6 ms·
Yep. It's fun when your entire etcd cluster dies at the same time (because you started a global unit that pulls a huge docker container). You'd think they have
by Gigablah 12y ago
Yep. It's fun when your entire etcd cluster dies at the same time (because you started a global unit that pulls a huge docker container).
You'd think they have safeguards to automatically rebalance on certain conditions (metadata filling up, etc) but no, you have to babysit btrfs.
- saryant 12y agoI'm glad I'm not the only one with these problems. I was starting to think I was going crazy with my etcd+btrfs issues.
- 23david 12y agohrm... what's the failure scenario here? Are there any good btrfs management utils that can help with running btrfs in prod?
- throwaway_nsa2 12y agoStop it with the anti-American attitude. We are keeping tabs on you.
- throwaway_nsa3 12y agoStop it with the anti-American attitude. We are keeping tabs on you.