There is a new voltadmin command to safely remove a single cluster from a multi-cluster XDCR environment without disrupting the connection between the remaining clusters. The DR DROP command tells a running cluster to send any queued binary logs, stop DR and inform the remaining clusters to remove it from the XDCR environment. See the description of the voltadmin utility in the Using VoltDB manual for details and an explanation of the difference between DR DROP and DR RESET.

The csvloader utility has a new option, --charset, that lets you specify the character set of the input file. See the description of the csvloader utility in the Using VoltDB manual for details.

A new column, QUEUE_GAP, has been added to the return values of the DRPRODUCER selector of the @Statistics system procedure. This column indicates if there is a gap between the last acknowledged transaction and the next transaction queued for transmission to the DR consumer. See the description of the @Statistics system procedure in the Using VoltDB manual for details.

Previously, the sqlcmd FILE directive let you process a single file containing SQL statements (and other sqlcmd directives). The directive has been extended to support a list of files, separated by spaces. If you include the -batch argument, all of the files are processed as a single batch. See the description of the sqlcmd utility in the Using VoltDB manual for details.

Java BigDecimal values are now allowable input to VoltDB integer datatypes, including TINYINT, SHORTINT, INTEGER, and BIGINT. As with any datatype conversion, if the BigDecimal value overflows the target integer type it will generate a runtime error. See the appendix on datatype compatibility in the Using VoltDB manual for details.

The --schema option to the voltdb init command lets you load a preliminary schema when initializing the database root. However, originally the schema file could only contain SQL DML statements. So certain files that were acceptable to sqlcmd were not acceptable to the voltdb init command because the files contained sqlcmd directives. The voltdb init command now ignores these directives and successfully processes the accompanying DML statements.

As part of an ongoing effort to optimize and reduce the impact of changes to the running database — specifically changes to the schema, stored procedures, or configuration — the internal code to implement these actions is being reorganized. One consequence of this change is that the name of internal procedures reported by the @Statistics system procedure may change. For example, previously, schema changes were reported by the @Statistics INITIATOR and PROCEDURE selectors as the UpdateApplicationCatalog procedure. Now these actions are executed by the UpdateCore procedure.

VoltDB uses tree indexes by default. It has been possible to request a hash index by including the string "hash" in the index name. However, testing shows that in most cases, tree indexes are more effective or at least as effective as hash indexes. Consequently, hash indexes are now deprecated and will be removed at an upcoming major release.

In addition to the new features and capabilities described above, the following limitations in previous versions have been resolved:

There is a new process available for upgrading VoltDB software on a running cluster using XDCR without requiring a second set of cluster hardware. Upgrading the software between two XDCR clusters is the easiest method. However, if you do not want or have the resources for a second separate cluster, the new process uses XDCR and K-safety to split an existing cluster into two and upgrade the two halves through XDCR. See the section on "Performing an Online Upgrade with Limited Hardware" in the VoltDB Administrator's Guide for details.

There is a new selector for the @Statistics system procedure, LATENCY, which provides periodic latency statistics. The statistics are generated every five seconds and provide both maximum and various levels of percentile measurement up to 99.999%. See the description of the @Statistics system procedure in the Using VoltDB manual for details.

Kerberos support has been extended to provide integration with the VoltDB command line utilities, such as sqlcmd and voltadmin. See the discussion of "Accessing the Database from the Command Line" in the Kerberos section of the Using VoltDB manual for details.

There is a new version of the kafkaloader utility, named kafkaloader10, that works with Kafka version 0.10.2. There are enough differences between Kafka 0.8.2 and 0.10.2 that two separate utilities will be maintained for the foreseeable future.

Customers interested in using kafkaloader10 are encouraged to try this beta release and provide us feedback. Note however, it has not yet received the rigorous testing of the default 0.8.2 version of kafkaloader and should not be considered production-ready quite yet. See the online help for kafkaloader10 in the bin directory for more information.

It is now possible to select one cluster in a multi-cluster XDCR configuration as the preferred host of the initial synchronizing snapshot when the current cluster joins the environment. You use the preferred-source attribute in the <connection> configuration to identify the cluster ID of the cluster that will send the snapshot. For example, the following configuration specifies that cluster ID 2 will send the snapshot when cluster ID 3 first joins the XDCR environment:

<dr id="3" role="xdcr" >
   <connection source="svra,svrb,svrc" preferred-source="2" />
</dr>

Note that preferred-source only affects the initial snapshot. All other DR communication occurs among all clusters within the XDCR environment.

