPH_ENTITY_ADDR_TYPE_LEGACY;
1130 struct ceph_options *opt = monc->client->options;
1131 int num_mon = opt->num_mon;
1134 /* build initial monmap */
1135 monc->monmap = kzalloc(struct_size(monc->monmap, mon_inst, num_mon),
1140 for (i = 0; i < num_mon; i++) {
1141 struct ceph_entity_inst *inst = &monc->monmap->mon_inst[i];
1143 memcpy(&inst->addr.in_addr, &opt->mon_addr[i].in_addr,
1144 sizeof(inst->addr.in_addr));
1145 inst->addr.type = my_type;
1146 inst->addr.nonce = 0;
1147 inst->name.type = CEPH_ENTITY_TYPE_MON;
1148 inst->name.num = cpu_to_le64(i);
1150 monc->monmap->num_mon = num_mon;
1154 int ceph_monc_init(struct ceph_mon_client *monc, struct ceph_client *cl)
1159 memset(monc, 0, sizeof(*monc));
1161 mutex_init(&monc->mutex);
1163 err = build_initial_monmap(monc);
1168 /* authentication */
1169 monc->auth = ceph_auth_init(cl->options->name, cl->options->key,
1170 cl->options->con_modes);
1171 if (IS_ERR(monc->auth)) {
1172 err = PTR_ERR(monc->auth);
1175 monc->auth->want_keys =
1176 CEPH_ENTITY_TYPE_AUTH | CEPH_ENTITY_TYPE_MON |
1177 CEPH_ENTITY_TYPE_OSD | CEPH_ENTITY_TYPE_MDS;
1181 monc->m_subscribe_ack = ceph_msg_new(CEPH_MSG_MON_SUBSCRIBE_ACK,
1182 sizeof(struct ceph_mon_subscribe_ack),
1184 if (!monc->m_subscribe_ack)
1187 monc->m_subscribe = ceph_msg_new(CEPH_MSG_MON_SUBSCRIBE, 128,
1189 if (!monc->m_subscribe)
1190 goto out_subscribe_ack;
1192 monc->m_auth_reply = ceph_msg_new(CEPH_MSG_AUTH_REPLY, 4096,
1194 if (!monc->m_auth_reply)
1197 monc->m_auth = ceph_msg_new(CEPH_MSG_AUTH, 4096, GFP_KERNEL, true);
1198 monc->pending_auth = 0;
1200 goto out_auth_reply;
1202 ceph_con_init(&monc->con, monc, &mon_con_ops,
1203 &monc->client->msgr);
1206 monc->had_a_connection = false;
1207 monc->hunt_mult = 1;
1209 INIT_DELAYED_WORK(&monc->delayed_work, delayed_work);
1210 monc->generic_request_tree = RB_ROOT;
1213 monc->fs_cluster_id = CEPH_FS_CLUSTER_ID_NONE;
1218 ceph_msg_put(monc->m_auth_reply);
1220 ceph_msg_put(monc->m_subscribe);
1222 ceph_msg_put(monc->m_subscribe_ack);
1224 ceph_auth_destroy(monc->auth);
1226 kfree(monc->monmap);
1230 EXPORT_SYMBOL(ceph_monc_init);
1232 void ceph_monc_stop(struct ceph_mon_client *monc)
1235 cancel_delayed_work_sync(&monc->delayed_work);
1237 mutex_lock(&monc->mutex);
1238 __close_session(monc);
1240 mutex_unlock(&monc->mutex);
1243 * flush msgr queue before we destroy ourselves to ensure that:
1244 * - any work that references our embedded con is finished.
1245 * - any osd_client or other work that may reference an authorizer
1246 * finishes before we shut down the auth subsystem.
