StreamLogs

Same as ListLogs but using server-side streaming. Also allows for tail -f semantics.

TypeScript

import { cloudApi, serviceClients, Session } from "@yandex-cloud/nodejs-sdk";

const StreamClusterLogsRequest =
  cloudApi.mdb.clickhouse_cluster_service.StreamClusterLogsRequest;
const StreamClusterLogsRequest_ServiceType =
  cloudApi.mdb.clickhouse_cluster_service.StreamClusterLogsRequest_ServiceType;

(async () => {
  const authToken = process.env["YC_OAUTH_TOKEN"];
  const session = new Session({ oauthToken: authToken });
  const client = session.client(serviceClients.ClusterServiceClient);

  const result = await client.streamLogs(
    StreamClusterLogsRequest.fromPartial({
      clusterId: "clusterId",
      // columnFilter: ["columnFilter"],
      serviceType: StreamClusterLogsRequest_ServiceType.CLICKHOUSE,
      // fromTime: {
      // seconds: 0,
      // nanos: 0
      // },
      // toTime: {
      // seconds: 0,
      // nanos: 0
      // },
      // recordToken: "recordToken",
      // filter: "filter"
    })
  );
  console.log(result);
})();

Python

import os
import grpc
import yandexcloud

from yandex.cloud.dataproc.v1.cluster_service_pb2_grpc import ClusterServiceStub
from yandex.cloud.mdb.clickhouse.v1.cluster_service_pb2 import StreamClusterLogsRequest

token = os.getenv("YC_OAUTH_TOKEN")

sdk = yandexcloud.SDK(token=token)

service = sdk.client(ClusterServiceStub)
response = service.streamLogs()
print(response)

StreamClusterLogsRequest

ServiceType

• SERVICE_TYPE_UNSPECIFIED

• CLICKHOUSE

  Logs of ClickHouse activity.

clusterId : string

Required. ID of the ClickHouse cluster.

columnFilter : string

Columns from logs table to get in the response.

serviceType : ServiceType

fromTime : google.protobuf.Timestamp

Start timestamp for the logs request.

toTime : google.protobuf.Timestamp

End timestamp for the logs request. If this field is not set, all existing logs will be sent and then the new ones as they appear. In essence it has tail -f semantics.

recordToken : string

Record token. Set record_token to the StreamLogRecord.next_record_token returned by a previous StreamLogs request to start streaming from next log record.

filter : string

A filter expression that filters resources listed in the response. The expression must specify:

1. The field name. Currently filtering can be applied to the LogRecord.logs.message.hostname, LogRecord.logs.message.severity fields.
2. An = operator.
3. The value in double quotes ("). Must be 1-63 characters long and match the regular expression [a-z0-9.-]{1,61}. Examples of a filter:

• message.hostname='node1.db.cloud.yandex.net'
• message.severity IN ('Error', 'Fatal') AND message.hostname != 'node2.db.cloud.yandex.net'.

StreamLogRecord

record : LogRecord

One of the requested log records.

nextRecordToken : string

This token allows you to continue streaming logs starting from the exact same record. To continue streaming, specify value of [next_record_token[ as value for the StreamClusterLogsRequest.record_token parameter in the next StreamLogs request. This value is interchangeable with the ListClusterLogsResponse.next_page_token from ListLogs method.

LogRecord

timestamp : google.protobuf.Timestamp

Log record timestamp in RFC3339 text format.

message : string

Contents of the log record.

---