There is a new tab available in the VoltDB Management Center that helps you monitor the progress of import connectors. The new Import tab is only visible when import connectors are configured and active.

The VoltDB Management Center (VMC) has been optimized to reduce its bandwidth use and thereby improve overall performance.

By default, VoltDB sends SQL statements through the JDBC export connector using all uppercase table and column names. However, a few databases (most notably, PostgreSQL) are case-sensitive. There is a new property for the JDBC connector that tells VoltDB to use lowercase table and column names. The lowercase property accepts either true or false, with false being the default. See the description of the JDBC export connector in the Using VoltDB manual for details.

There is a method in the VoltDB Java client API, updateApplicationCatalog(), designed for internal use based on features that are no longer available in the product. This method is now deprecated and be removed from the Java client library in the next major release of VoltDB.

In addition to the new features and capabilities described above, the following limitations in previous versions have been resolved:

It is now possible to add a starting schema and procedure classes when you initialize the database root directory. The new options --schema and --classes let you specify a file containing DDL commands and a JAR file of stored procedure classes, respectively. These items are loaded automatically the first time the database is started. (And subsequent starts, if command logging is not enabled.) For example, the following command pre-loads DDL commands from the file myschema.sql and a set of procedure classes from myprocs.jar:

$ voltdb init --config=deployment.xml \
              --schema=myschema.sql --classes=myprocs.jar

See the description of the voltdb init command in the Using VoltDB manual for details.

In multi-cluster XDCR environments, it is now possible to drop one cluster from the network without disrupting replication between the remaining clusters by using the voltadmin dr reset --cluster command. See the description of the voltadmin command in the Using VoltDB manual for details.

Use of the @SwapTables system procedure is now supported by database replication (DR). Swapping tables on a producer cluster acts like a schema change, causing the consumer cluster to pause until you execute the same @SwapTables statement on the consumer. See the description of the @SwapTables system procedure in the Using VoltDB manual for details.

Data manipulation language (DML) statements, such as INSERT, UPDATE, and DELETE, now support subqueries. See the description of the individual statements in the appendix of supported SQL statements in the Using VoltDB manual for details.

There is a new command, voltadmin status, that reports the current status of a running cluster. Information in the output includes the number and IP addresses of the servers in the cluster and if any nodes are missing from a K-safe cluster. You can also add the --dr option to include information on the current state of data replication for the cluster. See the description of the voltadmin command in the Using VoltDB manual for details.

The syntax for the voltdb collect command has been enhanced to be consistent with the other voltdb commands, such as get, init, and start. The voltdb collect command now let's you specify the location of the root directory using the --dir option, like other voltdb commands. It also has an --output option that lets you specify both the name and location of the output file. Note that, although deprecated, use of the old syntax will continue to be supported during the 7.x release cycle.

VoltDB now provides more thorough logging of security events. Specifically, every authorization failure is logged, each user who successfully authenticates is logged, and every configuration change is logged prior to the change being made.

A new chapter on writing custom exporters, importers and formatters has been added to the VoltDB Guide to Performance and Customization.

In addition to the new features and capabilities described above, the following limitations in previous versions have been resolved:

VoltDB now supports Secure Socket Layer (SSL) encryption for all externally accessible ports, not just the HTTP port. With SSL enabled, data passing through both the client and admin ports is encrypted. If encryption is enabled for the client and admin ports, you must use the --ssl option on the VoltDB command line tools (such as sqlcmd, voltadmin, and voltdb) to connect to the database. The Java and C++ client libraries also now support connecting through SSL-enabled ports.

<httpd>
   <https enabled="true">
      [ . . . ]
   </https>
</httpd>

<ssl enabled="true">
   [ . . . ]
</ssl>

See the section on using SSL encryption in the VoltDB Administrator's Guide for details on setting up SSL for a VoltDB cluster.

Three new voltdb commands have been added that allow you to extract the schema, stored procedures, and configuration from the local database root directory. These commands can be used whether the database is running or not. The new commands are:

See the description of the voltdb command in the Using VoltDB manual for details.

