# Restore backup using xtrabackup **URL:** \ **Category:** Percona XtraDB Cluster 5.x\ **Created:** [March 6, 2025, 6:41pm UTC](https://forums.percona.com/t/restore-backup-using-xtrabackup/37082 "2025-03-06T18:41:51Z")\ **Posts on this page:** 11\ **Page:** 1
**Author:** ![ALFR3D022](https://avatars.discourse-cdn.com/v4/letter/a/f08c70/32.png) [@ALFR3D022](https://forums.percona.com/u/ALFR3D022)\ **Post date:** [March 6, 2025, 6:41pm UTC](https://forums.percona.com/t/restore-backup-using-xtrabackup/37082/1 "2025-03-06T18:41:51Z")
I need to restore a backup made by xtrabackup, but I am using xtradbcluster to replicate the main database in the secondary database in an almost synchronous way. I was able to restore it to the main database using resync, but it did not replicate to the second database. ---
**Author:** ![matthewb](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/matthewb/32/34_2.png) [@matthewb](https://forums.percona.com/u/matthewb)\ **Post date:** [March 6, 2025, 9:12pm UTC](https://forums.percona.com/t/restore-backup-using-xtrabackup/37082/2 "2025-03-06T21:12:14Z")
> [@ALFR3D022](#): > > I was able to restore it to the main database using resync What is “resync”? > [@ALFR3D022](#): > > but I am using xtradbcluster to replicate the main database in the secondary database in an almost synchronous way I’m confused. Can you provide a diagram of your topology? ---
**Author:** ![ALFR3D022](https://avatars.discourse-cdn.com/v4/letter/a/f08c70/32.png) [@ALFR3D022](https://forums.percona.com/u/ALFR3D022)\ **Post date:** [March 6, 2025, 10:41pm UTC](https://forums.percona.com/t/restore-backup-using-xtrabackup/37082/3 "2025-03-06T22:41:57Z")
Sorry, I ended up writing it wrong, it would be “rsync”, it was the only method that worked. I saw about it in this documentation [Restore full, incremental, compressed backups - Percona XtraBackup](https://docs.percona.com/percona-xtrabackup/8.0/restore-a-backup.html). Yes, I will make the diagram: ![image](https://us1.discourse-cdn.com/flex019/uploads/percona1/original/3X/d/4/d4abf2ffa02815a9b218bd2f69dcce75e397a8fc.png) I restored the backup from node 1, but it did not replicate the data to node 2. ![image](https://us1.discourse-cdn.com/flex019/uploads/percona1/original/3X/c/5/c54dbd6d59bba04588d0faf5561679a8fc632a7b.png) ---
**Author:** ![satya.bodapati](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/satya.bodapati/32/1322_2.png) [@satya.bodapati](https://forums.percona.com/u/satya.bodapati)\ **Post date:** [March 7, 2025, 1:57pm UTC](https://forums.percona.com/t/restore-backup-using-xtrabackup/37082/4 "2025-03-07T13:57:12Z")
Do I understand it correctly that you have 2 nodes PXC setup. Node A and Node B. You stopped B gracefully(I hope) and restored the backup from Node A on Node B. Then you start server on node B and expect PXC to work as before? Please clarify if you use async replication or PXC cluster between nodes? ---
**Author:** ![ALFR3D022](https://avatars.discourse-cdn.com/v4/letter/a/f08c70/32.png) [@ALFR3D022](https://forums.percona.com/u/ALFR3D022)\ **Post date:** [March 7, 2025, 2:18pm UTC](https://forums.percona.com/t/restore-backup-using-xtrabackup/37082/5 "2025-03-07T14:18:33Z")
OK, I stopped both nodes, restored the backup on node 1, and started node 1 and node 2. However, the backup did not replicate to node 2. Should I have done something else? I am using PXC Cluster. ---
**Author:** ![matthewb](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/matthewb/32/34_2.png) [@matthewb](https://forums.percona.com/u/matthewb)\ **Post date:** [March 7, 2025, 9:57pm UTC](https://forums.percona.com/t/restore-backup-using-xtrabackup/37082/6 "2025-03-07T21:57:06Z")
> [@ALFR3D022](#): > > OK, I stopped both nodes, restored the backup on node 1, and started node 1 Correct. You should have started node1 in bootstrap mode (systemctl start mysql@bootstrap). This creates the cluster. You must always bootstrap the first node when starting the cluster. > [@ALFR3D022](#): > > However, the backup did not replicate to node 2. Did you configure node2 wsrep\_cluster\_address to include node1? If you do not configure node2 to connect to node1, then yea, nothing will happen. ---
