# Problem reboot all node **URL:** \ **Category:** Percona XtraDB Cluster 5.x\ **Created:** [April 13, 2017, 2:21am UTC](https://forums.percona.com/t/problem-reboot-all-node/5544 "2017-04-13T02:21:41Z")\ **Posts on this page:** 6\ **Page:** 1
**Author:** ![vuxitrum99](https://avatars.discourse-cdn.com/v4/letter/v/278dde/32.png) [@vuxitrum99](https://forums.percona.com/u/vuxitrum99)\ **Post date:** [April 13, 2017, 2:21am UTC](https://forums.percona.com/t/problem-reboot-all-node/5544/1 "2017-04-13T02:21:41Z")
I have 3 node. Then i restart 3 node in the same time that Making it for many times to run the system But failed I Try run: # /etc/init.d/mysql bootstrap-pxc Bootstrapping PXC (Percona XtraDB Cluster)MySQL (Percona XtraDB Cluster) is not running, but lock file (/var/lock/subsys/mysql) exists [FAILED] Starting MySQL (Percona XtraDB Cluster)…The server quit without updating PID file (/var/lib/mysql/CentOS64-IP203.pid). [FAILED] MySQL (Percona XtraDB Cluster) server startup failed! [FAILED] # service mysql restart Shutting down MySQL (Percona XtraDB Cluster)MySQL (Percona [FAILED]luster) PID file could not be found! MySQL (Percona XtraDB Cluster) is not running, but lock fil[FAILED]lock/subsys/mysql) exists Starting MySQL (Percona XtraDB Cluster)…The server quit without updating PID file (/var/lib/mysql/CentOS64-IP203.pid). [FAILED] MySQL (Percona XtraDB Cluster) server startup failed! [FAILED] Failed to restart server. [FAILED] You have new mail in /var/spool/mail/root # /etc/init.d/mysql start --wsrep-cluster-address=“gcomm://” MySQL (Percona XtraDB Cluster) is not running, but lock fil[FAILED]lock/subsys/mysql) exists Starting MySQL (Percona XtraDB Cluster)…The server quit without updating PID file (/var/lib/mysql/CentOS64-IP203.pid). [FAILED] MySQL (Percona XtraDB Cluster) server startup failed! [FAILED] My Log: # tail -f CentOS64-IP203.err 2017-04-13T07:48:24.197488Z 0 [ERROR] WSREP: It may not be safe to bootstrap the cluster from this node. It was not the last one to leave the cluster and may not contain all the updates. To force cluster bootstrap with this node, edit the grastate.dat file manually and set safe\_to\_bootstrap to 1 . 2017-04-13T07:48:24.197503Z 0 [ERROR] WSREP: wsrep::connect(gcomm://) failed: 7 2017-04-13T07:48:24.197510Z 0 [ERROR] Aborting 2017-04-13T07:48:24.197518Z 0 [Note] Giving 0 client threads a chance to die gracefully 2017-04-13T07:48:24.197529Z 0 [Note] WSREP: Service disconnected. 2017-04-13T07:48:27.197656Z 0 [Note] WSREP: Some threads may fail to exit. 2017-04-13T07:48:27.197726Z 0 [Note] Binlog end 2017-04-13T07:48:27.197935Z 0 [Note] /usr/sbin/mysqld: Shutdown complete please help me ---
**Author:** ![przemek](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/przemek/32/3_2.png) [@przemek](https://forums.percona.com/u/przemek)\ **Post date:** [April 18, 2017, 5:38am UTC](https://forums.percona.com/t/problem-reboot-all-node/5544/2 "2017-04-18T05:38:26Z")
In recent PXC versions an additional check was added, hence you have the message telling you it may be not safe to bootstrap the cluster from this particular node, and what you can do to force it anyway: ```auto 2017-04-13T07:48:24.197488Z 0 [ERROR] WSREP: It may not be safe to bootstrap the cluster from this node. It was not the last one to leave the cluster and may not contain all the updates. To force cluster bootstrap with this node, edit the grastate.dat file manually and set safe_to_bootstrap to 1 . ``` But what you really should do it to check all the nodes and see which one was stopped as the last one, as it may have the furthest writeset position. Also, such node, which was last one stopped, should have the > [@](#): > > safe\_to\_bootstrap: 1 value in the grastate.dat file. ---
**Author:** ![vuxitrum99](https://avatars.discourse-cdn.com/v4/letter/v/278dde/32.png) [@vuxitrum99](https://forums.percona.com/u/vuxitrum99)\ **Post date:** [September 15, 2017, 3:16am UTC](https://forums.percona.com/t/problem-reboot-all-node/5544/3 "2017-09-15T03:16:52Z")
thanks but I need an automatic startup solution. ---
**Author:** ![DBennett](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/dbennett/32/17_2.png) [@DBennett](https://forums.percona.com/u/DBennett)\ **Post date:** [September 15, 2017, 1:41pm UTC](https://forums.percona.com/t/problem-reboot-all-node/5544/4 "2017-09-15T13:41:34Z")
Hi Phan, Our policy has always been to avoid data loss. In this particular case, forcing a specific PXC node to bootstrap may result in the loss of some data. Other nodes may have committed transactions that are not reflected on this node. If you are willing to accept the possibility of data loss, you may be able to force a specific node to bootstrap by editing the [B]. [B] –Dave ---
**Author:** ![vuxitrum99](https://avatars.discourse-cdn.com/v4/letter/v/278dde/32.png) [@vuxitrum99](https://forums.percona.com/u/vuxitrum99)\ **Post date:** [September 18, 2017, 4:39am UTC](https://forums.percona.com/t/problem-reboot-all-node/5544/5 "2017-09-18T04:39:47Z")
Hi DBennett I tried your way but it was not right with my idea. I want no need to run /etc/init.d/mysql start --wsrep-cluster-address=“gcomm://”. when a node returns it will auto start service mysql. Is there anyway to configure ? Pls help me thanks ---
**Author:** ![lance.jud](https://avatars.discourse-cdn.com/v4/letter/l/0ea827/32.png) [@lance.jud](https://forums.percona.com/u/lance.jud)\ **Post date:** [January 16, 2019, 8:48pm UTC](https://forums.percona.com/t/problem-reboot-all-node/5544/6 "2019-01-16T20:48:13Z")
Dear [URL=“[https://www.percona.com/forums/member/22487-dbennett](https://www.percona.com/forums/member/22487-dbennett)”]DBennett, I have percona cluster 5.7 on redhat 6.9 and going to have failover test. Would[/URL] you tell me more how to setup Percona cluster without forcing a specific PXC node to bootstrap ? Rgds, Lance