Ask Your Question

What is the reason to send writes to all replica nodes?

asked 2015-04-15 15:45:11 -0600

Gayana gravatar image

updated 2015-04-15 16:29:25 -0600

smaffulli gravatar image

Swift now send the writes to all the replicas. But if we that behaviour to only send object to one replica system should perform fast.

But it seems the replica maintenance services running on Swift-storage nodes slows down the cluster.

Is this the reason to send the writes to all the replica nodes? Because it is eventual consistent, why we write to all the nodes every time?

edit retag flag offensive close merge delete

1 answer

Sort by ยป oldest newest most voted

answered 2015-04-15 22:46:00 -0600

torgomatic gravatar image

When a client performs an object PUT and Swift returns a 201 response, that means that the object has been safely stored and will be retrievable in the future.

If Swift were to write to only one disk and that disk failed before a second replica was written, then the object would be lost.

Disks fail at a high enough rate that writing to one disk is not enough to be safe.

edit flag offensive delete link more

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower


Asked: 2015-04-15 15:45:11 -0600

Seen: 62 times

Last updated: Apr 15 '15