Import/Export Individual Graphs

The GSQL EXPORT GRAPH and IMPORT GRAPH commands perform a logical backup and restore of selected individual graphs specified by the user in a GraphNameList. A graph export contains the graph’s data, and optionally some types of metadata. This data can be subsequently imported in order to recreate the same graph data, in the original or in a different TigerGraph platform instance.

To see how to Export/Inport all graphs in a database at once refer to Database Import/Export All Graphs.

Beginning with version 4.1, a full export package now includes access policies and template queries.

Import/export is a complement to Backup and Restore, but not a substitute.

Import Prerequisite

To import an exported graph, ensure that the export files are from a database that was running the exact same version of TigerGraph as the database that you are importing into.

Export Selected Graphs

The EXPORT GRAPH command reads the data and metadata for all indvdually selected graphs in the GraphNameList and writes the information to a .zip file in the designated folder. It inherits Export All options but has the following differences.

If the user provides no options to the command, it will export:

  • Data

  • Schema Change Jobs

  • Loading Jobs

  • Vertex/Edge Types

  • Queries

  • Indices

  • UDFs

  • All tuples & typedef accumulators

  • And other features that belong to the specified graph

It will NOT export:

  • Packages

  • Additionally, by default, it will NOT export, unless specified option --user:

    • Users

    • Roles

    • Secrets

    • And tokens

Required privilege

EXPORT_GRAPH

Example

EXPORT GRAPH (ALL|GraphNameList) to "/path/to/export"

GraphNameList

GraphNameList is a list of graph names. The separator is comma , . ( e.g., graphName1, graphName1,graphName2 )

Synopsis

EXPORT GRAPH (ALL|GraphNameList) [<export_options>] TO "<directory_name>"

exportOptions ::=
(-S | --SCHEMA | -T | --TEMPLATE | -D | --DATA | -U | --USERS | -P | --PASSWORD pwd)


    -S, --SCHEMA        Only Schema will be exported
    -T, --TEMPLATE      Only Schema, Queries, Loading Jobs, UDFs
    -D, --DATA          Only Data Sources will be exported
    -U, --USERS         Includes Permissions, Secrets, and Tokens
    -P, --PASSWORD      Encrypt with password. User will be prompted
        --SEPARATOR     Specify a column separator
        --EOL           Specify a line separator

Output

The export output contains the following categories of files:

  • Data files in CSV format, one file for each type of vertex and each type of edge in a single node.

    However, for cluster environments, since the data files are spread across the nodes, they will not be included in the zip file.

  • GSQL DDL command files created by the export command. The import command uses these files to recreate the graph schema(s) and reload the data.

  • Copies of the database’s queries, loading jobs, and user-defined functions.

  • GSQL command files used to recreate the users and their privileges.

The following files are created in the specified directory when exporting and are then zipped into a single file named ExportedGraph.zip.

If the file is password-protected, it can only be unzipped using the GSQL command IMPORT GRAPH. The security features prevent users from directly unzipping it.

  • A DBImportExport_<Graph_Name>.gsql for each graph called Graph_Name in a multigraph database that contains a series of GSQL DDL statements that do the following:

    • Create the exported graph, along with its local vertex, edge, and tuple types,

    • Create the loading jobs from the exported graphs

    • Create data source file objects

    • Create queries

  • A graph_<Graph_Name>/ folder for each graph in a multigraph database containing data for local vertex/edge types in <Graph_Name>. For each vertex or edge type called <type>, there is one of the following two data files:

    • vertex_<type>.csv

    • edge_<type>.csv

  • global.gsql - DDL job to create all global vertex and edge types, data sources, packages, template queries and functions.

  • tuple.gsql - DDL job to create all User Defined Tuples.

  • policies.gsql - DDL job to create policies.

  • Exported data and jobs used to restore the data:

    • GlobalTypes/ - folder containing data for global vertex/edge types

      • vertex_name.csv

      • edge_name.csv

    • run_loading_jobs.gsql - DDL created by the export command which will be used during import:

      • Temporary global schema change job to add user-defined indexes. This schema job is dropped after it has run.

