What exactly is Hare and why is it needed? #597
-
|
Hare is used to as specified in this doc:
|
Beta Was this translation helpful? Give feedback.
Replies: 2 comments 1 reply
-
|
@hessio, OK but I would say, we can even manage multiple services with systemctl, https://opensource.com/article/20/12/multiple-service-instances-systemctl utility as well |
Beta Was this translation helpful? Give feedback.
-
|
@mukul-seagate11 you are missing the word "distributed", I'm afraid. I'd recommend learning more about Consul and consensus protocols (like Paxos and Raft) - why they are needed and what problems they are solving. |
Beta Was this translation helpful? Give feedback.
-
|
Because CORTX is a distributed application (running on many nodes). It's not easy to start/stop/monitor such applications so we need some application which can keep track of the nodes. Hare will also generate the cluster configuration from CDF (Cluster Description File). Hare is built on top of Consul - watch this video to see what Consul is and that should make more sense why we need Hare. Hare does not do any recovery, it will notice the problem and inform all instances of the issue. CORTX-HA will take care of failover/failback control flow for affected services and stabilizes them across CORTX cluster. |
Beta Was this translation helpful? Give feedback.
@hessio, OK but I would say, we can even manage multiple services with systemctl, https://opensource.com/article/20/12/multiple-service-instances-systemctl utility as well