We’ve been receiving email warnings about “Connections % of configured limit has gone above 80” for a couple hours this morning, but can’t identify anything we’re doing as the source of these extra connections. We have a hosted M10 cluster, AWS N. California, v4.4.16, with a Realm sync instance. Realm sync says there has been 23 requests in the last hour, so not a lot.
Looking at the realtime data, local.oplog.rs looks to be consuming the most cycles, followed by _realm_sync.history and _realm.sync.resume_tokens. Is this just a lot of sync activity? Or should I restart the sync?
We had a similar issue last week that just “resolved itself” after a few hours.
Is this just a lot of sync activity? Or should I restart the sync?
It’s hard to say what the cause of the connection surge could be with the information at hand. However, have you had a chance to investigate the mongod logs starting from just before the surge and ending just after it dips back to your regular connection levels? This may provide some clue or further insight into which client(s) have caused the connection surge.
In saying so, perhaps it may be best to raise this with the Atlas chat support team as they have more insight into the associated Atlas project / cluster.
We figured out what was happening (we think!). We were doing some smaller sized updates to our MongoDB data, which then gets updated in Realm. We made the updates bigger and less frequent, and that appears to have solved it…thanks!
