This is the guide line for seamless migration of existing sync jobs into a new job - File cache or Hybrid work.
1. It's highly advisable that the files remain untouched during the migration process. If it's not possible at all, the changes files will need to be synced later, which may lead to undefined behavior or even file corrupt.
2. Don't change the number of Agents in the job during migration. All Agents from the old job must be added to the new job, no new Agents must be added. Do that after the new job is aligned.
3. Check Agent profile. If you have parameter priority_peers here - remove them. It's advisable not to have this parameter in the Agent profile for the new jobs, but configure it right in the job.
4. Check the size of the sync job on the Agents' storage (size of data + size of .sync/Archive). Be sure to set the cache size limit in File policy at least 10-15% bigger than that size, or no limit at all.
To update the existing Synchronization jobs to the File cache job, follow the steps below.
1. Ensure that Agents are in 'synced' state. Safest is to pause the Agents and wait for a while to make sure that Agents cease any file activity.
2. Shutdown the Agents.
3. On each Agent copy database files (all *.db and *.db.files) from current sync job to a safe place. How to discover the database name?
4. Start the Agents.
5. Note the Agents' paths in the sync job and delete the current sync job from the MC.
6. Create new File cache or Hybrid work job and arrange the Agents in it as needed
- at this step don't create any High availability or Gateway Scaleout groups.
- be sure to use the same paths as the Agents were using the sync job
- be sure to set the cache size limit in File policy at least 10-15% bigger than the size of sync job, including the Archive, or no limit at all.
7. Wait until the Agents create new databases, find these database files in the Agent's storage. Status should change from "deploying policy" to 'synced'.
8. Shutdown Agents.
9. Rename databases you copied on step 3 - give them the name of the newly created with new databases names that agents created in new job. First part of name is same on all agents, but the number in the middle is different. See example below.
10. Start the Agents.
How to discover database names?
Go to Agents -> <an agent in the sync job> -> Configuration and click to open Debug status. Scroll to "transfers" section and find the sync job in question by name. Note its "id". Open the Agent's storage folder and find the db files starting with same string as the ID.
Example
For example, there are 2 Agents in sync job:
Agent1 has ABC.123.db and ABC.123.files2.db;
Agent2 has QWE.789.db and QWE.789.files2.db.
Follow the steps 1-6 above.
Now there are two new databases:
Agent1 created DEF.654.db and DEF.654.files2.db;
Agent2 created XYZ.551.db and XYZ.551.files2.db.
Files must be renamed the following way:
ABC.123.db --> DEF.654.db ABC.123.files2.db --> DEF.654.files2.db QWE.789.db --> XYZ.551.db QWE.789.files.db --> XYZ.551.files2.db