ContextQMD
Back to Airbyte

Salesforce

docs/integrations/sources/salesforce.md

2.0.054.7 KB
Original Source

Salesforce

<HideInUI>

This page contains the setup guide and reference information for the Salesforce source connector.

</HideInUI>

Prerequisites

  • A Salesforce Account with Enterprise edition or Professional edition with API access purchased as an add-on

You may also need the following. This article explains how to get them.

<!-- env:oss -->
  • For Airbyte Open Source: Salesforce OAuth credentials
<!-- /env:oss -->

Setup guide

Set up Salesforce

Create a dedicated Salesforce user (optional, but recommended) {#dedicated-salesforce-user}

Follow the instructions below to create a Minimum Access standard profile and assign custom permission sets to grant the new user the read access needed for data you want to access with Airbyte.

While you can set up the Salesforce connector using any Salesforce user with read permission, we recommend creating a dedicated user with the Minimum Access standard profile for Airbyte. This allows you to granularly control the data Airbyte can read.

Using Permission Sets, you should grant this user read access to the data you want Airbyte to have access to. Learn more about Permission sets by referring to Salesforce's documentation. If you want to sync permissions data from Salesforce, you also need the permissions necessary to sync security-related data.

Log in to Salesforce with an admin account.

Step 1: Create a new user
  1. On the top right of the screen, click the gear icon and then click Setup.
  2. In the left navigation bar, under Administration, click Users > Users. Create a new User, entering details for the user's first name, last name, alias, and email. Filling in the email field auto-populates the username field and nickname.
    1. Leave role unspecified
    2. Select Salesforce for the User License
    3. Select Standard User for Profile.
    4. Decide whether to generate a new password and notify the user.
    5. Select save
Step 2: Create a new permission set
  1. Using the left navigation bar, select Users > Permission Sets
  2. Click New to create a new Permission Set.
  3. Give your permission set a descriptive label name (e.g., "Airbyte Read Only Access"). The API name auto-populates based on the label you give the permission set.
  4. For licence, leave this set to –None— and click save.
  5. Now that you see the permission set is created, define the permissions via Object Settings.
    1. Click "Object Settings."
    2. Select the Object Name for each object you want the user to have read-only access to (for example, Accounts, Contacts, Opportunities).
    3. Select “Edit” and check the "Read" permission and clear all other permissions (Create, Edit, Delete, etc.)
    4. Click Save
    5. Continue to add read permissions for any objects you want Airbyte to have access to.
  6. To grant access to uninstalled connected apps, you need to enable additional permission.
    1. Click "System Permissions"
    2. Select “Edit”
    3. If API Access Control is enabled, need to check the "Use Any API Client" permission. If API Access Control isn't enabled, need to check “Approve Uninstalled Connected Apps” permission.
    4. Click Save
Step 3: Assign the permission set to the new user
  1. From the Permission Sets page, click "Manage Assignments" next to the read-only permission set you just created.
  2. Click "Add Assignments."
  3. Find and select the user you created in Step 1.
  4. Click Assign

Log into the email you used above and verify your new Salesforce account user. You'll need to set a password as part of this process. Keep this password accessible.

:::info Profile vs. Permission Set: remember that the user's profile provides their baseline permissions. The permission set adds or restricts permissions on top of that. Object-Level vs. Field-Level Security: This guide focuses on object-level read-only access. While setting up your permission set, you can stick with object-level security or define more granular controls by scrolling down within each object settings page to select read access for only needed fields. :::

<!-- env:oss -->

Get Salesforce OAuth credentials (Airbyte Open Source only)

If you are using Airbyte Open Source, obtain the following OAuth credentials to authenticate:

  • Client ID
  • Client Secret
  • Refresh Token

To obtain these credentials, follow this walkthrough with the following modifications:

  1. If your Salesforce URL is not in the X.salesforce.com format, use your Salesforce domain name. For example, if your Salesforce URL is awesomecompany.force.com then use that instead of awesomecompany.salesforce.com.
  2. When running a curl command, run it with the -L option to follow any redirects.
  3. If you created a read-only user, use the user credentials when logging in to generate OAuth tokens.
<!-- /env:oss -->

