Node is not connecting.

MySQL & MariaDB Percona XtraDB Cluster 5.x
1

Hi ,
We are using three node Percona XtraDB Cluster. Right now only one node(Node No: 2) is running. All the (Node No: 1, Node No: 3) remaining nodes are down. here I attached the Down nodes error log.
Please provide your suggestion to solve.
Advance Thanks.

Node No: 3
Error Log:

131224 4:16:06 [ERROR] WSREP: failed to open gcomm backend connection: 110: failed to reach primary view: 110 (Connection timed out)
at gcomm/src/pc.cpp:connect():157
131224 4:16:06 [ERROR] WSREP: gcs/src/gcs_core.c:gcs_core_open():195: Failed to open backend connection: -110 (Connection timed out)
131224 4:16:06 [ERROR] WSREP: gcs/src/gcs.c:gcs_open():1290: Failed to open channel ‘ecomm’ at ‘gcomm://X.X.X.2’: -110 (Connection timed out)
131224 4:16:06 [ERROR] WSREP: gcs connect failed: Connection timed out
131224 4:16:06 [ERROR] WSREP: wsrep::connect() failed: 6
131224 4:16:06 [ERROR] Aborting

131224 4:16:06 [Note] WSREP: Service disconnected.
131224 4:16:07 [Note] WSREP: Some threads may fail to exit.
131224 4:16:07 [Note] /usr/sbin/mysqld: Shutdown complete

131224 04:16:07 mysqld_safe mysqld from pid file /mnt/data//FD-R14-db3.pid ended
131224 04:19:56 mysqld_safe Starting mysqld daemon with databases from /mnt/data/
131224 4:19:56 [Note] Flashcache bypass: disabled
131224 4:19:56 [Note] Flashcache setup error is : ioctl failed

