MariaDB (Galera Cluster) startet nicht - richtige config?

[anony]

Hi,

habe auf 3 Nodes (Debian 7 Wheezy) MariaDB Galera Cluster (Version 10.0.15, x64) installiert und die secure-installation (sudo /usr/bin/mysql_secure_installation) ausgeführt.

Soweit, so gut. Wenn ich jetzt aber meine Config anpasse /etc/mysql/conf.d/mariadb.cnf (Rest ist bis jetzt "Standard") und folgende Zeilen anfüge, startet es nicht mehr - es lädt und failed.

Folgende Zeilen habe ich an die Config /etc/mysql/conf.d/mariadb.cnf angefügt:

Spoiler

binlog_format=ROW
default-storage-engine=innodb
innodb_autoinc_lock_mode=2
innodb_locks_unsafe_for_binlog=1
query_cache_size=0
query_cache_type=0
bind-address=0.0.0.0
datadir=/var/lib/mysql
innodb_log_file_size=100M
innodb_file_per_table
innodb_flush_log_at_trx_commit=2
wsrep_provider=/usr/lib64/galera/libgalera_smm.so
wsrep_cluster_address="gcomm://$ip1,$ip2,$ip3"
wsrep_cluster_name='galera_cluster'
wsrep_node_address='$ip1'
wsrep_node_name='$name1'
wsrep_sst_method=rsync
wsrep_sst_auth=userass

Die Platzhalter $name1 und $ip1, $ip2 sowie $ip3 tausche ich natürlich durch die Namen / IPs der jeweiligen Nodes. Sonst ist die Config auf allen 3 Nodes gleich...

Versuche ich nun mit sudo /etc/init.d/mysql start --wsrep-new-cluster am Node1 zu starten, failt es. Ebenso mit sudo service mysql start an den anderen beiden.

Hat jemand Ahnung? Sieht jemand den Fehler?

 

[Hank Moody]

Was sagt denn das Log?

 

[anony]

Das schweigt, leider. /var/log/mysql.err, */mysql.log sind leer.

 

[Hank Moody]

Im syslog steht auch nichts?

 

[anony]

Hier die logs aus 2 runs:

Spoiler

Jan 15 09:57:16 vm-mariadb1 mysqld_safe: Starting mysqld daemon with databases from /var/lib/mysql
Jan 15 09:57:16 vm-mariadb1 mysqld_safe: WSREP: Running position recovery with --log_error='/var/lib/mysql/wsrep_recovery.lSUIRl' --pid-file='/var/lib/mysql/vm-mariadb1-recover.pid'
Jan 15 09:57:19 vm-mariadb1 mysqld_safe: WSREP: Recovered position 00000000-0000-0000-0000-000000000000:-1
Jan 15 09:57:19 vm-mariadb1 mysqld: 150115 9:57:19 [Note] WSREP: wsrep_start_position var submitted: '00000000-0000-0000-0000-000000000000:-1'
Jan 15 09:57:19 vm-mariadb1 mysqld: 150115 9:57:19 [Note] WSREP: Read nil XID from storage engines, skipping position init
Jan 15 09:57:19 vm-mariadb1 mysqld: 150115 9:57:19 [Note] WSREP: wsrep_load(): loading provider library '/usr/lib64/galera/libgalera_smm.so'
Jan 15 09:57:19 vm-mariadb1 mysqld: 150115 9:57:19 [ERROR] WSREP: wsrep_load(): dlopen(): /usr/lib64/galera/libgalera_smm.so: cannot open shared object file: No such file or directory
Jan 15 09:57:19 vm-mariadb1 mysqld: 150115 9:57:19 [ERROR] WSREP: wsrep_load(/usr/lib64/galera/libgalera_smm.so) failed: Invalid argument (22). Reverting to no provider.
Jan 15 09:57:19 vm-mariadb1 mysqld: 150115 9:57:19 [Note] WSREP: Read nil XID from storage engines, skipping position init
Jan 15 09:57:19 vm-mariadb1 mysqld: 150115 9:57:19 [Note] WSREP: wsrep_load(): loading provider library 'none'
Jan 15 09:57:19 vm-mariadb1 mysqld: 150115 9:57:19 [ERROR] Aborting
Jan 15 09:57:19 vm-mariadb1 mysqld:
Jan 15 09:57:19 vm-mariadb1 mysqld: 150115 9:57:19 [Note] WSREP: Service disconnected.
Jan 15 09:57:20 vm-mariadb1 mysqld: 150115 9:57:20 [Note] WSREP: Some threads may fail to exit.
Jan 15 09:57:20 vm-mariadb1 mysqld: 150115 9:57:20 [Note] /usr/sbin/mysqld: Shutdown complete
Jan 15 09:57:20 vm-mariadb1 mysqld:
Jan 15 09:57:20 vm-mariadb1 mysqld_safe: mysqld from pid file /var/run/mysqld/mysqld.pid ended
Jan 15 09:57:48 vm-mariadb1 /etc/init.d/mysql[4750]: 0 processes alive and '/usr/bin/mysqladmin --defaults-file=/etc/mysql/debian.cnf ping' resulted in
Jan 15 09:57:48 vm-mariadb1 /etc/init.d/mysql[4750]: #007/usr/bin/mysqladmin: connect to server at 'localhost' failed
Jan 15 09:57:48 vm-mariadb1 /etc/init.d/mysql[4750]: error: 'Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (111 "Connection refused")'
Jan 15 09:57:48 vm-mariadb1 /etc/init.d/mysql[4750]: Check that mysqld is running and that the socket: '/var/run/mysqld/mysqld.sock' exists!
Jan 15 09:57:48 vm-mariadb1 /etc/init.d/mysql[4750]:
Jan 15 09:58:11 vm-mariadb1 mysqld_safe: Starting mysqld daemon with databases from /var/lib/mysql
Jan 15 09:58:11 vm-mariadb1 mysqld_safe: WSREP: Running position recovery with --log_error='/var/lib/mysql/wsrep_recovery.ybaCNe' --pid-file='/var/lib/mysql/vm-mariadb1-recover.pid'
Jan 15 09:58:14 vm-mariadb1 mysqld_safe: WSREP: Recovered position 00000000-0000-0000-0000-000000000000:-1
Jan 15 09:58:14 vm-mariadb1 mysqld: 150115 9:58:14 [Note] WSREP: wsrep_start_position var submitted: '00000000-0000-0000-0000-000000000000:-1'
Jan 15 09:58:14 vm-mariadb1 mysqld: 150115 9:58:14 [Note] WSREP: Read nil XID from storage engines, skipping position init
Jan 15 09:58:14 vm-mariadb1 mysqld: 150115 9:58:14 [Note] WSREP: wsrep_load(): loading provider library '/usr/lib64/galera/libgalera_smm.so'
Jan 15 09:58:14 vm-mariadb1 mysqld: 150115 9:58:14 [ERROR] WSREP: wsrep_load(): dlopen(): /usr/lib64/galera/libgalera_smm.so: cannot open shared object file: No such file or directory
Jan 15 09:58:14 vm-mariadb1 mysqld: 150115 9:58:14 [ERROR] WSREP: wsrep_load(/usr/lib64/galera/libgalera_smm.so) failed: Invalid argument (22). Reverting to no provider.
Jan 15 09:58:14 vm-mariadb1 mysqld: 150115 9:58:14 [Note] WSREP: Read nil XID from storage engines, skipping position init
Jan 15 09:58:14 vm-mariadb1 mysqld: 150115 9:58:14 [Note] WSREP: wsrep_load(): loading provider library 'none'
Jan 15 09:58:14 vm-mariadb1 mysqld: 150115 9:58:14 [ERROR] Aborting
Jan 15 09:58:14 vm-mariadb1 mysqld:
Jan 15 09:58:14 vm-mariadb1 mysqld: 150115 9:58:14 [Note] WSREP: Service disconnected.
Jan 15 09:58:15 vm-mariadb1 mysqld: 150115 9:58:15 [Note] WSREP: Some threads may fail to exit.
Jan 15 09:58:15 vm-mariadb1 mysqld: 150115 9:58:15 [Note] /usr/sbin/mysqld: Shutdown complete
Jan 15 09:58:15 vm-mariadb1 mysqld:
Jan 15 09:58:15 vm-mariadb1 mysqld_safe: mysqld from pid file /var/run/mysqld/mysqld.pid ended
Jan 15 09:58:44 vm-mariadb1 /etc/init.d/mysql[5425]: 0 processes alive and '/usr/bin/mysqladmin --defaults-file=/etc/mysql/debian.cnf ping' resulted in
Jan 15 09:58:44 vm-mariadb1 /etc/init.d/mysql[5425]: #007/usr/bin/mysqladmin: connect to server at 'localhost' failed
Jan 15 09:58:44 vm-mariadb1 /etc/init.d/mysql[5425]: error: 'Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (111 "Connection refused")'
Jan 15 09:58:44 vm-mariadb1 /etc/init.d/mysql[5425]: Check that mysqld is running and that the socket: '/var/run/mysqld/mysqld.sock' exists!
Jan 15 09:58:44 vm-mariadb1 /etc/init.d/mysql[5425]:

 

[Hank Moody]

[src=bash]Jan 15 09:57:19 vm-mariadb1 mysqld: 150115 9:57:19 [ERROR] WSREP: wsrep_load(): dlopen(): /usr/lib64/galera/libgalera_smm.so: cannot open shared object file: No such file or directory
Jan 15 09:57:19 vm-mariadb1 mysqld: 150115 9:57:19 [ERROR] WSREP: wsrep_load(/usr/lib64/galera/libgalera_smm.so) failed: Invalid argument (22). Reverting to no provider.[/src]

Hier ist dein Problem - Google hat jetzt leider auf die Schnelle nichts sinnvolles aufgezeigt... wenn du den Fehler beheben kannst, wird MariaDB wohl starten.