Set up the Salesforce connector in Airbyte

<!-- env:cloud -->

For Airbyte Cloud:

  1. Log into your Airbyte Cloud account.
  2. Click Sources and then click + New source.
  3. On the Set up the source page, select Salesforce from the Source type dropdown.
  4. Enter a name for the Salesforce connector.
  5. To authenticate: For Airbyte Cloud: Click Authenticate your account to authorize your Salesforce account. Airbyte will authenticate the Salesforce account you are already logged in to. Please make sure you are logged into the right account.
  6. Toggle whether your Salesforce account is a Sandbox account or a production account.
  7. (Optional) For Start Date, use the provided datepicker or enter the date programmatically in either YYYY-MM-DD or YYYY-MM-DDTHH:MM:SSZ format. The data added on and after this date will be replicated. If this field is left blank, Airbyte will replicate the data for the last two years by default. Please note that timestamps are in UTC.
  8. (Optional) In the Filter Salesforce Object section, you may choose to target specific data for replication. To do so, click Add, then select the relevant criteria from the Search criteria dropdown. For Search value, add the search terms relevant to you. You may add multiple filters. If no filters are specified, Airbyte will replicate all data.
  9. (Optional) For Lookback Window, enter an ISO 8601 duration (e.g., PT10M, PT30M, PT1H) to control how far back the connector re-reads data on each incremental sync. The default is PT10M (10 minutes). Increase this value if you observe missing records in your destination, which can occur due to Salesforce API eventual consistency delays.
  10. Click Set up source and wait for the tests to complete.
<!-- /env:cloud --> <!-- env:oss -->

For Airbyte Open Source:

  1. Navigate to the Airbyte Open Source dashboard.
  2. In the left navigation bar, click Sources. In the top-right corner, click + New source.
  3. Find and select Salesforce from the list of available sources.
  4. Enter a Source name of your choosing to help you identify this source.
  5. To authenticate: For Airbyte Open Source: Enter your Client ID, Client Secret, and Refresh Token.
  6. Toggle whether your Salesforce account is a Sandbox account or a production account.
  7. (Optional) For Start Date, use the provided datepicker or enter the date programmatically in either YYYY-MM-DD or YYYY-MM-DDTHH:MM:SSZ format. The data added on and after this date will be replicated. If this field is left blank, Airbyte will replicate the data for the last two years by default. Please note that timestamps are in UTC.
  8. (Optional) In the Filter Salesforce Object section, you may choose to target specific data for replication. To do so, click Add, then select the relevant criteria from the Search criteria dropdown. For Search value, add the search terms relevant to you. You may add multiple filters. If no filters are specified, Airbyte will replicate all data.
  9. (Optional) For Lookback Window, enter an ISO 8601 duration (e.g., PT10M, PT30M, PT1H) to control how far back the connector re-reads data on each incremental sync. The default is PT10M (10 minutes). Increase this value if you observe missing records in your destination, which can occur due to Salesforce API eventual consistency delays.
  10. Click Set up source and wait for the tests to complete.
<!-- /env:oss --> <HideInUI>

Supported sync modes

The Salesforce source connector supports the following sync modes:

Supported Streams

The Salesforce connector supports reading both Standard Objects and Custom Objects from Salesforce. Each object is read as a separate stream. See a list of all Salesforce Standard Objects here.

Airbyte allows exporting all available Salesforce objects dynamically based on:

  • If the authenticated Salesforce user has the Role and Permissions to read and fetch objects. This would be set as part of the Permission Set you assign to the Airbyte user. See Create a dedicated Salesforce user for more information.
  • If the Salesforce object has the queryable property set to true. Airbyte can only fetch objects which are queryable. If you don’t see an object available via Airbyte, and it is queryable, check if it is API-accessible to the Salesforce user you authenticated with.

Syncing Permissions Data from Salesforce

The Salesforce connector can be used to sync security-related objects that can be synced to understand user permissions, roles, and access patterns within your Salesforce organization.