131224 4:19:56 [Note] WSREP: Read nil XID from storage engines, skipping position init
131224 4:19:56 [Note] WSREP: wsrep_load(): loading provider library ‘/usr/lib64/libgalera_smm.so’
131224 4:19:56 [Note] WSREP: wsrep_load(): Galera 2.2(r114) by Codership Oy <info@codership.com> loaded succesfully.
131224 4:19:56 [Note] WSREP: Found saved state: 00000000-0000-0000-0000-000000000000:-1
131224 4:19:56 [Note] WSREP: Reusing existing ‘/mnt/data//galera.cache’.
131224 4:19:56 [Note] WSREP: Passing config to GCS: base_host = X.X.X.3; gcache.dir = /mnt/data/; gcache.keep_pages_size = 0; gcache.mem_size = 0; gcache.name = /mnt/data//galera.cache; gcache.page_size = 128M; gcache.size = 128M; gcs.fc_debug = 0; gcs.fc_factor = 0.5; gcs.fc_limit = 16; gcs.fc_master_slave = NO; gcs.max_packet_size = 64500; gcs.max_throttle = 0.25; gcs.recv_q_hard_limit = 9223372036854775807; gcs.recv_q_soft_limit = 0.25; gcs.sync_donor = NO; replicator.causal_read_timeout = PT30S; replicator.commit_order = 3
131224 4:19:56 [Note] WSREP: Assign initial position for certification: -1, protocol version: -1
131224 4:19:56 [Note] WSREP: wsrep_sst_grab()
131224 4:19:56 [Note] WSREP: Start replication
131224 4:19:56 [Note] WSREP: Setting initial position to 00000000-0000-0000-0000-000000000000:-1
131224 4:19:56 [Note] WSREP: protonet asio version 0
131224 4:19:56 [Note] WSREP: backend: asio
131224 4:19:56 [Note] WSREP: GMCast version 0
131224 4:19:56 [Note] WSREP: (a5560fc4-6c52-11e3-0800-6de37b8550ef, ‘tcp://0.0.0.0:4567’) listening at tcp://0.0.0.0:4567
131224 4:19:56 [Note] WSREP: (a5560fc4-6c52-11e3-0800-6de37b8550ef, ‘tcp://0.0.0.0:4567’) multicast: , ttl: 1
131224 4:19:56 [Note] WSREP: EVS version 0
131224 4:19:56 [Note] WSREP: PC version 0
131224 4:19:56 [Note] WSREP: gcomm: connecting to group ‘ecomm’, peer ‘X.X.X.2:’
131224 4:19:57 [Note] WSREP: declaring 771b0086-6c52-11e3-0800-6686b0fba2c7 stable
131224 4:19:57 [Note] WSREP: view(view_id(PRIM,771b0086-6c52-11e3-0800-6686b0fba2c7,2) memb {
771b0086-6c52-11e3-0800-6686b0fba2c7,
a5560fc4-6c52-11e3-0800-6de37b8550ef,
} joined {
} left {
} partitioned {
})
131224 4:19:57 [Note] WSREP: gcomm: connected
131224 4:19:57 [Note] WSREP: Changing maximum packet size to 64500, resulting msg size: 32636
131224 4:19:57 [Note] WSREP: Shifting CLOSED → OPEN (TO: 0)
131224 4:19:57 [Note] WSREP: Opened channel ‘ecomm’
131224 4:19:57 [Note] WSREP: New COMPONENT: primary = yes, bootstrap = no, my_idx = 1, memb_num = 2
131224 4:19:57 [Note] WSREP: STATE EXCHANGE: Waiting for state UUID.
131224 4:19:57 [Note] WSREP: Waiting for SST to complete.
131224 4:19:57 [Note] WSREP: STATE EXCHANGE: sent state msg: a59fd8b1-6c52-11e3-0800-74c11aaf856b
131224 4:19:57 [Note] WSREP: STATE EXCHANGE: got state msg: a59fd8b1-6c52-11e3-0800-74c11aaf856b from 0 (FD-R14-db2)
131224 4:19:57 [Note] WSREP: STATE EXCHANGE: got state msg: a59fd8b1-6c52-11e3-0800-74c11aaf856b from 1 (FD-R14-db3)
131224 4:19:57 [Note] WSREP: Quorum results:
version = 2,
component = PRIMARY,
conf_id = 1,
members = 1/2 (joined/total),
act_id = 172,
last_appl. = -1,
protocols = 0/4/2 (gcs/repl/appl),
group UUID = 493d95e8-6c50-11e3-0800-b099dbf50e3d
131224 4:19:57 [Note] WSREP: Flow-control interval: [12, 23]
131224 4:19:57 [Note] WSREP: Shifting OPEN → PRIMARY (TO: 172)
131224 4:19:57 [Note] WSREP: State transfer required:
Group state: 493d95e8-6c50-11e3-0800-b099dbf50e3d:172
Local state: 00000000-0000-0000-0000-000000000000:-1
131224 4:19:57 [Note] WSREP: New cluster view: global state: 493d95e8-6c50-11e3-0800-b099dbf50e3d:172, view# 2: Primary, number of nodes: 2, my index: 1, protocol version 2
131224 4:19:57 [Warning] WSREP: Gap in state sequence. Need state transfer.
131224 4:19:59 [Note] WSREP: Running: ‘wsrep_sst_rsync ‘joiner’ ‘X.X.X.3’ ‘’ ‘/mnt/data/’ ‘/etc/my.cnf’ ‘15049’ 2>sst.err’
131224 4:19:59 [Note] WSREP: Prepared SST request: rsync|X.X.X.3:4444/rsync_sst
131224 4:19:59 [Note] WSREP: wsrep_notify_cmd is not defined, skipping notification.
131224 4:19:59 [Note] WSREP: Assign initial position for certification: 172, protocol version: 2
131224 4:19:59 [Warning] WSREP: Failed to prepare for incremental state transfer: Local state UUID (00000000-0000-0000-0000-000000000000) does not match group state UUID (493d95e8-6c50-11e3-0800-b099dbf50e3d): 1 (Operation not permitted)
at galera/src/replicator_str.cpp:prepare_for_IST():440. IST will be unavailable.
131224 4:19:59 [Note] WSREP: Node 1 (FD-R14-db3) requested state transfer from ‘any’. Selected 0 (FD-R14-db2)(SYNCED) as donor.
131224 4:19:59 [Note] WSREP: Shifting PRIMARY → JOINER (TO: 172)
131224 4:19:59 [Note] WSREP: Requesting state transfer: success, donor: 0

