Bloom filter (strange behavior)

There’s the problem! Now, I have no idea why your node is getting shut down, but for sure, the current code will consider the GC run failed and delete the bloom filter.

We are working on changes that will allow a storage node to resume after a restart from where it left off, but these changes will take some time. Here is one: https://review.dev.storj.io/c/storj/storj/+/12806

Anyway, it is certainly not the case that we’re expecting storage nodes to restart regularly. Looking at a random sample of nodes, the average node has no process restarts for over a week, and many have not been restarted at all in 20 days. Notably, we’re in a middle of a rollout, so node process uptime is lower than it might otherwise be.

Whether or not we get pause/resume completed soon, I think the key takeaway is your node will be happier if you can help it stay running. Do you have any indication of why it restarted?

2 Likes