Loading
Salesforce now sends email only from verified domains. Read More

Instance Refreshes and Consolidations

Publish Date: Sep 16, 2025
Description

Last updated on September 16, 2025

To support your org’s continued growth in our Salesforce-managed (first-party) data centers, and provide the availability and resiliency you expect from Salesforce, we occasionally need to relocate your org onto more modern infrastructure. Hyperforce is the premiere infrastructure for Salesforce. However, for our customers still in first-party, we remain committed to providing a best-in-class environment. Salesforce is consolidating our first-party footprint into fewer, modern data centers, and retiring older data centers. As a result of this work, two maintenance operations can affect your first-party org.

  • Instance Refresh (IR) - An operation that moves all customer orgs off of an older source instance, and onto a newer target instance. After the operation completes, the source instance is decommissioned. 

  • Instance Consolidation (IC) - An operation that adds customer orgs to an existing, operating target instance.

Your preparation for these maintenance activities differs. If you follow these best practices, the maintenance will be seamless, regardless of which operation is occurring for your org. This article provides a list of frequently asked questions to help you prepare for these operations.

NOTE: This document is for informational purposes only, and is not part of any legal or otherwise binding agreement. The policies and practices described in this document are subject to change at Salesforce's sole discretion.

Resolution

How to Prepare for an Instance Consolidation

  • What is an instance consolidation (IC)?
    An instance consolidation adds new orgs to the instance that hosts your org. 

  • What happens to my org during an instance consolidation?
    Your org will undergo a site-switch, and experience up to 2 hours of downtime. You can't access the org during downtime.

  • What steps should I take to prepare for an instance consolidation?
    An instance consolidation includes a site-switch. Follow the instructions in Site Switching Overview and FAQ for preparation steps, and to understand how site switching affects different services. Instance consolidations include up to 2 hours of downtime. You can't access the org during downtime. If you allowlist IPs or have custom DNS timeouts, you may need to update your network settings.

  • Can I opt-out of an instance consolidation?
    No. Instance consolidations affect the entire instance. It’s not possible for orgs to opt out.

  • Is there any way to avoid instance consolidations?
    Yes. Upgrade to Hyperforce. Orgs on Hyperforce aren’t affected by ICs. Salesforce is strategically scheduling as many orgs as possible to upgrade to Hyperforce before they’re subject to ICs. If you’re scheduled to upgrade, don’t defer. If you’re not yet scheduled to upgrade, you can contact your account team at any time to request a migration to Hyperforce. You don’t need to wait for an IR or IC notice to make the request.

  • How often do instance consolidations happen?
    There is no fixed frequency for instance consolidations. Salesforce endeavors to limit IRs and ICs to one event per year per org. However, to accommodate all necessary relocations, a single instance can undergo multiple consolidations, and it's possible for an org to experience more than one in a year.

  • How will I know when my org will undergo an instance consolidation?
    Salesforce sends Product & Service Notification emails to the org’s admins 60 days before the instance consolidation. You can also monitor the Maintenance tab for Salesforce Services on trust.salesforce.com. Read Preferred Salesforce Maintenance Schedule to see when Salesforce typically schedules different types of maintenance.  

  • Will my org ID change after an instance consolidation?
    No. 

  • What happens to sandbox refreshes during an IC?
    Sandbox refreshes are stopped shortly before the site switch begins. In-progress sandbox refreshes are restarted, not resumed, after the site switch completes. You can’t start a sandbox refresh during the site switch.

  • What happens to my Trust notification subscriptions after an IC?
    Nothing. Your org’s instance isn’t changing, so there’s no change to your Trust notifications.

 

How to Prepare for an Instance Refresh

  • What is an instance refresh (IR)?
    An instance refresh moves your org from an older, source instance, to a newer, target instance. This operation provides your org with better infrastructure, more room to grow, and the best availability and resiliency possible in our first-party data centers.

  • Can I opt out of an instance refresh?
    No. An IR is an instance-wide operation that affects every org. Because the source instance is decommissioned after the operation, no orgs can remain behind.

  • What happens to my org during an instance refresh?
    Your org moves to a new data center, and the name of your instance changes. The maintenance takes up to 4 hours. During this time, if your production org is moving, it is placed in read-only mode. If your sandbox org is moving, it will experience downtime and won't be accessible. Learn more about read-only mode in Read-Only Mode Overview. 

Be prepared to take these actions after the instance refresh.

    • If you experience integration issues after the IR, refresh your integrations.
    • If you continue to see a "Maintenance in Progress" alert on your login page after the IR completes, refresh your DNS cache.
    • If you have Einstein Bots enabled on your org, trigger a new org authentication to Einstein.AI by disabling and re-enabling Einstin Bots under Setup.

  • Will my org ID change after an instance refresh?
    No.

  • Do the standard maintenance and release windows change after an IR?
    Sometimes. Usually, when your org reaches the new instance, it will have the same release and maintenance windows as on the source instance. In the rare cases when an org’s maintenance or release windows must change, Salesforce notifies the org admins before the IR.

  • What happens to the data in the old data center after an IR?
    After the IR completes, we decommission the hardware that hosted your instance in the old data center. Our procedures meet compliance and security requirements, and ensure that your data doesn’t leave the previous data center. Maintaining your trust is our #1 value.

  • What happens to my Trust notification subscriptions after an IR?
    Salesforce auto-subscribes existing subscribers on the source instance to notifications on the target instance.

  • What happens to sandbox refreshes during an IR?
    Sandbox refresh queues are suspended starting 4 hours before the IR begins. In-progress sandbox refreshes are restarted, not resumed, 24 hours after the IR completes.