Node No 1 :
Error Log:
at galera/src/replicator_str.cpp:prepare_for_IST():440. IST will be unavailable.
131224 4:12:17 [Note] WSREP: Node 1 (FD-R14-db1) requested state transfer from ‘any’. Selected 0 (FD-R14-db2)(SYNCED) as donor.
131224 4:12:17 [Note] WSREP: Shifting PRIMARY → JOINER (TO: 172)
131224 4:12:17 [Note] WSREP: Requesting state transfer: success, donor: 0
131224 04:17:09 mysqld_safe Starting mysqld daemon with databases from /mnt/data/
131224 4:17:09 [Note] Flashcache bypass: disabled
131224 4:17:09 [Note] Flashcache setup error is : ioctl failed

131224 4:17:09 [Note] WSREP: Read nil XID from storage engines, skipping position init
131224 4:17:09 [Note] WSREP: wsrep_load(): loading provider library ‘/usr/lib64/libgalera_smm.so’
131224 4:17:09 [Note] WSREP: wsrep_load(): Galera 2.2(r114) by Codership Oy <info@codership.com> loaded succesfully.
131224 4:17:09 [Note] WSREP: Found saved state: 00000000-0000-0000-0000-000000000000:-1
131224 4:17:09 [Note] WSREP: Reusing existing ‘/mnt/data//galera.cache’.
131224 4:17:09 [Note] WSREP: Passing config to GCS: base_host = X.X.X.1; gcache.dir = /mnt/data/; gcache.keep_pages_size = 0; gcache.mem_size = 0; gcache.name = /mnt/data//galera.cache; gcache.page_size = 128M; gcache.size = 128M; gcs.fc_debug = 0; gcs.fc_factor = 0.5; gcs.fc_limit = 16; gcs.fc_master_slave = NO; gcs.max_packet_size = 64500; gcs.max_throttle = 0.25; gcs.recv_q_hard_limit = 9223372036854775807; gcs.recv_q_soft_limit = 0.25; gcs.sync_donor = NO; replicator.causal_read_timeout = PT30S; replicator.commit_order = 3
131224 4:17:09 [Note] WSREP: Assign initial position for certification: -1, protocol version: -1
131224 4:17:09 [Note] WSREP: wsrep_sst_grab()
131224 4:17:09 [Note] WSREP: Start replication
131224 4:17:09 [Note] WSREP: Setting initial position to 00000000-0000-0000-0000-000000000000:-1
131224 4:17:09 [Note] WSREP: protonet asio version 0
131224 4:17:09 [Note] WSREP: backend: asio
131224 4:17:09 [Note] WSREP: GMCast version 0
131224 4:17:09 [Note] WSREP: (41e1954d-6c52-11e3-0800-a5888b35da99, ‘tcp://0.0.0.0:4567’) listening at tcp://0.0.0.0:4567
131224 4:17:09 [Note] WSREP: (41e1954d-6c52-11e3-0800-a5888b35da99, ‘tcp://0.0.0.0:4567’) multicast: , ttl: 1
131224 4:17:09 [Note] WSREP: EVS version 0
131224 4:17:09 [Note] WSREP: PC version 0
131224 4:17:09 [Note] WSREP: gcomm: connecting to group ‘ecomm’, peer ‘X.X.X.2:’
131224 4:17:10 [Note] WSREP: declaring 493c5523-6c50-11e3-0800-25c9bc4e1fff stable
131224 4:17:10 [Note] WSREP: view(view_id(NON_PRIM,41e1954d-6c52-11e3-0800-a5888b35da99,8) memb {
41e1954d-6c52-11e3-0800-a5888b35da99,
493c5523-6c50-11e3-0800-25c9bc4e1fff,
} joined {
} left {
} partitioned {
0975f0b3-6c52-11e3-0800-2d07df149f30,
9183e102-6c51-11e3-0800-62536ac728a9,
})
131224 4:17:40 [ERROR] WSREP: failed to open gcomm backend connection: 110: failed to reach primary view: 110 (Connection timed out)
at gcomm/src/pc.cpp:connect():157
131224 4:17:40 [ERROR] WSREP: gcs/src/gcs_core.c:gcs_core_open():195: Failed to open backend connection: -110 (Connection timed out)
131224 4:17:40 [ERROR] WSREP: gcs/src/gcs.c:gcs_open():1290: Failed to open channel ‘ecomm’ at ‘gcomm://X.X.X.2’: -110 (Connection timed out)
131224 4:17:40 [ERROR] WSREP: gcs connect failed: Connection timed out
131224 4:17:40 [ERROR] WSREP: wsrep::connect() failed: 6
131224 4:17:40 [ERROR] Aborting
131224 4:17:40 [Note] WSREP: Service disconnected.
131224 4:17:41 [Note] WSREP: Some threads may fail to exit.
131224 4:17:41 [Note] /usr/sbin/mysqld: Shutdown complete
131224 04:17:41 mysqld_safe mysqld from pid file /mnt/data//FD-R14-db1.pid ended