A new selector, PROCEDUREDETAIL, for the @Statistics system procedure provides detailed performance information about the execution of individual statements within stored procedures. See the description of the @Statistics system procedure in the Using VoltDB manual for details.

A new system procedure, @SwapTables, lets you swap the contents of two tables. The two tables must have the identical schema (that is, the same column names, datatypes, and constraints) and cannot have views defined. See the description of the @SwapTables system procedure in the Using VoltDB manual for details.

It is now possible to restart a K-safe cluster with fewer than the original number of nodes. For example, if one server fails on a five node cluster and you want to restart the cluster but the failed server is not ready, you can start the cluster with the remaining four nodes by using the --missing argument on the voltdb start command, like so:

$ voltdb start --count=5 --missing=1

Note that the partial cluster is not fully K-safe to its specified K-safety value until the full complement of nodes rejoin the cluster. See the description of the voltdb command in the Using VoltDB manual for details.

There are four new SQL functions to convert internet addresses from text strings to binary and back again. Two functions (INET_ATON and INET_NTOA) convert IPv4 addresses and two (INET6_ATON and INET6_NTOA) convert IPv6 addresses. See the Appendix of SQL Functions in the Using VoltDB manual for details.

Previously, the information gathered by the @Statistics system procedure related to schema changes (from DML statements or calls to @UpdateClasses or @UpdateApplicationCatalog) was reset each time the database schema changed. Schema changes no longer reset the statistics on each invocation. Also, every schema change is recorded, rather than the 5% sampling done for other procedures.

VoltDB can send SNMP traps for selected database events, such as node failures or when resource limits are reached. Starting with V7.1, it also reports database replication (DR) events when DR is enabled.

We continue to extend support for the use of indexes to optimize the execution of SQL statements. Two new uses of indexes are when evaluating:

Three new features have been added to the kafkaloader utility:

See the online help for the kafkaloader utility for details.

Previously, updating the stored procedure classes (using the sqlcmd LOAD CLASSES directive or the @UpdateClasses system procedure) updated all of the classes in the schema. As a result, the more classes in the schema, the longer the update would take, even if the update itself was small. VoltDB now updates only those classes affected by the change, significantly reducing the time required for the update to complete.

You can create a DR environment using clusters running different versions of VoltDB. However, to support new functionality (in particular, multi-cluster XDCR) the DR protocol changed in V7.0. That major version acts as a gateway between earlier and later releases. In other words, you can run a V7.0 cluster with earlier versions, or you can run a 7.0 cluster with later versions. However, you cannot run pre-V7.0 clusters with post-V7.0 clusters.

As a consequence, if you want to perform an in-service upgrade (using different version clusters) to upgrade a DR environment using a pre-V7.0 version (say, V6.9) to V7.1 or later, you must upgrade all clusters to V7.0 first, then upgrade to your final target version. Of course, you can always use the normal single-cluster upgrade process to go directly from V6.x to V7.x if you choose.

To ensure the security of VoltDB servers, the database no longer accepts SHA-1 for hashed data. All hashed data (such as passwords etc) must be hashed using SHA-256. Note: this is an incompatible change. Some earlier client libraries will not be able to connect to the V7.1 servers because they do not support the more robust SHA-256 hashing.

In addition to the new features and capabilities described above, the following limitations in previous versions have been resolved:

The following limitations in previous versions have been resolved:

The operating system and software requirements for VoltDB have been updated based on changes to the supported versions of the underlying technologies. Specifically:

As mentioned earlier, the addition of the role attribute to the <dr> element in the configuration file eliminates the need for the SET DR=ACTIVE SQL statement and the --replica argument to the voltdb start command. In addition, two columns for the local and remote cluster IDs have been added to the DRPRODUCER and DRCONSUMER selectors for the @Statistics system procedure. The new columns make it easier to identify to which DR connection the current row of data applies. However, their presence also changes the order of columns from previous releases. If an application used the DRPRODUCER or DRCONSUMER selectors and retrieves column values by index instead of by name, the application will need to be modified to accommodate the new columns in the results.

Previously, VoltDB allowed any 8-byte integer as input to a TIMESTAMP column. However, the TIMESTAMP functions limit the values to those defined by the MIN_VALID_TIMESTAMP() and MAX_VALID_TIMESTAMP() functions. Starting with V7.0, VoltDB no longer accepts values outside of the valid timestamp range for TIMESTAMP columns.