      • Loading jobs to load data for global and local vertex/edges.

  • Database’s saved queries, loading jobs, and schema change jobs

    • SchemaChangeJob/ -* folder containing DDL for schema change jobs. See section "Schema Change Jobs" for more information

      • Global_Schema_Change_Jobs.gsql contains all global schema change jobs

      • Graph_Name_Schema_Change_Jobs.gsql contains schema change jobs for each graph Graph_Name

  • User-defined functions

    • Tokenbank.cpp - copy of <tigergraph.root.dir>/app/<VERSION_NUM>/dev/gdk/gsql/src/TokenBank/TokenBank.cpp

    • ExprFunctions.hpp - copy of <tigergraph.root.dir>/app/<VERSION_NUM>dev/gdk/gsql/src/QueryUdf/ExprFunctions.hpp

    • ExprUtil.hpp - copy of <tigergraph.root.dir>/app/<VERSION_NUM>/dev/gdk/gsql/src/QueryUdf/ExprUtil.hpp

  • Users (Only if specified with --users):

    • users.gsql - DDL to create all exported users, import Secrets and Tokens and grant permissions.

Insufficient disk space

If not enough disk space is available for the data to be exported, the system returns an error message indicating not all data has been exported. Some data may have already been written to disk. If an insufficient disk error occurs, the files will not be zipped, due to the possibility of corrupted data which would then corrupt the zip file. The user should clear enough disk space, including deleting the partially exported data, before reattempting the export.

It is possible for all the files to be written to disk and then to run out of disk space during the zip operation. If that is the case, the system will report this error. The unzipped files will be present in the specified export directory.

Export timeout

If the timeout limit is reached during export, the system returns an error message indicating not all data has been exported. Some data may have already been written to disk. If a timeout error occurs, the files will not be zipped. The user should delete the export files, increase the timeout limit and then rerun the export.

The timeout limit is controlled by the session parameter export_timeout. The default timeout is ~138 hours. To change the timeout limit, use the command:

SET EXPORT_TIMEOUT = <timeout_in_ms>

Import Selected Graphs

The IMPORT GRAPH command unzips the .zip file ExportedGraph.zip located in the designated folder, and then runs the GSQL command files.

IMPORT GRAPH does not execute like drop all as with IMPORT GRAPH ALL before importing specified graphs.

Instead they import with these conditons:

  • If there are global level any VERTEX/EDGE, and the importing tarball also contains the VERTEX/EDGE type with the same name 2 things could happen:

    1. If the attributes between them are different, the import process will be aborted, and report error.

    2. If the signature (attributeName, attributeType, primary_id_as_attribute, primaryKey, compositeKeys, edgeDiscriminator, or UDT types) are exactly same, the VERTEX/EDGE creation process will be skipped, and continue.

  • If a graph name specified in the arguments is not found within the tarball, an error will be thrown, and the import statement does nothing.

  • If a graph name specified in the arguments is already existing in current schema, the importing process will be aborted.

  • The UDFs will be merged automatically for [tg_]ExprFunctions.hpp if possible, else an error will be reported.

  • For [tg_]ExprUtils.hpp & TokenBank.hpp, a warning will be reported if the MD5sum are different between importing one and existing one.

  • If GraphNameList --user is provided, the existent users will be retained. If users are exported, the users will be imported as well.

Please be extra cautious when importing databases as it can overwrite the current solution, resulting in the deletion of existing schemas, load jobs, queries, and data files. Importing a new solution cannot be undone to restore the previous state, regardless of whether the import succeeds or fails.

Therefore, create a complete backup beforehand in case you need to restore the database: Back up a Database Cluster

For security purposes, TigerGraph has two gadmin commands, GSQL.UDF.Policy.Enable and GSQL.UDF.Policy.HeaderAllowlist to prevent malicious code execution during import. Please refer to the section on UDF Security to ensure that UDFs comply with the security specifications. This will help you import the solution successfully.