Common use cases for syncing Salesforce permission data include:

  • Permission Replication: Leverage permissions from the Salesforce source to guide application of permissions downstream.
  • Security Auditing: Generate reports on user access rights and permission assignments.

The following streams contain security and permission-related data:

  • User - Core user accounts with security-related fields including profiles, roles, and user permissions. Contains information about user status, login history, and assigned licenses.
  • ActivePermSetLicenseMetric - Tracks permission set license usage metrics including assigned user counts, active user counts, and total available licenses.
  • ActiveProfileMetric - Provides metrics about user profile usage including user license associations and assignment counts.

For comprehensive information about Salesforce security objects, refer to the Salesforce Object Reference documentation.

How Salesforce Permission Syncing Works

Salesforce provides some security-related data through its standard object model:

  1. Dynamic Object Discovery: The connector automatically discovers available Salesforce objects (sobjects) based on the authenticated user's permissions.
  2. Permission-Based Access: Which security objects are available depends on the permissions granted to the Salesforce user used for authentication and your Salesforce environment configuration.
  3. Standard Object Syncing: Available security-related objects are synced as regular Salesforce objects through the same sync mechanisms as other business data.

:::note Security-related object availability varies by Salesforce environment (production vs sandbox) and user permissions. Security objects can be large datasets in organizations with many users, so monitor your Salesforce API limits accordingly. :::

Permissions Needed to Sync Permissions Data {#permissions-to-sync-permissions}

To sync security-related data from Salesforce, the authenticated Salesforce user must have appropriate permissions to read security objects. Consider granting these permissions through a dedicated permission set:

  • "View All Users" - Required to access comprehensive User data.
  • Standard read permissions for the specific objects you want to sync.

For more information about Salesforce security and permissions, refer to the official Salesforce documentation on User Permissions and Permission Sets.

Limitations & Troubleshooting

<details> <summary> Expand to see details about Salesforce connector limitations and troubleshooting. </summary>

Connector limitations

Rate limiting

The Salesforce connector is restricted by Salesforce's Daily Rate Limits. The connector syncs data until it hits the daily rate limit, then ends the sync early with success status, and starts the next sync from where it left off. Note that picking up from where it ends will work only for incremental sync, which is why we recommend using the Incremental Sync - Append + Deduped sync mode.

Syncing Formula Fields

The Salesforce connector syncs formula field outputs from Salesforce. If the formula of a field changes in Salesforce and no other field on the record is updated, you will need to reset the stream and sync a historical backfill to pull in all the updated values of the field.

Syncing Deletes

The Salesforce connector supports retrieving deleted records from the Salesforce recycle bin. For the streams which support it, a deleted record will be marked with isDeleted=true. To find out more about how Salesforce manages records in the recycle bin, please visit their docs.

Usage of the BULK API vs REST API

Salesforce allows extracting data using either the BULK API or REST API. To achieve fast performance, Salesforce recommends using the BULK API for extracting larger amounts of data (more than 2,000 records). For this reason, the Salesforce connector uses the BULK API by default to extract any Salesforce objects, unless any of the following conditions are met:

  • The Salesforce object has columns which are unsupported by the BULK API, like columns with a base64 or complexvalue type
  • The Salesforce object is not supported by BULK API. In this case we sync the objects via the REST API which will occasionally cost more of your API quota. This includes the following objects:
    • AcceptedEventRelation
    • Attachment
    • CaseStatus
    • ContractStatus
    • DeclinedEventRelation
    • FieldSecurityClassification
    • KnowledgeArticle
    • KnowledgeArticleVersion
    • KnowledgeArticleVersionHistory
    • KnowledgeArticleViewStat
    • KnowledgeArticleVoteStat
    • OrderStatus
    • PartnerRole
    • RecentlyViewed
    • ServiceAppointmentStatus
    • ShiftStatus
    • SolutionStatus
    • TaskPriority
    • TaskStatus
    • UndecidedEventRelation

More information on the differences between various Salesforce APIs can be found here.