  • What happens to previously scheduled activities such as weekly exports and Apex callouts during an IR?
    Scheduled activities, like weekly exports and Apex jobs, are paused before the IR and resumed after the IR completes. Any jobs scheduled to run during the IR will start when the maintenance window is over.

    If a job that was running before the IR returns an error afterward, you can restart the job to get the expected results. For the best experience, Salesforce recommends rescheduling large or long-running jobs to begin after the IR completes.

    Apex callouts to external services continue to run during the IR. However, because the Salesforce application is in read-only mode, any follow-up updates to Salesforce may fail. To prevent issues, Salesforce recommends taking steps to stop these callouts from executing while in read-only mode. For guidance, read Apex Callouts in Read-Only Mode.

  • What happens to integrations after an IR?
    Integrations, including 3rd-party integrations, usually resume automatically, but if you have problems, try restarting the integration. Restarting clears the DNS lookup cache so that your integrations detect the IP addresses of the new data center. Because restart steps can vary, check with the integration developer or on AppExchange for specific instructions for logging in and restarting your specific integrations.

  • What happens to certificates after an IR?
    Because each instance has its own unique certificate, the certificates change after an instance refresh. Salesforce strongly advises against pinning server certificates. If you must pin a certificate, make a pinset of the Mozilla roots, and follow the Certificate Changes Trailblazer Community group to see notifications of certificate changes for 1P production orgs. If you pin certificates, we encourage you to stop that practice as soon as possible. Learn more in Certificates in Salesforce.

  • What happens to email-to-service addresses after an IR?
    Email-to-service addresses, such as Email-to-Case and Email-to-Salesforce are unaffected. The previous instance is still appended to the address. However, delivery may be delayed until after the IR completes.

  • What happens to Live Agent or SOS implementations after an IR?
    If you use Live Agent or SOS, avoid hard-coding the endpoint URL on your webpage. The URL can change after an IR, and while our deployment code has a built-in redirect, it's best to update the URL in your code after the IR completes to prevent any service interruptions. Read more in Chat no longer works after server endpoint URL is updated.

  • What happens to Salesforce for Outlook (SFO) OAuth tokens after an IR?
    After the IR, you’ll be logged out of SFO and then prompted to log back in via the Setup Wizard. If you have trouble logging back in, read Salesforce for Outlook Reauthentication, and get help in the Salesforce for Outlook & Email Connect Trailblazer Community group.

  • What happens to High Volume Platform Events and Change Data Capture during an IR?

    • Some events aren’t migrated. Due to how our Event Bus works, some events published before the IR won’t be available on the new instance.
    • The 72-hour retention window is lost. You won’t be able to replay or access events that occurred before the IR, even if they are within the usual 72-hour window.
    • ReplayIDs aren’t continuous. The ReplayIDs for new events published after the IR won’t relate to the IDs from before. Subscribers won’t be able to replay events from a point in the stream that existed before the IR.

To retain events, drain the Event Bus to an external system before the IR. Make subscribers to the events, such as Streaming API clients, Apex triggers, and flows, aware of the change. For more information, read Publishing Events in Read-Only Mode.

  • What happens to Event Log Data after an IR?
    Event log data isn’t fully preserved after an IR. The system relies on a nightly process to populate monitoring events. As a result, any log data created during the maintenance period—from the day the IR begins until it's finished—won’t be moved to the new instance. This will create a gap in your event logs for that timeframe.

  • What happens after an IR if I use a third-party content delivery network (CDN)?
    To maintain your connection through the CDN, you must proactively update your CDN's configuration and certificates before the IR, and then thoroughly test the changes in a staging environment immediately after the maintenance is complete.

    • For your primary customer.my.salesforce.com domain:

      • Before the IR, you must update the origin in your CDN's configuration and push the changes to a staging environment.
      • After the IR, immediately test the new configuration in staging. Only deploy to production after you confirm that everything is working correctly.

    • For your VisualForce or ContentForce domains (e.g., customer--c.$NA10.content.force.com)

      • You must obtain a new SSL certificate that covers your customer.my.salesforce.com domain as well as both the old and new instance names for your VisualForce/ContentForce domains. For example: customer--c.NA10.content.force.com and customer--c.NA42.content.force.com.

      • Share this new certificate with your CDN provider so they can deploy it.
      • After the new certificate is deployed, update your CDN configuration with the new VisualForce/ContentForce hostnames and origin for all three endpoints, and push these changes to a staging environment.
      • As with your primary domain, test the staging configuration immediately after the IR before deploying to production.

NOTE: Because acquiring a new certificate and working with your third-party CDN provider can take several days, start the process at least 2 weeks before your IR.

NOTE: If you have a specialized acceleration of your MyDomain.my.salesforce.com, MyDomain.lightning.force.com, MyDomain—c.documentforce.com, MyDomain--PackageName.visualforce.com, MyDomain—c.InstanceName.content.force.com, or MyDomain--PackageName.InstanceName.visual.force.com hostnames using Akamai, you must update the origin and push the configuration to staging within the CDN before the IR. Immediately after the IR and before deploying to production, test that direct access and staging work as expected. This is a rare configuration. If you’re not sure what this is, then it doesn’t apply to your org.

This type of specialized CDN acceleration is being replaced by the Salesforce Edge Network. Salesforce doesn’t generally support adding hostnames to this type of specialized configuration. The recommended path forward is to work with Salesforce Support to migrate to the Salesforce Edge Network. This process will involve Salesforce R&D removing the DNS configuration overrides used by the specialized acceleration.

  • Where can I go if I have more questions?
    Post questions and collaborate with other Salesforce users on the Salesforce Service Delivery group in the Trailblazer Community, or open a support case via Salesforce Help.

Knowledge Article Number

000387056

 
Loading
Salesforce Help | Article