Incompatible changes in ArangoDB 3.9

It is recommended to check the following list of incompatible changes before upgrading to ArangoDB 3.9, and adjust any client programs if necessary.

The following incompatible changes have been made in ArangoDB 3.9:

Architecture requirements

The minimum architecture requirements have been raised from the Westmere architecture to the Sandy Bridge architecture. 256 bit AVX instructions are now expected to be present on all targets that run ArangoDB 3.9 executables. If a target does not support AVX instructions, it may fail with SIGILL at runtime.

Deployment mode “leader-follower” no longer supported

The Leader/Follower deployment mode in which two single servers were set up as a leader and follower pair (without any kind of automatic failover) was deprecated and removed from the documentation.

Recommended alternatives are the Active Failover deployment option and the OneShard feature in a cluster.

Extended naming convention for databases

There is a new startup option allowing database names to contain most UTF-8 characters. The option name is --database.extended-names-databases.

The feature is disabled by default to ensure compatibility with existing client drivers and applications that only support ASCII names according to the traditional database naming convention used in previous ArangoDB versions.

If the feature is enabled, then any endpoints that contain database names in the URL may contain special characters that were previously not allowed (percent-encoded). They are also to be expected in payloads that contain database names. If client applications assemble URLs with database names programmatically, they need to ensure that database names are properly URL-encoded and also NFC-normalized if they contain UTF-8 characters.

The ArangoDB client tools arangobench, arangodump, arangoexport, arangoimport, arangorestore, and arangosh ship with full support for the extended database naming convention.

Please be aware that dumps containing extended database names cannot be restored into older versions that only support the traditional naming convention. In a cluster setup, it is required to use the same database naming convention for all Coordinators and DB-Servers of the cluster. Otherwise the startup will be refused. In DC2DC setups it is also required to use the same database naming convention for both datacenters to avoid incompatibilities.

Also see Database Naming Conventions.


The following complexity limits have been added in 3.9 for AQL queries, Additional complexity limits have been added for AQL queries, in order to prevent programmatically generated large queries from causing trouble (too deep recursion, enormous memory usage, long query optimization and distribution passes etc.).

The following limits have been added:

  • a recursion limit for AQL query expressions. An expression can now be up to 500 levels deep. An example expression is 1 + 2 + 3 + 4, which is 3 levels deep 1 + (2 + (3 + 4)). The recursion of expressions is limited to 500 levels.
  • a limit for the number of execution nodes in the initial query execution plan. The number of execution nodes is limited to 4000. This number includes all execution nodes of the initial execution plan, even if some of them could be optimized away later by the query optimizer during plan optimization.

AQL queries that violate these limits will fail to run, and instead abort with error 1524 (“too much nesting or too many objects”) during setup.

Also see Known limitations for AQL queries.

Startup options

Installing Foxx apps from remote URLS

The --foxx.allow-install-from-remote option controls whether installing Foxx apps from remote URL sources other than Github is allowed. If set to false, installing Foxx apps is blocked for any remote sources other than Github. Installing Foxx apps from Github or from uploaded zip files is still possible with this option. Setting it to true will allow installing Foxx apps from any remote URL sources.

In ArangoDB 3.9, the default value for this option is false, meaning that installing Foxx apps from remote sources other than Github is now disallowed. This also inactivates the Remote tab in the Services section of the web interface. Compared to the previous versions of ArangoDB, this is a downwards-incompatible default value change, which was made for security reasons. To enable installing apps from remote sources again, set this option to true.

RocksDB options

The default value for the startup --rocksdb.max-subcompactions option was changed from 1 to 2. This allows compactions jobs to be broken up into disjoint ranges which can be processed in parallel by multiple threads.

Rebalance shards

The new --cluster.max-number-of-move-shards option limits the maximum number of move shards operations that can be made when the Rebalance Shards button is clicked in the Web UI. For backwards compatibility purposes, the default value is 10. If the value is 0, the tab containing the button is not clickable.

Timeout for JWT sessions

The lifetime for tokens that can be obtained from the POST /_open/auth endpoint is now configurable via the --server.session-timeout startup option.

The value for the option can be specified in seconds. The default timeout is one hour. Previous versions of ArangoDB had a longer, hard-coded timeout.

The web interface uses JWT for authentication. However, the session will be renewed automatically as long as you regularly interact with the Web UI in your browser. You will not get logged out while actively using it.