:::info Force Using Bulk API If you set the Force Use Bulk API option to true, the connector will ignore unsupported properties and sync streams using BULK API. :::

Missing Records (Salesforce API Eventual Consistency)

Salesforce does not guarantee that recently created or updated records are immediately available through its API. A record may have its SystemModStamp set, but the underlying transaction may not yet be committed. During an incremental sync, the connector can advance its cursor past such records, causing them to be permanently missed in subsequent syncs.

Symptoms:

  • Records are missing in the destination but present when queried directly in Salesforce
  • The same source synced to a different destination at a later time does not have missing records
  • No errors appear in sync logs

Solution: Increase the Lookback Window in the connector configuration. This controls how far back the connector re-reads data from the last cursor position on each incremental sync. The default is PT10M (10 minutes). If you observe missing records, try increasing it to PT30M (30 minutes) or PT1H (1 hour). Because the connector uses append-dedup mode, re-reading overlapping data does not create duplicates in the destination.

The lookback window uses the ISO 8601 duration format. The format is PT<number><unit>, where P marks the start of the duration and T separates date from time components. Common examples:

ValueMeaning
PT10M10 minutes
PT30M30 minutes
PT1H1 hour
PT2H2 hours
P1D1 day
</details>

Changelog

<details> <summary>Expand to review</summary>
VersionDatePull RequestSubject
2.7.182026-02-2573501fix(source-salesforce): skip time-based slicing for full_refresh syncs (AI-Triage PR)
2.7.172026-02-1073235Make lookback window configurable to address Salesforce API eventual consistency
2.7.162025-10-2969078Promoting release candidate 2.7.16-rc.1 to a main version.
2.7.16-rc.12025-10-2766136Minor performance tuning
2.7.152025-10-2268166Add ActivityFieldHistory to UNSUPPORTED_FILTERING_STREAMS to fix missing records
2.7.142025-10-2168455Update dependencies
2.7.132025-10-1460432Update dependencies
2.7.122025-09-1566136Update to CDK v7
2.7.112025-05-1460271Define suggested streams
2.7.102025-05-1060100Update dependencies
2.7.92025-05-0459644Update dependencies
2.7.82025-04-2758997Update dependencies
2.7.72025-04-1958453Update dependencies
2.7.62025-04-1257976Update dependencies
2.7.52025-04-0557424Update dependencies
2.7.42025-03-2753689catch JSONDecodeError for error response
2.7.32025-03-2956776Update dependencies
2.7.22025-03-2455898Fix input state serialization issues
2.7.12025-03-2251921Update dependencies
2.7.02025-03-2055186Update manifest for adapting changes with AsyncRetriever
2.6.52025-02-2054178Promoting release candidate 2.6.5-rc.1 to a main version.
2.6.52025-02-2054178Promoting release candidate 2.6.5-rc.1 to a main version.
2.6.5-rc.12025-02-1853229Upgrade to API v62.0
2.6.42025-01-1148635Starting with this version, the Docker image is now rootless. Please note that this and future versions will not be compatible with Airbyte versions earlier than 0.64
2.6.32024-11-0546835Update dependencies
2.6.22024-10-10Bump minimum CDK to 5.10.2
2.6.12024-10-0546436Update dependencies, including CDK fix in v5.10.2
2.6.02024-10-0245678Have bulk streams use CDK components
2.5.342024-09-2846187Update dependencies
2.5.332024-09-2145779Update dependencies
2.5.322024-09-1445579Update dependencies
2.5.312024-09-0745329Update dependencies
2.5.302024-08-3144959Update dependencies
2.5.292024-08-2444747Update dependencies
2.5.282024-08-1744327Update dependencies
2.5.272024-08-1243732Update dependencies
2.5.262024-08-1043673Update dependencies
2.5.252024-08-0343211Update dependencies
2.5.242024-07-2742685Update dependencies
2.5.232024-07-2042166Update dependencies
2.5.222024-07-1341752Update dependencies