Note that this is an incompatible change and may require changes to applications that are inadvertently inserting invalid timestamp values. If so, you can use the IS_VALID_TIMESTAMP() function to validate input values before inserting them into TIMESTAMP columns.

In addition to the new features and capabilities described above, the following limitations in previous versions have been resolved:

VoltDB reserves the subfolder "segments" under the command log directory for storing the actual command log files. Do not add, remove, or modify any files in this directory. In particular, do not set the command log snapshot directory to a subfolder "segments" of the command log directory, or else the server will hang on startup.

Because DR data is buffered on the master database and then delivered asynchronously to the replica, there is always the danger that data does not reach the replica if a master node stops. This situation is mitigated in a K-safe environment by all copies of a partition buffering on the master cluster. Then if a sending node goes down, another node on the master database can take over sending logs to the replica. However, if multiple nodes go down and rejoin in rapid succession, it is possible that some buffered DR data — from transactions when one or more nodes were down — could be lost when another node with the last copy of that buffer also goes down.

If this occurs and the replica recognizes that some binary logs are missing, DR stops and must be restarted.

To avoid this situation, especially when cycling through nodes for maintenance purposes, the key is to ensure that all buffered DR data is transmitted before stopping the next node in the cycle. You can do this using the @Statistics system procedure to make sure the last ACKed timestamp (using @Statistitcs DR on the master cluster) is later than the timestamp when the previous node completed its rejoin operation.

Bulk operations, such as large deletes, inserts, or updates are possible within a single stored procedure. However, if the binary logs generated for DR are larger than 45MB, the operation will fail. To avoid this situation, it is best to break up large bulk operations into multiple, smaller transactions. A general rule of thumb is to multiply the size of the table schema by the number of affected rows. For deletes and inserts, this value should be under 45MB to avoid exceeding the DR binary log size limit. For updates, this number should be under 22.5MB (because the binary log contains both the starting and ending row values for updates).

There are a number of VoltDB features that help manage the database by constraining memory size and resource utilization. These features are extremely useful in avoiding crashes as a result of unexpected or unconstrained growth. However, these features could interfere with the normal operation of DR when passing data from one cluster to another, especially if the two clusters are different sizes. Therefore, as a general rule of thumb, DR overrides these features in favor of maintaining synchronization between the two clusters.

Specifically, DR ignores any resource monitor limits defined in the deployment file when applying binary logs on the consumer cluster. DR also ignores any partition row limits defined in the database schema when applying binary logs. This means, for example, if the replica database in passive DR has less memory or fewer unique partitions than the master, it is possible that applying binary logs of transactions that succeeded on the master could cause the replica to run out of memory. Note that these resource monitor and tables row limits are applied on any original transactions local to the cluster (for example, transactions on the master database in passive DR).

Database Replication (DR) now supports replication across clusters of different sizes. However, if the replica cluster is smaller than the master cluster, it may require a significantly larger Java heap setting. Specifically, if the replica has fewer unique partitions than the master, each partition on the replica must manage the incoming binary logs from more partitions on the master, which places additional pressure on the Java heap.

A simple rule of thumb is that the worst case scenario could require an additional P * R * 20MB space in the Java heap , where P is the number of sites per host on the replica server and R is the ratio of unique partitions on the master to partitions on the replica. For example, if the master cluster is 5 nodes with 10 sites per host and a K factor of 1 (i.e. 25 unique partitions) and the replica cluster is 3 nodes with 8 sites per host and a K factor of 1 (12 unique partitions), the Java heap on the replica cluster may require approximately 320MB of additional space in the heap:

An alternative is to reduce the size of the DR buffers on the master cluster by setting the DR_MEM_LIMIT Java property. For example, you can reduce the DR buffer size from the default 10MB to 5MB using the VOLTDB_OPTS environment variable before starting the master cluster.

Changing the DR buffer limit on the master from 10MB to 5MB proportionally reduces the additional heap size needed. So in the previous example, the additional heap on the replica is reduced from 320MB to 160MB.

Part of the replication process for XDCR is to verify that the record's starting and ending states match on both clusters, otherwise known as conflict resolution. To do that, XDCR must find the record first. Finding uniquely indexed records is efficient; finding non-unique records is not and can impact overall database performance.

