* If there is a snapshot, HTTP server won't start.
* Resotring form snapshot occurs after replaying WAL.
* When taking a snapshot, the last change is not applied to the state machine yet.
Currently our sample systemd service file `contrib/systemd/etcd.service`
have startup/shutdown dependency as below:
[Unit]
After=network.target
For some rare condition, e.g. bare matel deployment with slow network
startup, IP could not be assigned e arly enough before etcd default
`ETCD_HEARTBEAT_INTERVAL="100"` and `ETCD_ELECTION_TIMEOUT="1000"` get
timeouted, after graceful system reboot.
This cause etcd false negative classify itself use unhealthy, therefore
stop rejoining the remaining online cluster members.
This PR introduce:
- `etcd.service`: Ensure startup after `network-online.target` and
`time-sync.target`, so effective network connectivity and synced
time is available.
The logic is concept proof by
<https://github.com/alvistack/ansible-role-etcd/tree/develop>; also
works as expected with Ceph + Kubernetes deployment by
<https://github.com/alvistack/ansible-collection-kubernetes/tree/develop>.
No more deadlock happened during graceful system reboot, both AIO
single/multiple node with loopback mount.
Also see:
- <https://github.com/ceph/ceph/pull/36776>
- <https://github.com/etcd-io/etcd/pull/12259>
- <https://github.com/cri-o/cri-o/pull/4128>
- <https://github.com/kubernetes/release/pull/1504>
Signed-off-by: Wong Hoi Sing Edison <hswong3i@gmail.com>
This change makes the etcd package compatible with the existing Go
ecosystem for module versioning.
Used this tool to update package imports:
https://github.com/KSubedi/gomove
Update to Go 1.12.5 testing. Remove deprecated unused and gosimple
pacakges, and mask staticcheck 1006. Also, fix unconvert errors related
to unnecessary type conversions and following staticcheck errors:
- remove redundant return statements
- use for range instead of for select
- use time.Since instead of time.Now().Sub
- omit comparison to bool constant
- replace T.Fatal and T.Fatalf in tests with T.Error and T.Fatalf respectively because the goroutine calls T.Fatal must be called in the same goroutine as the test
- fix error strings that should not be capitalized
- use sort.Strings(...) instead of sort.Sort(sort.StringSlice(...))
- use he status code of Canceled instead of grpc.ErrClientConnClosing which is deprecated
- use use status.Errorf instead of grpc.Errorf which is deprecated
Related #10528#10438
The suggested pattern for Raft proposals is that they be retried
periodically until they succeed. This turns out to be an issue
when a leader cannot commit entries because the leader will continue
to append re-proposed entries to its log without committing anything.
This can result in the uncommitted tail of a leader's log growing
without bound until it is able to commit entries.
This change add a safeguard to protect against this case where a
leader's log can grow without bound during loss of quorum scenarios.
It does so by introducing a new, optional ``MaxUncommittedEntriesSize
configuration. This config limits the max aggregate size of uncommitted
entries that may be appended to a leader's log. Once this limit
is exceeded, proposals will begin to return ErrProposalDropped
errors.
See cockroachdb/cockroach#27772