2.5.212024-07-1041529Update dependencies
2.5.202024-07-0941255Update dependencies
2.5.192024-07-0841043Use the latest CDK version possible
2.5.182024-07-0640835Update dependencies
2.5.172024-06-2540329Update dependencies
2.5.162024-06-2139927Update dependencies
2.5.152024-06-1639517Salesforce refactor: add CheckpointMixin for state management
2.5.142024-06-0639269[autopull] Upgrade base image to v1.2.2
2.5.132024-05-2338563Use HttpClient to perform HTTP requests for bulk, authentication and schema discovery
2.5.122024-05-1638255Replace AirbyteLogger with logging.Logger
2.5.112024-05-0938205Use new delete method of HttpMocker for test_bulk_stream
2.5.102024-05-0938065Replace deprecated authentication mechanism to up-to-date one
2.5.92024-05-0237749Adding mock server tests for bulk streams
2.5.82024-04-3037340Source Salesforce: reduce info logs
2.5.72024-04-2436657Schema descriptions
2.5.62024-04-1937448Ensure AirbyteTracedException in concurrent CDK are emitted with the right type
2.5.52024-04-1837419Ensure python return code != 0 in case of error
2.5.42024-04-1837392Update CDK version to have partitioned state fix
2.5.32024-04-1737376Improve rate limit error message during check command
2.5.22024-04-1537105Raise error when schema generation fails
2.5.12024-04-1137001Update airbyte-cdk to flush print buffer for every message
2.5.02024-04-1136942Move Salesforce to partitioned state in order to avoid stuck syncs
2.4.42024-04-0836901Upgrade CDK for empty internal_message empty when ExceptionWithDisplayMessage raised
2.4.32024-04-0836885Add missing retry on REST API
2.4.22024-04-0536862Upgrade CDK for updated error messaging regarding missing streams
2.4.12024-04-0336385Retry HTTP requests and jobs on various cases
2.4.02024-03-1235978Upgrade CDK to start emitting record counts with state and full refresh state
2.3.32024-03-0435791Fix memory leak (OOM)
2.3.22024-02-1935421Add Stream Slice Step option to specification
2.3.12024-02-1235147Manage dependencies with Poetry.
2.3.02023-12-1533522Sync streams concurrently in all sync modes
2.2.22024-01-0433936Prepare for airbyte-lib
2.2.12023-12-1233342Added new ContentDocumentLink stream
2.2.02023-12-1233350Sync streams concurrently on full refresh
2.1.62023-11-2832535Run full refresh syncs concurrently
2.1.52023-10-1831543Base image migration: remove Dockerfile and use the python-connector-base image
2.1.42023-08-1729538Fix encoding guess
2.1.32023-08-1729500handle expired refresh token error
2.1.22023-08-1028781Fix pagination for BULK API jobs; Add option to force use BULK API
2.1.12023-07-0628021Several Vulnerabilities Fixes; switched to use alpine instead of slim, CVE-2022-40897, CVE-2023-29383, CVE-2023-31484, CVE-2016-2781
2.1.02023-06-2627726License Update: Elv2
2.0.142023-05-0425794Avoid pandas inferring wrong data types by forcing all data type as object
2.0.132023-04-3025700Remove pagination and query limits
2.0.122023-04-2525507Update API version to 57
2.0.112023-04-2025352Update API version to 53
2.0.102023-04-0524888Add more frequent checkpointing
2.0.92023-03-2924660Set default start_date. Sync for last two years if start date is not present in config
2.0.82023-03-3024690Handle rate limit for bulk operations
2.0.72023-03-1424071Remove regex pattern for start_date, use format validation instead
2.0.62023-03-0322891Specified date formatting in specification
2.0.52023-03-0123610Handle different Salesforce page size for different queries
2.0.42023-02-2422636Turn on default HttpAvailabilityStrategy for all streams that are not of class BulkSalesforceStream
2.0.32023-02-1723190In case properties are chunked, fetch primary key in every chunk
2.0.22023-02-1322896Count the URL length based on encoded params
2.0.12023-02-0822597Make multiple requests if a REST stream has too many properties
2.0.02023-02-0222322Remove ActivityMetricRollup stream
