We will be upgrading our backend systems from Docker Swarm to Kubernetes in order to enhance our infrastructure to provide you with a better experience. This upgrade is scheduled to begin in 2024, and we anticipate the process to be ongoing for several months as we coordinate upgrading customer subscriptions and migrating the existing apps over to the new infrastructure. Only Flogo, BW, and NodeJS apps will be migrated, Scribe apps are not a part of this process. 

During this time, we will be notifying users within TCI when it's time for their apps to move. You can review all known issues below as well as how to be as ready as possible for the coming changes. 

We encourage you to plan accordingly and thank you for your understanding and patience during this upgrade. If you have any questions or concerns, please feel free to contact our support team.

Readiness Checklist

• Ensure you are using the latest tibagent software

• Leverage new tiblogs binaries to retrieve your older logs, please open a support case to gain access to the tiblog tool.

• Configure Timer Triggers to account for a rerun when the app is migrated

• Ensure you have whitelisted ALL relevant TIBCO IP addresses

• App Endpoint URLs WILL change post-migration. Client-side code addressing the original endpoints will continue working, but there will be some added latency as we re-route internally to the new endpoints. Over a period of time, it would be good to update client-side code to use the new app endpoint urls.

• Mock Apps are not migrated 

Known Issues

• App rescaling infinitely (IPAS-21522) Rescaling an App with an already expired SFDC Token can cause an infinite loop of scaling up. The workaround is to reconfigure the connection and rescale the app.
   

• Cannot Restore old pinned version of App (IPAS-21644) Cannot restore a pin from CIC1 version of App.
   

• Existing Logs are missing in the UI (IPAS-21722) Existing logs are not gone, but will not be available in the UI after migration. You can leverage this process to retrieve your logs via the tiblog CLI binary found here.
   

• New Logs are missing in the UI (IPAS-21467 & IPAS-20680) Log out and in again to see newly generated logs in the UI.
   

• Seeing IPs instead of Instance IDs for Resource Metrics (IPAS-20398) Log out and back in again
   

• App running off schedule (IPAS-21499) - Since the app is restarted during migration the app will automatically reset its timer and be re-triggered, either if it had an OnStart/OnShut trigger or if it leveraged a Timer trigger.
   

• App using VPN Connection fails (IPAS-21962) - After the Upgrade is completed, update your VPN password and scale up the app to take effect.
   

• Latency on old CIC1 URLs () When hitting an old URL Endpoint after your Apps have been migrated to the new infrastructure, there will be a slightly increased latency on the processing as we reroute your call internally to the new endpoint. To maintain peak performance get your client-side calls updated as soon as you can to eliminate the overhead.
   

• Cloud Mesh Endpoint Not Working (IPAS-22063) - When a BW App is pushed post migration from a parent/child configuration and is called via an app that has yet to be migrated then the Cloud Mesh Endpoint is unavailable
   

• Secrets API endpoints return “Operation not allowed for your subscription” (IPAS-21911) - This feature is only available in the new backend Kubernetes infrastructure, you will gain access to it once your subscription has been moved.
   

• Issue applicable to the app having an old image (IPAS-22500) - If you have an app created before Feb 2022,  after migrating this app if it is copied to another org then you will see logs showing only dates, metrics are not recorded, updated password is not picked up by the app. To resolve this please re-push the app.

• Websocket server to emits one message at migration (IPAS-21500) - If there is a websocket server trigger in flogo that gets invoked as soon as an instance is up and running, post-migration it emits a message which the Websocket subscriber consumes as soon as it is up and running.
   

• URL Published to Mashery will need to be deleted and re-added (IPAS-21989) - If you have previously published an endpoint to Mashery, in order to update that to the newly created URL you will need to delete and re-add it. The previous URL will STILL work until this action is taken.