To make you aware of possible performance impact, VoltDB issues a warning if you declare a table as a DR table and it does not have a unique index.

You cannot start XDCR if both databases already have data in the DR tables. Only one of the two participating databases can have preexisting data when DR starts for the first time.

When starting XDCR for the first time, where one database already contains data, a snapshot of that data is sent to the other database. While receiving and processing that snapshot, the receiving database is paused. That is, it is in read-only mode. Once the snapshot is completed and the two database are synchronized, the receiving database is automatically unpaused, resuming normal read/write operations.

Normally, XDCR will automatically restart where it left off after one of the clusters stops and recovers from its command logs (using the voltdb recover command). However, if the workload is predominantly multi-partition write transactions, a failed cluster may not be able to restart XDCR after it recovers. In this case, XDCR must be restarted from scratch, using the content from one of the clusters as the source for synchronizing and recreating the other cluster (using the voltdb create --force command) without any content in the DR tables.

When using XDCR, if the binary log from one cluster includes a TRUNCATE TABLE statement and the other cluster performs any write operation to the same table before the binary log is processed, the TRUNCATE TABLE operation will be reported as a conflict. Note that currently DELETE operations always supercede other actions, so the TRUNCATE TABLE will be executed on both clusters.

It is possible to place a limit on the number of rows that any partition can hold for a specific table using the LIMIT PARTITION ROWS clause of the CREATE TABLE statement. When close to the limit, transactions on either or both clusters can exceed the limit simultaneously, resulting in a potentially large number of delete operations that then generate conflicts when the the associated binary log reaches the other cluster.

VoltDB provides a way to generate a deterministically unique ID within a stored procedure using the getUniqueId method. This method guarantees uniqueness within the current cluster. However, the method could generate the same ID on two distinct database clusters. Consequently, when using XDCR, you should combine the return values of VoltProcedure.getUniqueId with VoltProcedure.getClusterId, which returns the current cluster's unique DR ID, to generate IDs that are unique across all clusters in your environment.

You cannot use cross-datacenter replication (XDCR) with the deprecated export syntax, that is the EXPORT TABLE statement. To use XDCR with export, you must use the current CREATE STREAM syntax for declaring the source streams for export targets.

A bug in the Apache Kafka client can result in file descriptors being allocated but not released if the producer.type attribute is set to "sync" (which is the default). The consequence is that the system eventually runs out of file descriptors and the VoltDB server process will crash.

Until this bug is fixed, use of synchronous Kafka export is not recommended. The workaround is to set the Kafka producer.type attribute to "async" using the VoltDB export properties.

If, while Kafka import is enabled, the Kafka broker that VoltDB is connected to stops (for example, if the server crashes or is taken down for maintenance), some messages may be lost between Kafka and VoltDB. To ensure no data is lost, we recommend you disable VoltDB import before taking down the associated Kafka broker. You can then re-enable import after the Kafka broker comes back online.

There is an issue with the Kafka importer where, if multiple nodes in the cluster fail and restart, the importer can lose track of some of the data that was being processed when the nodes failed. Normally, these pending imports are replayed properly on restart. But if multiple nodes fail, it is possible for some in-flight imports to get lost. This issue will be addressed in an upcoming release.

When entering a multi-line statement at the sqlcmd prompt, if a line ends in a comment (indicated by two hyphens) and the comment contains an unmatched single quote character, the following lines of input are not interpreted correctly. Specifically, the comment is incorrectly interpreted as continuing until the next single quote character or a closing semi-colon is read. This is most likely to happen when reading in a schema file containing comments. This issue is specific to the sqlcmd utility.

A fix for this condition is planned for an upcoming point release

VoltDB currently intercepts assertions as part of its handling of stored procedures. Attempts to use assertions in stored procedures for debugging or to find programmatic errors will not work as expected.

The UPPER() and LOWER() functions return a string converted to all uppercase or all lowercase letters, respectively. However, for the initial release, these functions only operate on characters in the ASCII character set. Other case-sensitive UTF-8 characters in the string are returned unchanged. Support for all case-sensitive UTF-8 characters will be included in a future release.

There is a problem with how the math library used to build the C++ client library handles large decimal values on 32-bit operating systems. As a result, the C++ library cannot serialize and pass Decimal datatypes reliably on these systems.