Nachtrag: welche Pakete hast du genau installiert und welche Architektur hat dein Debian? 64-Bit?

 

[anony]

Danke dir, bin dann mal googlen

You do not have permission to view link please Anmelden or Registrieren

Node1 läuft jetzt, sieht gut aus. Alles aus ON, alles läuft. :-) (64bit, ja)
mysql -u root -p -e "show status like 'wsrep%'" spuckt folgendes aus (unwichtige lines entfernt):

Spoiler

+------------------------------+--------------------------------------+
| Variable_name | Value |
+------------------------------+--------------------------------------+
| wsrep_local_state_comment | Synced |
| wsrep_incoming_addresses | $ip1:3306 |
| wsrep_cluster_conf_id | 1 |
| wsrep_cluster_size | 1 |
| wsrep_cluster_status | Primary |
| wsrep_connected | ON |
| wsrep_ready | ON |

Man sollte also davon ausgehen, dass alles passt, richtig? Dem ist aber nicht so, da die anderen beiden Nodes 2 und 3 leider nicht connecten können. Log folgt.

Log:

Spoiler

Jan 15 10:36:37 vm-mariadb2 mysqld_safe: Starting mysqld daemon with databases from /var/lib/mysql
Jan 15 10:36:37 vm-mariadb2 mysqld_safe: WSREP: Running position recovery with --log_error='/var/lib/mysql/wsrep_recovery.g3toCI' --pid-file='/var/lib/mysql/vm-mariadb2-recover.pid'
Jan 15 10:36:39 vm-mariadb2 mysqld_safe: WSREP: Recovered position 00000000-0000-0000-0000-000000000000:-1
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: wsrep_start_position var submitted: '00000000-0000-0000-0000-000000000000:-1'
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: Read nil XID from storage engines, skipping position init
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: wsrep_load(): loading provider library '/usr/lib64/galera/libgalera_smm.so'
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: wsrep_load(): Galera 25.3.5-wheezy(rXXXX) by Codership Oy <info@codership.com> loaded successfully.
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: CRC-32C: using hardware acceleration.
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: Found saved state: 00000000-0000-0000-0000-000000000000:-1
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: Passing config to GCS: base_host = $ip2; base_port = 4567; cert.log_conflicts = no; debug = no; evs.inactive_check_period = PT0.5S; evs.inactive_timeout = PT15S; evs.join_retrans_period = PT1S; evs.max_install_timeouts = 1; evs.send_window = 4; evs.stats_report_period = PT1M; evs.suspect_timeout = PT5S; evs.user_send_window = 2; evs.view_forget_timeout = PT24H; gcache.dir = /var/lib/mysql/; gcache.keep_pages_size = 0; gcache.mem_size = 0; gcache.name = /var/lib/mysql//galera.cache; gcache.page_size = 128M; gcache.size = 128M; gcs.fc_debug = 0; gcs.fc_factor = 1.0; 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; gmcast.segment = 0; gmcast.version = 0; pc.announce_timeout = PT3S; pc.checksum = false; pc.ignore_quorum = false; pc.ignore_sb = false; pc.npvo = false; pc.version = 0; pc.wait_prim = true; pc.wait_prim_timeout = P30S; p
Jan 15 10:36:39 vm-mariadb2 mysqld: c.weight = 1; protonet.
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: Service thread queue flushed.
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: Assign initial position for certification: -1, protocol version: -1
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: wsrep_sst_grab()
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: Start replication
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: Setting initial position to 00000000-0000-0000-0000-000000000000:-1
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: protonet asio version 0
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: Using CRC-32C (optimized) for message checksums.