1250 ceph_auth_destroy(monc->auth);
1252 WARN_ON(!RB_EMPTY_ROOT(&monc->generic_request_tree));
1254 ceph_msg_put(monc->m_auth);
1255 ceph_msg_put(monc->m_auth_reply);
1256 ceph_msg_put(monc->m_subscribe);
1257 ceph_msg_put(monc->m_subscribe_ack);
1259 kfree(monc->monmap);
1261 EXPORT_SYMBOL(ceph_monc_stop);
1263 static void finish_hunting(struct ceph_mon_client *monc)
1265 if (monc->hunting) {
1266 dout("%s found mon%d\n", __func__, monc->cur_mon);
1267 monc->hunting = false;
1268 monc->had_a_connection = true;
1270 __schedule_delayed(monc);
1274 static void finish_auth(struct ceph_mon_client *monc, int auth_err,
1277 dout("%s auth_err %d was_authed %d\n", __func__, auth_err, was_authed);
1278 WARN_ON(auth_err > 0);
1280 monc->pending_auth = 0;
1282 monc->client->auth_err = auth_err;
1283 wake_up_all(&monc->client->auth_wq);
1287 if (!was_authed && ceph_auth_is_authenticated(monc->auth)) {
1288 dout("%s authenticated, starting session global_id %llu\n",
1289 __func__, monc->auth->global_id);
1291 monc->client->msgr.inst.name.type = CEPH_ENTITY_TYPE_CLIENT;
1292 monc->client->msgr.inst.name.num =
1293 cpu_to_le64(monc->auth->global_id);
1295 __send_subscribe(monc);
1296 __resend_generic_request(monc);
1298 pr_info("mon%d %s session established\n", monc->cur_mon,
1299 ceph_pr_addr(&monc->con.peer_addr));
1303 static void handle_auth_reply(struct ceph_mon_client *monc,
1304 struct ceph_msg *msg)
1309 mutex_lock(&monc->mutex);
1310 was_authed = ceph_auth_is_authenticated(monc->auth);
1311 ret = ceph_handle_auth_reply(monc->auth, msg->front.iov_base,
1313 monc->m_auth->front.iov_base,
1314 monc->m_auth->front_alloc_len);
1316 __send_prepared_auth_request(monc, ret);
1318 finish_auth(monc, ret, was_authed);
1319 finish_hunting(monc);
1321 mutex_unlock(&monc->mutex);
1324 static int __validate_auth(struct ceph_mon_client *monc)
1328 if (monc->pending_auth)
1331 ret = ceph_build_auth(monc->auth, monc->m_auth->front.iov_base,
1332 monc->m_auth->front_alloc_len);
1334 return ret; /* either an error, or no need to authenticate */
1335 __send_prepared_auth_request(monc, ret);
1339 int ceph_monc_validate_auth(struct ceph_mon_client *monc)
1343 mutex_lock(&monc->mutex);
1344 ret = __validate_auth(monc);
1345 mutex_unlock(&monc->mutex);
1348 EXPORT_SYMBOL(ceph_monc_validate_auth);
1350 static int mon_get_auth_request(struct ceph_connection *con,
1351 void *buf, int *buf_len,
1352 void **authorizer, int *authorizer_len)
1354 struct ceph_mon_client *monc = con->private;
1357 mutex_lock(&monc->mutex);
1358 ret = ceph_auth_get_request(monc->auth, buf, *buf_len);
1359 mutex_unlock(&monc->mutex);
1365 *authorizer_len = 0;
1369 static int mon_handle_auth_reply_more(struct ceph_connection *con,
1370 void *reply, int reply_len,
1371 void *buf, int *buf_len,
1372 void **authorizer, int *authorizer_len)
1374 struct ceph_mon_client *monc = con->private;
1377 mutex_lock(&monc->mutex);
1378 ret = ceph_auth_handle_reply_more(monc->auth, reply, reply_len,
1380 mutex_unlock(&monc->mutex);
1386 *authorizer_len = 0;
1390 static int mon_handle_auth_done(struct ceph_connection *con,
1391 u64 global_id, void *reply, int reply_len,
1392 u8 *session_key, int *session_key_len,
1393 u8 *con_secret, int *con_secret_len)
1395 struct ceph_mon_client *monc = con->private;
1399 mutex_lock(&monc->mutex);
1400 WARN_ON(!monc->hunting);
1401 was_authed = ceph_auth_is_authenticated(monc->auth);
1402 ret = ceph_auth_handle_reply_done(monc->auth, global_id,
1404 session_key, session_key_len,
1405 con_secret, con_secret_len);
1406 finish_auth(monc, ret, was_authed);
1408 finish_hunting(monc);
1409 mutex_unlock(&monc->mutex);
1413 static int mon_handle_auth_bad_method(struct ceph_connection *con,
1414 int used_proto, int result,
1415 const int *allowed_protos, int proto_cnt,
1416 const int *allowed_modes, int mode_cnt)
1418 struct ceph_mon_client *monc = con->private;
1421 mutex_lock(&monc->mutex);
