Upgrading With Databases Down
Upgrading with databases down involves stopping the NuoDB databases, installing the latest NuoDB software on all the hosts, then restarting the databases. If you prefer to upgrade your database with no downtime, see Rolling Upgrade: Databases Remain Available.
|All NuoDB processes on the host must be stopped to upgrade the NuoDB software on that host, even if they are from different databases or domains.|
For each database in the domain, shut down all NuoDB Storage Managers (SMs) and Transaction Engines (TEs) for that database using the
nuocmd shutdown database command.
For more information, see Shutting down a Database and NuoDB Command Reference.
While this step is optional, it is recommended as a best practice.
For each host in the domain, make a copy of the content in each SM’s archive and journal directories, as well as the NuoDB Admin
raftlog file (
Store each copy separately.
For information on backing up databases, see Backing Up and Restoring Databases.
The location of the database archive and journal directories can be determined using the
nuocmd show archives command.
For more information, see NuoDB Command Reference.
nuocmd show archive output entry for an
archive-id displays an
archive directory and a
By default the journal is stored inside the archive; if it is stored outside the archive then both need to be backed up.
Shut down all NuoDB Admin processes (APs) on each host machine. For more information, see the Domain Operations section, Starting and Stopping NuoDB Admin.
Start an AP on each host in the domain; see Starting and Stopping NuoDB Admin.
|After upgrading the database, it may be necessary to upgrade the database protocol. For more information, see Upgrading the Database Protocol.|
After completing the above steps, see After You Upgrade.