Advance Thanks,
Periyannan N.

2

looks like both node1 & 3 needs SST.

make node 2 as primary
in mysql prompt give
SET GLOBAL wsrep_provider_options=“pc.bootstrap=1”;
(if u have done already then leave it)

then put =node2 address in both node 1& node3 my.cnf file and then try to start again.

If you provide some more information about you setup it will be helpful, like OS,percona version,status of innobackup.backup.log,innobackup.prepare.log (if any).

3

Hi Madhusudan,

First Thanks for your reply. Please find the below are my server setup:

OS Version: CentOS 6.3 (Final)

Percona Version: Percona-XtraDB-Cluster-server-5.5.27-23.6.356.rhel6.x86_64

And then we are using rsync for SST. I think we dont have innobackup.backup.log,innobackup.prepare.log.

Please provide any more suggestion, i am new to this Percona XtraDB Cluster.

Advance Thanks,

Regards,
Periyannan N.

4

did u try, what I suggested.? and your cluster is a production or in testing…?

more info about recovering cluster…
[URL]Percona XtraDB Cluster

5

Hi,

This production database, we took down time tomorrow only, we could not able to check right now.

And then for information, we are using galera-2.0

Thanks and Regards,
Periyannan N.

6

Hi,

For your Information,

And then here, three nodes are act as a master only(multi master).

Now the node no: 2 is running,
we didnt execute the “SET GLOBAL wsrep_provider_options=“pc.bootstrap=1”;” from begin onwards.
we didnt put “wsrep_sst_donor=node2” in my.cnf file from begin onwards.

Thanks and Regards,
Periyannan N.

7

Hi,

whether i need to delete the below log before execute the rsync command.
GRA_1_151865761.log
rsync_sst.conf
rsync_sst.pid

Thanks and Regards,
Periyannan N.

8

While adding node to the Cluster server I m not able to start mysql service.
And I have read the logs and I found below error on logs 
Failed to open Gcomm backend connections:  -110 (connection timeout) 
Failed to open channel ‘mysql1-cluster’ at ‘Gcomm:/(cluster IP)’ : -110 (connection time out)

Please help to resolve this issue…!