Required privileges

WRITE_SCHEMA, CREATE_QUERY, WRITE_LOADINGJOB, EXECUTE_LOADINGJOB, DROP ALL, WRITE_USERS

Example

IMPORT GRAPH (ALL|GraphNameList) from "/path/to/exported/data"

Synopsis

IMPORT GRAPH (ALL|GraphNameList) [import_options] FROM "<filename>"

importOptions ::= [-P | --PASSWORD ] [ (-KU | -- keep-users]
-P,  --PASSWORD     Decrypt with password. User will be prompted.
-KU, --KEEP-USERS   Do not delete user identities before importing

Loading Jobs

There are two sets of loading jobs:

  • Those that were in the catalog of the database which was exported. These are embedded in the file DBImportExport_Graph_Name.gsql

  • Those that are created by EXPORT GRAPH and are used to assist with the import process. These are embedded in the file run_loading_jobs.gsql.

The catalog loading jobs are not needed to restore the data. They are included for archival purposes.

Some special rules apply to importing loading jobs. Some catalog loading jobs will not be imported.

  1. If a catalog loading job contains DEFINE FILENAME F = "/path/to/file/", the path will be removed and the imported loading job will only contain DEFINE FILENAME F. This is to allow a loading job to still be imported even though the file may no longer exist or the path may be different due to moving to another TigerGraph instance.

  2. If a specific file path is used directly in the LOAD statement, and the file cannot be found, the loading job cannot be created and will be skipped. For example, LOAD "/path/to/file" to vertex v1 cannot be created if /path/to/file does not exist.

  3. Any file path using $sys.data_root will be skipped. This is because the value of $sys.data_root is not retained from an export. During an import, $sys.data_root is set to the root folder of the import location.

Schema Change Jobs

There are two sets of schema change jobs:

  1. Those that were in the catalog of the database which was exported. These are stored in the folder /SchemaChangeJobs.

  2. Those that were created by EXPORT GRAPH and are used to assist with the import process. These are in the run_loading_jobs.gsql command file. The jobs are dropped after the import command is finished with them.

The database’s schema change jobs are not executed during the import process. This is because if a schema change job had been run before the export, then the exported schema already reflects the result of the schema change job. The directory /SchemaChangeJobs contains these files:

  • Global_Schema_Change_Jobs.gsql contains all global schema change jobs

  • <Graph_Name>_Schema_Change_Jobs.gsql contains schema change jobs for each graph <Graph_Name>.

Known Issues and Workarounds

[tg_]ExprFunction.hpp

[tg_]ExprFunction.hpp will be automatically merged while importing single graphs. In some cases below, query compilation may fail if one of the below issues are not solved. There are several known issues:

  1. If the [tg_]ExprUtil.hpp or TokenBank.cpp are different from the existing ones, users will get a warning message and a workaround will be shown as below:

    Importing UDF Functions
Failed to merge ExprFunctions automatically.
Warning: The import file /TIGERGRAPH/tmp/import_UDF_automatically_merge/exported/ExprFunctions.hpp is inconsistent with the existing file ExprFunctions.hpp.
Manually fix should be done by following commands:
1. Run commands: GSQL 'GET ExprFunctions TO "/path/to/store/file"'
2. Cherry-pick the changes from /TIGERGRAPH/tmp/import_UDF_automatically_merge/exported/ExprFunctions.hpp to "/path/to/store/file"
3. Run commands: GSQL 'PUT ExprFunctions FROM "/path/to/store/file"'
PUT ExprFunctions skipped.

  2. Inline variables, including headers, namespace, macro should be handled manually. For example, the below statements will not be auto-merged and moving the below component to an existing UDF as in the steps above, is the workaround.

    #include <stdio.h>
#define MACRO 1
using namespace std;
int a = 1;
typedef std::string string;