Note that the C++ client interface can send and receive Decimal values properly on 64-bit platforms.

Enabling SNMP can take up to 2 minutes to complete. This delay does not always occur and can vary in length. If SNMP is enabled when the database server starts, the delay occurs after the server logs the message "Initializing SNMP" and before it attempts to connect to the cluster. If you enable SNMP while the database is running, the delay can occur when you issue the voltadmin update command or modify the setting in the VoltDB Management Center Admin tab. This issue results from a Java constraint related to secure random numbers used by the SNMP library.

With VoltDB V7.0, cross-datacenter replication (XDCR) supports multiple clusters in an XDCR network. However, the VoltDB Management Center currently reports on only one such connection per cluster. In the future, the Management Center will provide monitoring and statistics for all connections to the current cluster.

Any time the database schema or stored procedures are changed, the data table showing stored procedure statistics at the bottom of the Monitor section of the VoltDB Management Center get reset. As soon as new invocations of the stored procedures occur, the statistics table will show new values based on performance after the schema update. Until invocations occur, the procedure table is blank.

The ASSUMEUNIQUE attribute is designed for identifying columns in partitioned tables where the column values are known to be unique but the table is not partitioned on that column, so VoltDB cannot verify complete uniqueness across the database. Using interactive DDL, you can create a table with a column marked as ASSUMEUNIQUE, but if you try to partition the table on the ASSUMEUNIQUE column, you receive an error. The solution is to drop and add the column using the UNIQUE attribute instead of ASSUMEUNIQUE.

You cannot add or remove a column constraint such as UNIQUE or ASSUMEUNIQUE using the ALTER TABLE ALTER COLUMN statement. Instead to add or remove such constraints, you must first drop then add the modified column. For example:

ALTER TABLE employee DROP COLUMN empID;
ALTER TABLE employee ADD COLUMN empID INTEGER UNIQUE;

For partitioned tables, the value of the column used to partition the table determines what partition the row belongs to. If you use UPDATE to change this value and the new value belongs in a different partition, the UPDATE request will fail and the stored procedure will be rolled back.

Updating the partition column value may or may not cause the record to be repartitioned (depending on the old and new values). However, since you cannot determine if the update will succeed or fail, you should not use UPDATE to change the value of partitioning columns.

The workaround, if you must change the value of the partitioning column, is to use both a DELETE and an INSERT statement to explicitly remove and then re-insert the desired rows.

If you refer to a table or column name that does not exist, VoltDB reports that the "user lacks privilege or object not found". This can happen, for example, if you misspell a table or column name.

Another situation where this occurs is if you mistakenly use double quotation marks to enclose a string literal (such as WHERE ColumnA="True"). ANSI SQL requires single quotes for string literals and reserves double quotes for object names. In the preceding example, VoltDB interprets "True" as an object name, cannot resolve it, and issues the "user lacks privilege" error.

The workaround is, if you receive this error, to look for misspelled table or columns names or string literals delimited by double quotes in the offending SQL statement.

Starting with VoltDB 6.0, ambiguous column references are no longer allowed. For example, if both the Customer and Placedorder tables have a column named Address, the reference to Address in the following SELECT statement is ambiguous:

SELECT OrderNumber, Address FROM Customer, Placedorder
   . . .

Previously, VoltDB would select the column from the leftmost table (Customer, in this case). Ambiguous column references are no longer allowed and you must use table prefixes to disambiguate identical column names. For example, specifying the column in the preceding statement as Customer.Address.

A corollary to this change is that a column declared in a USING clause can now be referenced using a prefix. For example, the following statement uses the prefix Customer.Address to disambiguate the column selection from a possibly similarly named column belonging to the Supplier table:

SELECT OrderNumber, Vendor, Customer.Address
   FROM Customer, Placedorder Using (Address), Supplier
    . . .

VoltDB opens a file descriptor for every client connection to the database. In normal operation, this use of file descriptors is transparent to the user. However, if there are an inordinate number of concurrent client connections, or clients open and close many connections in rapid succession, it is possible for VoltDB to exceed the process limit on file descriptors. When this happens, new connections may be rejected or other disk-based activities (such as snapshotting) may be disrupted.

In environments where there are likely to be an extremely large number of connections, you should consider increasing the operating system's per-process limit on file descriptors.