**Author:** ![ALFR3D022](https://avatars.discourse-cdn.com/v4/letter/a/f08c70/32.png) [@ALFR3D022](https://forums.percona.com/u/ALFR3D022)\ **Post date:** [March 7, 2025, 10:09pm UTC](https://forums.percona.com/t/restore-backup-using-xtrabackup/37082/7 "2025-03-07T22:09:17Z")
I configured it this way: Node 1: #Template my.cnf for PXC #Edit to your requirements. [client] socket=/var/run/mysqld/mysqld.sock [mysqld] server-id=1 datadir=/var/lib/mysql socket=/var/run/mysqld/mysqld.sock log-error=/var/log/mysql/error.log pid-file=/var/run/mysqld/mysqld.pid port=3306 #Binary log expiration period is 604800 seconds, which equals 7 days binlog\_expire\_logs\_seconds=604800 ######## wsrep ############### #Path to Galera library wsrep\_provider=/usr/lib/galera4/libgalera\_smm.so #Cluster connection URL contains IPs of nodes #If no IP is found, this implies that a new cluster needs to be created, #in order to do that you need to bootstrap this node wsrep\_cluster\_address=gcomm://10.48.77.227, 10.48.77.226 #In order for Galera to work correctly binlog format should be ROW binlog\_format=ROW #Slave thread to use wsrep\_slave\_threads=8 wsrep\_log\_conflicts #This changes how InnoDB autoincrement locks are managed and is a requirement for Galera innodb\_autoinc\_lock\_mode=2 #Node IP address wsrep\_node\_address=10.48.77.227 #Cluster name wsrep\_cluster\_name=pxc-cluster #If wsrep\_node\_name is not specified, then system hostname will be used wsrep\_node\_name=percona-cluster-node #pxc\_strict\_mode allowed values: DISABLED,PERMISSIVE,ENFORCING,MASTER pxc\_strict\_mode=ENFORCING #SST method wsrep\_sst\_method=xtrabackup-v2 * * * Node 2 #Template my.cnf for PXC #Edit to your requirements. [client] socket=/var/run/mysqld/mysqld.sock [mysqld] server-id=1 datadir=/var/lib/mysql socket=/var/run/mysqld/mysqld.sock log-error=/var/log/mysql/error.log pid-file=/var/run/mysqld/mysqld.pid port=3307 #Binary log expiration period is 604800 seconds, which equals 7 days binlog\_expire\_logs\_seconds=604800 ######## wsrep ############### #Path to Galera library wsrep\_provider=/usr/lib/galera4/libgalera\_smm.so #Cluster connection URL contains IPs of nodes #If no IP is found, this implies that a new cluster needs to be created, #in order to do that you need to bootstrap this node wsrep\_cluster\_address=gcomm://10.48.77.227, 10.48.77.226 #In order for Galera to work correctly binlog format should be ROW binlog\_format=ROW #Slave thread to use wsrep\_slave\_threads=8 wsrep\_log\_conflicts #This changes how InnoDB autoincrement locks are managed and is a requirement for Galera innodb\_autoinc\_lock\_mode=2 #Node IP address wsrep\_node\_address=10.48.77.226 #Cluster name wsrep\_cluster\_name=pxc-cluster #If wsrep\_node\_name is not specified, then system hostname will be used wsrep\_node\_name=application-cluster-node #pxc\_strict\_mode allowed values: DISABLED,PERMISSIVE,ENFORCING,MASTER pxc\_strict\_mode=ENFORCING #SST method wsrep\_sst\_method=xtrabackup-v2 * * * Is there something wrong that I need to fix? It is replicating normally when I create a database and insert values. However, it does not replicate the backup. ---
**Author:** ![matthewb](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/matthewb/32/34_2.png) [@matthewb](https://forums.percona.com/u/matthewb)\ **Post date:** [March 8, 2025, 1:11am UTC](https://forums.percona.com/t/restore-backup-using-xtrabackup/37082/8 "2025-03-08T01:11:10Z")
Did you bootstrap node1? What does this show on node1 and node2? `SHOW GLOBAL STATUS LIKE 'wsrep_cluster_size'` ---
**Author:** ![ALFR3D022](https://avatars.discourse-cdn.com/v4/letter/a/f08c70/32.png) [@ALFR3D022](https://forums.percona.com/u/ALFR3D022)\ **Post date:** [March 8, 2025, 1:20am UTC](https://forums.percona.com/t/restore-backup-using-xtrabackup/37082/9 "2025-03-08T01:20:43Z")
I started node 1 with ‘systemctl start mysql@bootstrap.service’ and node 2 with ‘systemctl start mysql’ It has the same value in both nodes ‘wsrep\_cluster\_size = 2’ ![image](https://us1.discourse-cdn.com/flex019/uploads/percona1/original/3X/a/f/afd4c22235a5ca5000e1e9f776d0ab60a56c5f09.png) ---
**Author:** ![matthewb](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/matthewb/32/34_2.png) [@matthewb](https://forums.percona.com/u/matthewb)\ **Post date:** [March 8, 2025, 5:31am UTC](https://forums.percona.com/t/restore-backup-using-xtrabackup/37082/10 "2025-03-08T05:31:22Z")
I would shut down node2. Erase everything in the datadir of node2. Then start node2 mysql normally. This should force an SST of the entire data set on node1 to copy to node2. ---
**Author:** ![ALFR3D022](https://avatars.discourse-cdn.com/v4/letter/a/f08c70/32.png) [@ALFR3D022](https://forums.percona.com/u/ALFR3D022)\ **Post date:** [March 9, 2025, 1:13am UTC](https://forums.percona.com/t/restore-backup-using-xtrabackup/37082/11 "2025-03-09T01:13:10Z")
It worked! I deleted everything that was inside /var/lib/mysql/ and tried to start it again, it gave an error because it didn’t have the certificates from node 1, I copied them and tried to start node 2 again, it started and synchronized the data. Thank you very much!