Matthew Jaffee

Thanks @benbjohnson this is awesome, I have a few notes: Making this a seemless upgrade seems like it will be pretty involved. Based on some of the other work we...

My thinking was that there is already an inherent upper limit on the number of rows you can have in Pilosa since all the rows of all the fields are...

I don't think I have a good enough feel for the tradeoffs between sharding/not sharding rows. Sharding+caching definitely gives us the most flexibility going forward though... just at the cost...

seems like a good place to start, thanks Ben!

possible this was resolved by #1717 - need to confirm

@yuzhichang yes, it should be possible to set a new coordinator when the coordinator is down. Honestly though, I'm not sure we have a test for this particular case, so...

using import-roaring works around this, so it's a bit lower priority.

@sydt2014 sorry for the delay here — the short answer is that while this is possible, you'd probably have to do quite a bit of custom work to make it...

While we are not able to actively contribute to Pilosa at this time, please ask any questions in the slack [support channel](https://www.pilosa.com/community/#support) and we will be glad to help. We...

Had to back this change out—seems there is a race between when the cluster knows about all the nodes and when holder opening occurs which was causing #1948 I think...