1422 WARN_ON(!monc->hunting);
1423 was_authed = ceph_auth_is_authenticated(monc->auth);
1424 ceph_auth_handle_bad_method(monc->auth, used_proto, result,
1425 allowed_protos, proto_cnt,
1426 allowed_modes, mode_cnt);
1427 finish_auth(monc, -EACCES, was_authed);
1428 mutex_unlock(&monc->mutex);
1433 * handle incoming message
1435 static void mon_dispatch(struct ceph_connection *con, struct ceph_msg *msg)
1437 struct ceph_mon_client *monc = con->private;
1438 int type = le16_to_cpu(msg->hdr.type);
1441 case CEPH_MSG_AUTH_REPLY:
1442 handle_auth_reply(monc, msg);
1445 case CEPH_MSG_MON_SUBSCRIBE_ACK:
1446 handle_subscribe_ack(monc, msg);
1449 case CEPH_MSG_STATFS_REPLY:
1450 handle_statfs_reply(monc, msg);
1453 case CEPH_MSG_MON_GET_VERSION_REPLY:
1454 handle_get_version_reply(monc, msg);
1457 case CEPH_MSG_MON_COMMAND_ACK:
1458 handle_command_ack(monc, msg);
1461 case CEPH_MSG_MON_MAP:
1462 ceph_monc_handle_map(monc, msg);
1465 case CEPH_MSG_OSD_MAP:
1466 ceph_osdc_handle_map(&monc->client->osdc, msg);
1470 /* can the chained handler handle it? */
1471 if (monc->client->extra_mon_dispatch &&
1472 monc->client->extra_mon_dispatch(monc->client, msg) == 0)
1475 pr_err("received unknown message type %d %s\n", type,
1476 ceph_msg_type_name(type));
1482 * Allocate memory for incoming message
1484 static struct ceph_msg *mon_alloc_msg(struct ceph_connection *con,
1485 struct ceph_msg_header *hdr,
1488 struct ceph_mon_client *monc = con->private;
1489 int type = le16_to_cpu(hdr->type);
1490 int front_len = le32_to_cpu(hdr->front_len);
1491 struct ceph_msg *m = NULL;
1496 case CEPH_MSG_MON_SUBSCRIBE_ACK:
1497 m = ceph_msg_get(monc->m_subscribe_ack);
1499 case CEPH_MSG_STATFS_REPLY:
1500 case CEPH_MSG_MON_COMMAND_ACK:
1501 return get_generic_reply(con, hdr, skip);
1502 case CEPH_MSG_AUTH_REPLY:
1503 m = ceph_msg_get(monc->m_auth_reply);
1505 case CEPH_MSG_MON_GET_VERSION_REPLY:
1506 if (le64_to_cpu(hdr->tid) != 0)
1507 return get_generic_reply(con, hdr, skip);
1510 * Older OSDs don't set reply tid even if the original
1511 * request had a non-zero tid. Work around this weirdness
1512 * by allocating a new message.
1515 case CEPH_MSG_MON_MAP:
1516 case CEPH_MSG_MDS_MAP:
1517 case CEPH_MSG_OSD_MAP:
1518 case CEPH_MSG_FS_MAP_USER:
1519 m = ceph_msg_new(type, front_len, GFP_NOFS, false);
1521 return NULL; /* ENOMEM--return skip == 0 */
1526 pr_info("alloc_msg unknown type %d\n", type);
1528 } else if (front_len > m->front_alloc_len) {
1529 pr_warn("mon_alloc_msg front %d > prealloc %d (%u#%llu)\n",
1530 front_len, m->front_alloc_len,
1531 (unsigned int)con->peer_name.type,
1532 le64_to_cpu(con->peer_name.num));
1534 m = ceph_msg_new(type, front_len, GFP_NOFS, false);
1541 * If the monitor connection resets, pick a new monitor and resubmit
1542 * any pending requests.
1544 static void mon_fault(struct ceph_connection *con)
1546 struct ceph_mon_client *monc = con->private;
1548 mutex_lock(&monc->mutex);
1549 dout("%s mon%d\n", __func__, monc->cur_mon);
1550 if (monc->cur_mon >= 0) {
1551 if (!monc->hunting) {
1552 dout("%s hunting for new mon\n", __func__);
1553 reopen_session(monc);
1554 __schedule_delayed(monc);
1556 dout("%s already hunting\n", __func__);
1559 mutex_unlock(&monc->mutex);
1563 * We can ignore refcounting on the connection struct, as all references
1564 * will come from the messenger workqueue, which is drained prior to
1565 * mon_client destruction.
1567 static struct ceph_connection *mon_get_con(struct ceph_connection *con)
1572 static void mon_put_con(struct ceph_connection *con)
1576 static const struct ceph_connection_operations mon_con_ops = {
1579 .alloc_msg = mon_alloc_msg,
1580 .dispatch = mon_dispatch,
1582 .get_auth_request = mon_get_auth_request,
1583 .handle_auth_reply_more = mon_handle_auth_reply_more,
1584 .handle_auth_done = mon_handle_auth_done,
1585 .handle_auth_bad_method = mon_handle_auth_bad_method,