TokuDB unknown error -100005 when starting up from a logical data backup

MySQL & MariaDB Other MySQL® Questions
1

I am getting the above error when trying to start my Target server after doing a data copy from Source to Target. Does anyone know what this means? My google-fu hasn’t provided any answers unfortunately, and i’m a bit concerned about the ‘unknown’ nature of it.
Some technical details below:

Source MySQL database version: MariaDB 10.1.37
Target MySQL database version: MariaDB 10.1.13
Steps with command performed while taking backup: use mylvmbackup command.
Backup file transfer/copy command:

mylvmbackup --user=XXXX --password=XXXX --vgname=G1SRV2-vg-storage –lvname=storage --backuptype=tar --backupdir=/mnt/g1srv1 --lvsize=8G --skip_mycnf

OS details of database servers:Source: Debian 9.1Destination: Debian 7.10
Restoration steps details with command:extract tar file on Target machine. give mysql user permissions to extracted mysql directory. Move extracted mysql directory to mysql data dir path. Restart target mysql server.

Backup tool version  (both Source and Target server):mylvmbackup 0.15-1.1





2

Can you provide more details from Error Log ?

Couple of things to point out - you’re doing Physical backup with LVM not Logical backup.     MyLVMBackup is NOT designed or tested with TokuDB as such it well may not copy all files needed.  Also  You’re copying backup from newer to older version  which may be problematic

3

Hi Peter,
I’ve pasted the error logs from the syslog here. The only useful things i can see is the TokuDB plugins already installed, then the unknown error below that.
Ah my apologies for the incorrect terminology. Yes i have done a bit of reading and it seems there isn’t really any good way to move a TokuDB database. Especially large ones where dump/restore is impractical. However i have manged to do it when i initially set up the Slave server a few years ago.
I’m certainly interested in supported ways to efficiently copy the data to restart a slave replication process.

Cheers