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: backend: asio
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: GMCast version 0
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: (01d92f99-9c9a-11e4-8809-b2c266ac5fb0, 'tcp://0.0.0.0:4567') listening at tcp://0.0.0.0:4567
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: (01d92f99-9c9a-11e4-8809-b2c266ac5fb0, 'tcp://0.0.0.0:4567') multicast: , ttl: 1
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: EVS version 0
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: PC version 0
Jan 15 10:36:39 vm-mariadb2 mysqld: 150115 10:36:39 [Note] WSREP: gcomm: connecting to group 'galera_cluster', peer '$ip1:,$ip2:,$ip3:'
Jan 15 10:36:42 vm-mariadb2 mysqld: 150115 10:36:42 [Warning] WSREP: no nodes coming from prim view, prim not possible
Jan 15 10:36:42 vm-mariadb2 mysqld: 150115 10:36:42 [Note] WSREP: view(view_id(NON_PRIM,01d92f99-9c9a-11e4-8809-b2c266ac5fb0,1) memb {
Jan 15 10:36:42 vm-mariadb2 mysqld: #01101d92f99-9c9a-11e4-8809-b2c266ac5fb0,0
Jan 15 10:36:42 vm-mariadb2 mysqld: } joined {
Jan 15 10:36:42 vm-mariadb2 mysqld: } left {
Jan 15 10:36:42 vm-mariadb2 mysqld: } partitioned {
Jan 15 10:36:42 vm-mariadb2 mysqld: })
Jan 15 10:36:43 vm-mariadb2 mysqld: 150115 10:36:43 [Warning] WSREP: last inactive check more than PT1.5S ago (PT3.50274S), skipping check
Jan 15 10:37:09 vm-mariadb2 /etc/init.d/mysql[4237]: 0 processes alive and '/usr/bin/mysqladmin --defaults-file=/etc/mysql/debian.cnf ping' resulted in
Jan 15 10:37:09 vm-mariadb2 /etc/init.d/mysql[4237]: #007/usr/bin/mysqladmin: connect to server at 'localhost' failed
Jan 15 10:37:09 vm-mariadb2 /etc/init.d/mysql[4237]: error: 'Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (111 "Connection refused")'
Jan 15 10:37:09 vm-mariadb2 /etc/init.d/mysql[4237]: Check that mysqld is running and that the socket: '/var/run/mysqld/mysqld.sock' exists!
Jan 15 10:37:09 vm-mariadb2 /etc/init.d/mysql[4237]:
Jan 15 10:37:12 vm-mariadb2 mysqld: 150115 10:37:12 [Note] WSREP: view((empty))
Jan 15 10:37:12 vm-mariadb2 mysqld: 150115 10:37:12 [ERROR] WSREP: failed to open gcomm backend connection: 110: failed to reach primary view: 110 (Connection timed out)
Jan 15 10:37:12 vm-mariadb2 mysqld: #011 at gcomm/src/pc.cpp:connect():141
Jan 15 10:37:12 vm-mariadb2 mysqld: 150115 10:37:12 [ERROR] WSREP: gcs/src/gcs_core.c:gcs_core_open():202: Failed to open backend connection: -110 (Connection timed out)
Jan 15 10:37:12 vm-mariadb2 mysqld: 150115 10:37:12 [ERROR] WSREP: gcs/src/gcs.c:gcs_open():1291: Failed to open channel 'galera_cluster' at 'gcomm://$ip1,$ip2,$ip3': -110 (Connection timed out)
Jan 15 10:37:12 vm-mariadb2 mysqld: 150115 10:37:12 [ERROR] WSREP: gcs connect failed: Connection timed out
Jan 15 10:37:12 vm-mariadb2 mysqld: 150115 10:37:12 [ERROR] WSREP: wsrep::connect() failed: 7
Jan 15 10:37:12 vm-mariadb2 mysqld: 150115 10:37:12 [ERROR] Aborting
Jan 15 10:37:12 vm-mariadb2 mysqld:
Jan 15 10:37:12 vm-mariadb2 mysqld: 150115 10:37:12 [Note] WSREP: Service disconnected.
Jan 15 10:37:13 vm-mariadb2 mysqld: 150115 10:37:13 [Note] WSREP: Some threads may fail to exit.
Jan 15 10:37:13 vm-mariadb2 mysqld: 150115 10:37:13 [Note] /usr/sbin/mysqld: Shutdown complete
Jan 15 10:37:13 vm-mariadb2 mysqld:
Jan 15 10:37:13 vm-mariadb2 mysqld_safe: mysqld from pid file /var/run/mysqld/mysqld.pid ended

--- [2015-01-15 10:46 CET] Automatisch zusammengeführter Beitrag ---

Update:
Sorry, war jetzt selbst ein bisschen blind. Würde vermutlich funktionieren, wenn die Hosts untereinander kommunizieren könnten...

 

[Hank Moody]

Darf man fragen, wie du das Ausgangs-Problem gelöst hast?

 

[anony]

War'n Galera-Problem, da hat einfach der Pfad in der config nicht gestimmt... (wsrep_provider)

Danke dir nochmal!