*: snapshot RPC #5012
*: snapshot RPC #5012
Conversation
| bw += n | ||
| } | ||
| } | ||
| pw.Close() |
There was a problem hiding this comment.
defer pw.Close()? Or do we only do this when it was exited with break?
There was a problem hiding this comment.
There are failure paths that do CloseWithError so defer pw.Close() would try to double close.
There was a problem hiding this comment.
Oh there's CloseWithError. Got it.
|
all fixed up ptal /cc @xiang90 I've been testing with: # feed keys into V3Demo cluster before this
ETCDCTL_API=3 ./bin/etcdctl snapshot save test.db
# kill V3Demo cluster here
rm -rf infra1-test.etcd/; ETCDCTL_API=3 ./bin/etcdctl snapshot restore test.db --name="infra1-test" --initial-cluster="infra1-test=http://localhost:2380" --initial-advertise-peer-urls="http://localhost:2380"
# start etcd on infra1-test |
| wr.CompactRevision = 1 | ||
| func NewSnapshotRestoreCommand() *cobra.Command { | ||
| cmd := &cobra.Command{ | ||
| Use: "restore <filename>", |
There was a problem hiding this comment.
I changed it to use the node name so it matches etcd directory naming behavior. Should it be inconsistent with that?
There was a problem hiding this comment.
oh... it should be consistent...
There was a problem hiding this comment.
--data-dir '${name}.etcd'
path to the data directory.
There was a problem hiding this comment.
OK; default to the name if --data-dir is not specified?
|
LGTM. Remove the last commit? |
|
fixed a bug in 3-cluster restore and added support for deleting member bucket contents. ptal /cc @xiang90 |
|
LGTM |
Includes updated etcdctl snapshot. Restore support still WIP