From 4ee84c51fa73c4ec7cbee922111a140a3aaf75df Mon Sep 17 00:00:00 2001 From: Jason Date: Sat, 10 Jan 2015 17:35:47 +0800 Subject: [PATCH] totem: Ignore duplicated commit tokens in recovery In active rrp mode, commit tokens are treated as mcast data messages, thus, rrp directly delivers them to srp layer by active_mcast_recv(). This will result in duplicated commit tokens being received by srp from different heartbeat links. If node is in recovery state and has already sent out the initial orf token, those duplicated commit tokens will cause message_handler_memb_commit_token() to send initial orf token again! This is wrong because it resets the orf token content in instance->orf_token_retransmit, which breaks the token retransmission state. Furthermore, by sending those initial orf tokens again and again, it may lead active_token_recv() to drop some subsequent orf tokens. It is OK for rrp because srp will do token retransmission, but as said above, srp retransmission state has already been broken, so finally we meet a "token lost in recovery state" condition caused by software. If token timeout value is large, then it will takes long time to create a new ring. This can be reproduced by having two noded set to active rrp mode, with two heartbeat links. Then with one node always on, let the other one do stop/start again and again. It has a low probability to reproduce. In theory, I think, the more heartbeat links used, the more easily it can be reproduced. This problem can be resolved by letting message_handler_memb_commit_token() to ignore duplicated commit tokens in recovery state if node (the ring representation) has already sent out the initial orf token. Different from prev take, this version do not depends on stored token data but uses originated_orf_token in totemsrp_instance to remember if initial orf token has been already originated for current membership. Signed-off-by: Jason Reviewed-by: Steven Dake Reviewed-by: Jan Friesse Reviewed-by: Christine Caulfield --- exec/totemsrp.c | 16 ++++++++++++++++ 1 files changed, 16 insertions(+), 0 deletions(-) diff --git a/exec/totemsrp.c b/exec/totemsrp.c index 95736b1..b05773a 100644 --- a/exec/totemsrp.c +++ b/exec/totemsrp.c @@ -508,6 +508,8 @@ struct totemsrp_instance { uint32_t orf_token_discard; + uint32_t originated_orf_token; + uint32_t threaded_mode_enabled; uint32_t waiting_trans_ack; @@ -731,6 +733,8 @@ static void totemsrp_instance_initialize (struct totemsrp_instance *instance) instance->orf_token_discard = 0; + instance->originated_orf_token = 0; + instance->commit_token = (struct memb_commit_token *)instance->commit_token_storage; instance->my_id.no_addrs = INTERFACE_MAX; @@ -1834,6 +1838,8 @@ static void memb_state_operational_enter (struct totemsrp_instance *instance) char left_node_msg[1024]; char joined_node_msg[1024]; + instance->originated_orf_token = 0; + memb_consensus_reset (instance); old_ring_state_reset (instance); @@ -2045,6 +2051,8 @@ static void memb_state_gather_enter ( { instance->orf_token_discard = 1; + instance->originated_orf_token = 0; + memb_set_merge ( &instance->my_id, 1, instance->my_proc_list, &instance->my_proc_list_entries); @@ -4510,6 +4518,14 @@ static int message_handler_memb_commit_token ( case MEMB_STATE_RECOVERY: if (totemip_equal (&instance->my_id.addr[0], &instance->my_ring_id.rep)) { + + /* Filter out duplicated tokens */ + if (instance->originated_orf_token) { + break; + } + + instance->originated_orf_token = 1; + log_printf (instance->totemsrp_log_level_debug, "Sending initial ORF token"); -- 1.7.1