Durability is an important part of your transactional guarantees. It means that once a transaction has been successfully committed, your application will always see the results of that transaction.
Of course, no software algorithm can guarantee durability in the face of physical data loss. Hard drives can fail, and if you have not copied your data to locations other than your primary disk drives, then you will lose data when those drives fail. Therefore, in order to truly obtain a durability guarantee, you need to ensure that any data stored on disk is backed up to secondary or alternative storage, such as secondary disk drives, or offline tapes.
There are three different types of backups that you can perform with DB databases and log files. They are:
Offline backups
This type of backup is perhaps the easiest to perform as it involves simply copying database and log files to an offline storage area. It also gives you a snapshot of the database at a fixed, known point in time. However, you cannot perform this type of a backup while you are performing writes to the database.
Hot backups
This type of backup gives you a snapshot of your database. Since your application can be writing to the database at the time that the snapshot is being taken, you do not necessarily know what the exact state of the database is for that given snapshot.
Incremental backups
This type of backup refreshes a previously performed backup.
Once you have performed a backup, you can perform catastrophic recovery to restore your databases from the backup. See Catastrophic Recovery for more information.
Note that you can also maintain a hot failover. See Using Hot Failovers for more information.
If you are copying database files you must copy databases atomically, in multiples of the database page size. In other words, the reads made by the copy program must not be interleaved with writes by other threads of control, and the copy program must read the databases in multiples of the underlying database page size. Generally, this is not a problem because operating systems already make this guarantee and system utilities normally read in power-of-2 sized chunks, which are larger than the largest possible Berkeley DB database page size.
On some platforms (most notably, some releases of Solaris), the copy utility (cp
) was
implemented using the mmap()
system call rather than the
read()
system call. Because mmap()
did not make the same
guarantee of read atomicity as did read()
, the cp
utility
could create corrupted copies of the databases.
Also, some platforms have implementations of the tar
utility that performs 10KB block
reads by default. Even when an output block size is specified, the utility will still not read the
underlying databases in multiples of the specified block size. Again, the result can be a corrupted backup.
To fix these problems, use the dd
utility instead of cp
or
tar
. When you use dd
, make sure you specify a block size that is
equal to, or an even multiple of, your database page size. Finally, if you plan to use a system
utility to copy database files, you may want to use a system call trace utility (for example,
ktrace
or truss
) to make sure you are not using a I/O size that is
smaller than your database page size. You can also use these utilities to make sure the system utility is
not using a system call other than read()
.
To create an offline backup:
Commit or abort all on-going transactions.
Pause all database writes.
Force a checkpoint. See Checkpoints for details.
Copy all your database files to the backup location.
Note that you can simply copy all of the database
files, or you can determine which database files
have been written during the lifetime of the current
logs. To do this, use either the
DbEnv::log_archive()
method with the DB_ARCH_DATA
option,
or use the db_archive
command with the -s
option.
However, be aware that backing up just the modified databases only works if you have all of your
log files. If you have been removing log files for any reason then using
log_archive()
can result in an
unrecoverable backup because you might not be notified of a database file that was modified.
Copy the last log file to your backup location.
Your log files are named
log.xxxxxxxxxx
,
where xxxxxxxxxx is a
sequential number. The last log file is the file
with the highest number.
To create a hot backup, you do not have to stop database operations. Transactions may be on-going and you can be writing to your database at the time of the backup. However, this means that you do not know exactly what the state of your database is at the time of the backup.
You can use the db_hotbackup command line utility to create a hot backup. This program optionally runs a checkpoint, and then copies all necessary files to a target directory.
Alternatively, you can manually create a hot backup as follows:
Set the DB_HOTBACKUP_IN_PROGRESS
flag in your environment.
For more information, see the
DbEnv::set_flags() API reference page.
Copy all your database files to the backup location.
Note that you can simply copy all of the database
files, or you can determine which database files
have been written during the lifetime of the current
logs. To do this, use either the
DbEnv::log_archive()
with the DB_ARCH_DATA
option,
or use the db_archive
command with the -s
option.
Copy all logs to your backup location.
Reset the DB_HOTBACKUP_IN_PROGRESS
flag.
It is important to copy your database files and then your logs. In this way, you can complete or roll back any database operations that were only partially completed when you copied the databases.
Once you have created a full backup (that is, either a offline or hot backup), you can create incremental backups. To do this, simply copy all of your currently existing log files to your backup location.
Incremental backups do not require you to run a checkpoint or to cease database write operations.
If your application uses the transactional bulk insert optimization, it is important to know that a database copy taken prior to a bulk loading event can no longer be used as the target of an incremental backup. This is true because bulk loading omits logging of some record insertions, so recovery cannot roll forward these insertions. It is recommended that a full backup be scheduled following a bulk loading event.
For more information, see the description of the DB_TXN_BULK
flag in the
DbEnv::txn_begin() API reference page.
When you are working with incremental backups, remember that the greater the number of log files contained in your backup, the longer recovery will take. You should run full backups on some interval, and then do incremental backups on a shorter interval. How frequently you need to run a full backup is determined by the rate at which your databases change and how sensitive your application is to lengthy recoveries (should one be required).
You can also shorten recovery time by running recovery against the backup as you take each incremental backup. Running recovery as you go means that there will be less work for DB to do if you should ever need to restore your environment from the backup.