This morning I noticed an app we used (Task-Tracker) is not syncing. I logged into console and navigated to the App Services section->Logs and and am seeing a message
Synchronization between Atlas and Device Sync has been stopped, due to error:
encountered non-recoverable resume token error. Sync cannot be resumed from this state and must be terminated and re-enabled to continue functioning: (ChangeStreamHistoryLost) PlanExecutor error during aggregation :: caused by :: Resume of change stream was not possible, as the resume point may no longer be in the oplog.
With a button to Restart Atlas Sync. Pressing it comes back to the same vague error. I am working with Chat Support now and will circle back with any additional information.
Clear server generated errors and suggestions for corrective action would be really helpful as I have no idea what to do what that.
The resolution, which I don’t quite understand was the following:
To follow up on this, the App services team informed that your cluster have fallen off the oplog, so you will have to terminate/reenable the Sync.
Note that as mentioned before, this errors is common on free tiers since are not suggested for using the Sync feature.
I assumed the button thats says Restart Atlas Sync (shown in the above post) indicated that it was already terminated. Apparently not, it has to be manually terminated and restarted. Either the button is broken or it doesn’t do what the title says it does.
The really interesting part is the response that this is a “common error” on a free tier and is “not recommended for syncing.” That doesn’t make a whole lot of sense to me if you just trying out MongoDB it would crash like that - but ok. Hmmm. My .02