1.0.302023-01-2722016Set AvailabilityStrategy for streams explicitly to None
1.0.292023-01-0520886Remove ActivityMetric stream
1.0.282022-12-2920927Fix tests; add expected records
1.0.272022-11-2919869Remove AccountHistory from unsupported BULK streams
1.0.262022-11-1519286Bugfix: fallback to REST API if entity is not supported by BULK API
1.0.252022-11-1319294Use the correct encoding for non UTF-8 objects and data
1.0.242022-11-0118799Update list of unsupported Bulk API objects
1.0.232022-11-0118753Add error_display_message for ConnectionError
1.0.222022-10-1217615Make paging work, if cursor_field is not changed inside one page
1.0.212022-10-1017778Add EventWhoRelation to the list of unsupported Bulk API objects.
1.0.202022-09-3017453Check objects that are not supported by the Bulk API (v52.0)
1.0.192022-09-2917314Fixed bug with decoding response
1.0.182022-09-2817304Migrate to per-stream states.
1.0.172022-09-2317094Tune connection check: fetch a list of available streams
1.0.162022-09-2117001Improve writing file of decode
1.0.152022-08-3016086Improve API type detection
1.0.142022-08-2916119Exclude KnowledgeArticleVersion from using bulk API
1.0.132022-08-2315901Exclude KnowledgeArticle from using bulk API
1.0.122022-08-0915444Fixed bug when Bulk Job was timeout by the connector, but remained running on the server
1.0.112022-07-0713729Improve configuration field descriptions
1.0.102022-06-0913658Correct logic to sync stream larger than page size
1.0.92022-05-0612685Update CDK to v0.1.56 to emit an AirbyeTraceMessage on uncaught exceptions
1.0.82022-05-0412576Decode responses as utf-8 and fallback to ISO-8859-1 if needed
1.0.72022-05-0312552Decode responses as ISO-8859-1 instead of utf-8
1.0.62022-04-2712335Adding fixtures to mock time.sleep for connectors that explicitly sleep
1.0.52022-04-2512304Add Describe stream
1.0.42022-04-2012230Update connector to use a spec.yaml
1.0.32022-04-0411692Optimised memory usage for BULK API calls
1.0.22022-03-0110751Fix broken link anchor in connector configuration
1.0.12022-02-2710679Reorganize input parameter order on the UI
1.0.02022-02-2710516Speed up schema discovery by using parallelism
0.1.232022-02-1010141Processing of failed jobs
0.1.222022-02-0210012Increase CSV field_size_limit
0.1.212022-01-289499If a sync reaches daily rate limit it ends the sync early with success status. Read more in Performance considerations section
0.1.202022-01-269757Parse CSV with "unix" dialect
0.1.192022-01-258617Update connector fields title/description
0.1.182022-01-209478Add available stream filtering by queryable flag
0.1.172022-01-199302Deprecate API Type parameter
0.1.162022-01-189151Fix pagination in REST API streams
0.1.152022-01-119409Correcting the presence of an extra else handler in the error handling
0.1.142022-01-119386Handling 400 error, while sobject doesn't support query or queryAll requests
0.1.132022-01-118797Switched from authSpecification to advanced_auth in specefication
0.1.122021-12-238871Fix examples for new field in specification
0.1.112021-12-238871Add the ability to filter streams by user
0.1.102021-12-239005Handling 400 error when a stream is not queryable
0.1.92021-12-078405Filter 'null' byte(s) in HTTP responses
0.1.82021-11-308191Make start_date optional and change its format to YYYY-MM-DD
0.1.72021-11-248206Handling 400 error when trying to create a job for sync using Bulk API.
0.1.62021-11-168009Fix retring of BULK jobs
0.1.52021-11-157885Add Transform for output records
0.1.42021-11-097778Fix types for anyType fields
0.1.32021-11-067592Fix getting anyType fields using BULK API
0.1.22021-09-306438Annotate Oauth2 flow initialization parameters in connector specification
0.1.12021-09-216209Fix bug with pagination for BULK API
0.1.02021-09-085619Salesforce Aitbyte-Native Connector
</details> </HideInUI>