*: support Writer with context

[fanminshi]

backup/writer interface now reasons about input context. The context controls the writer.Write operation.

ref: #1906

[fanminshi]

cc/ @hongchaodeng @hasbro17

[xiang90]

/cc @rjtsdl for abs changes.

[hongchaodeng]

LGTM for S3 part.
@rjtsdl Can you help implement the ABS part for context support?

[hongchaodeng]

@fanminshi
Can you try the manual test:

• set up 1s timeout and have large snapshot and expect it to fail due to timeout.
• increase timeout to something larger then make it succeed.

[fanminshi]

@hongchaodeng that was my plan for the 4th bullet point #1906

[rjtsdl]

@hongchaodeng yep, i can help on ABS ctx support. Do we want it in this PR? Or I can create a new one.

[fanminshi]

@rjtsdl let's support abs ctx in a new pr.

[fanminshi]

Manual test plan

Case: 1 second backup timeout

etcd size:

$ bin/etcdctl --endpoints=http://192.168.99.100:32379 endpoint status
http://192.168.99.100:32379, c257147bab1ca7cf, 3.2.13, 71 MB, false, 4, 51198, 0,

1 second timeout:

apiVersion: "etcd.database.coreos.com/v1beta2"
kind: "EtcdBackup"
metadata:
  name: example-cluster-backup
spec:
  etcdEndpoints: ['example-etcd-cluster-client:2379']
  storageType: S3
  backupPolicy:
    TimeoutInSecond: 1
  s3:
    path: jenkins-testing-operator/vault.etcd.backup
    awsSecret: aws

output:

time="2018-02-06T22:49:58Z" level=info msg="getMaxRev: endpoint example-etcd-cluster-client:2379 revision (52065)"
time="2018-02-06T22:49:59Z" level=error msg="error syncing etcd backup (backup/example-cluster-backup): failed to save snapshot (failed to write snapshot (MultipartUpload: upload multipart failed
	upload id: X1OsTyWaRuHqvD10V0DatRqrCkhdnbS_OXQEo0Uc0StSbNSgglRSFzHUsTOujJIik6Dex48U42akh01sj88f7ZYTxX4VfPgRtONWcJhRSnin1ws.Oakq4hcFyYN5fF7s
caused by: RequestCanceled: request context canceled
caused by: context deadline exceeded))" pkg=controller

Case: 100 seconds backup timeout

etcd size:

$ bin/etcdctl --endpoints=http://192.168.99.100:32379 endpoint status
http://192.168.99.100:32379, c257147bab1ca7cf, 3.2.13, 71 MB, false, 4, 51198, 0,

100 backup timeout:

apiVersion: "etcd.database.coreos.com/v1beta2"
kind: "EtcdBackup"
metadata:
  name: example-cluster-backup
spec:
  etcdEndpoints: ['example-etcd-cluster-client:2379']
  storageType: S3
  backupPolicy:
    TimeoutInSecond: 100
  s3:
    path: jenkins-testing-operator/vault.etcd.backup
    awsSecret: aws

output:

$ kubectl -n backup get etcdbackups.etcd.database.coreos.com example-cluster-backup -o yaml
apiVersion: etcd.database.coreos.com/v1beta2
kind: EtcdBackup
metadata:
  clusterName: ""
  creationTimestamp: 2018-02-06T22:53:34Z
  generation: 0
  initializers: null
  name: example-cluster-backup
  namespace: backup
  resourceVersion: "155599"
  selfLink: /apis/etcd.database.coreos.com/v1beta2/namespaces/backup/etcdbackups/example-cluster-backup
  uid: 8f8d58d9-0b90-11e8-b6a1-080027ae9197
spec:
  backupPolicy:
    timeoutInSecond: 100
  etcdEndpoints:
  - example-etcd-cluster-client:2379
  s3:
    awsSecret: aws
    path: jenkins-testing-operator/vault.etcd.backup
  storageType: S3
status:
  etcdRevision: 52065
  etcdVersion: 3.2.13
  succeeded: true

[fanminshi]

cc/ @hongchaodeng. Manually tested.

[hongchaodeng]

LGTM