I’ve got an Atlas Search Index on a collection with ~12M items. It’s been working fine for about 2 years and since yesterday it got stuck in the “initial sync” stage - it indexes a couple thousand items and then starts over. I’ve got multiple processes that update the items in the collection and the loop only happens when they’re active. Once I stop them the index remains in the “Active” state. I’m on a M50 tier with NVMe SSD and mongo version 4.4.13. Can you please assist me, or should I open a support ticket for it?
Welcome to the community @Dimitar_Tsonev. I would suggest a support case for this type of issue. However, I would keep an eye on if you have enough disk space to support to copies of your search index.
I’ve opened a support ticket actually and they investigated the issue - it seems I’ve got a couple of records with too much data in them and that’s breaking the index. Working on it. Thank you!