Andrey Vasnetsov

But you are also using custom sharding, which means that you need to explicitly place your data into one or another shard. May I ask which criteria you are using...

/bounty $200

Hi @vitali-ipquants, thanks for the suggestion! Unfortunately, the implementation of cluster level snapshot is associated with a number of technical difficulties, including: - Where to store unified snapshot? - How...

> A solution which could work without significant changes is to require each node to have access to a shared file system - something like NFS, CephFS. We can't assume...

/bounty $100

> What about going with a cluster deployment? Will that help? marginally. Still creating 2000 collections in the loop is a bad idea

> oh the loop is just a way to reproduce the issue. In our system we create the collections gradually over multiple weeks. that counts as well

Hey @AlexandruCihodaru, we unfortunately can't remove `Initializing` right away, it would create problems during cluster version upgrade, when some nodes have latest versions, while others don't. Here is a step-by-step...

> In this first PR I attempted to address the first bullet "Make Initializing behave as Active in all situations" Oh, ok. Let me have another look then

/bounty $200