Disallowed usage of collection names in AQL expressions

The startup option --query.allow-collections-in-expressions added in 3.8.0 controls whether collection names are allowed in arbitrary places in AQL expressions. The default was true, but is now changed to false in 3.9.0 to make queries like FOR doc IN collection RETURN collection fail, where it was probably intended to RETURN doc instead. Also see ArangoDB Server Query Options

Such unintentional usage of collection names in queries now makes the query fail with error 1568 (“collection used as expression operand”) by default. The option can be set to true to restore the 3.8 behavior. However, the option is deprecated from 3.9.0 on and will be removed in future versions. From then on, unintended usage of collection names will always be disallowed.

If you use queries like RETURN collection then you should replace them with FOR doc IN collection RETURN doc to ensure future compatibility.

Cluster-internal network protocol

The cluster-internal network protocol is hard-coded to HTTP/1 in ArangoDB 3.9. Any other protocol selected via the startup option --network.protocol will automatically be switched to HTTP/1. The startup option --network.protocol is now deprecated and hidden by default. It will be removed in a future version.

“Old” system collections

The option --database.old-system-collections was introduced in 3.6 and 3.7 to control if the obsolete system collections _modules and _routing should be created with every new database or not.

The option was introduced with a default value of true in 3.6 and 3.7 for downwards-compatibility reasons. With the introduction of the option, it was also announced that the default value would change to false in ArangoDB 3.8. This has happened, meaning that 3.8 installations by default will not create these system collections anymore.

In ArangoDB 3.9 the option --database.old-system-collections is now completely obsolete, and ArangoDB will never create these system collections for any new databases. The option can still be specified at startup, but it meaningless now.

Client tools

General changes

The default value for the --threads startup parameter was changed from 2 to the maximum of 2 and the number of available CPU cores for the following client tools:

  • arangodump
  • arangoimport
  • arangorestore

This change can help to improve performance of imports, dumps or restore processes on machines with multiple cores in case the --threads parameter was not previously used. As a trade-off, the change may lead to an increased load on servers, so any scripted imports, dumps or restore processes that want to keep the server load under control should set the number of client threads explicitly when invoking any of the above client tools.


The default value of arangodump’s --envelope option changes from true in 3.8 to false in 3.9. This change turns on the non-envelope dump format by default, which will lead to smaller and slightly faster dumps. In addition, the non-enveloped format allows higher parallelism when restoring dumps with arangorestore.

The non-enveloped dump format is different to the enveloped dump format used by default in previous versions of ArangoDB.

In the enveloped format, dumps were JSONL files with a JSON object in each line, and the actual database documents were placed inside a data attribute. There was also a type attribute for each line, which designated the type of object in that line (typically this will have been type "2300", meaning “document”). The old, enveloped format looks like this:

{"type":2300,"key":"test","data":{"_key":"test","_rev":..., ...}}

The non-enveloped format which is now enabled by default only contains the actual documents, e.g.

{"_key":"test","_rev":..., ...}

The change of the default dump format may have an effect on third-party backup tools or script. arangorestore will work fine with both formats. To switch between the formats, arangodump provides the --envelope option.


With the default dump format changing from the enveloped variant to the non-enveloped variant, arangorestore will now by default be able to employ higher parallelism when restoring data of large collections.

When restoring a collection from a non-enveloped dump, arangorestore can send multiple batches of data for the collection in parallel if it can read the dump files faster than the server can respond to arangorestore’s requests. This increased parallelism normally helps to speed up the restore process, but it can also lead to arangorestore saturating the server with its restore requests. In this case it is advised to decrease the value of arangorestore’s --threads option accordingly. The value of --threads will the determine the maximum parallelism used by arangorestore.


ArangoDB release packages install an executable named arangoimp as an alias for the arangoimport executable. This is done to provide compatibility with older releases, in which arangoimport did not yet exist and was named arangoimp. The renaming was actually carried out in the codebase in December 2017. Using the arangoimp executable is now deprecated, and it is always favorable to use arangoimport instead. While the arangoimport executable will remain, the arangoimp alias will be removed in a future version of ArangoDB, and its use is now highly discouraged.


The former --json and --pretty options of the arangovpack utility were removed and replaced with separate options for specifying the input and output types:

  • --input-type (json, json-hex, vpack, vpack-hex)
  • --output-type (json, json-pretty, vpack, vpack-hex)

The former --print-non-json option was replaced with the new --fail-on-non-json option, which makes arangovpack fail when trying to emit non-JSON types to JSON output.