Martin Blanchard pushed to branch mablanch/83-executed-action-metadata at BuildGrid / buildgrid
Commits:
-
79cd4f43
by Martin Blanchard at 2018-10-26T10:41:26Z
-
687469de
by Martin Blanchard at 2018-10-26T10:48:44Z
4 changed files:
- buildgrid/_app/commands/cmd_operation.py
- + buildgrid/_enums.py
- buildgrid/bot/bot_session.py
- buildgrid/server/job.py
Changes:
... | ... | @@ -20,15 +20,21 @@ Operations command |
20 | 20 |
Check the status of operations
|
21 | 21 |
"""
|
22 | 22 |
|
23 |
+from collections import OrderedDict
|
|
23 | 24 |
import logging
|
25 |
+from operator import attrgetter
|
|
24 | 26 |
from urllib.parse import urlparse
|
25 | 27 |
import sys
|
28 |
+from textwrap import indent
|
|
26 | 29 |
|
27 | 30 |
import click
|
31 |
+from google.protobuf import json_format
|
|
28 | 32 |
import grpc
|
29 | 33 |
|
34 |
+from buildgrid._enums import OperationStage
|
|
30 | 35 |
from buildgrid._protos.build.bazel.remote.execution.v2 import remote_execution_pb2, remote_execution_pb2_grpc
|
31 | 36 |
from buildgrid._protos.google.longrunning import operations_pb2, operations_pb2_grpc
|
37 |
+from buildgrid._protos.google.rpc import code_pb2
|
|
32 | 38 |
|
33 | 39 |
from ..cli import pass_context
|
34 | 40 |
|
... | ... | @@ -65,45 +71,145 @@ def cli(context, remote, instance_name, client_key, client_cert, server_cert): |
65 | 71 |
context.logger.debug("Starting for remote {}".format(context.remote))
|
66 | 72 |
|
67 | 73 |
|
74 |
+def _print_operation_status(operation, print_details=False):
|
|
75 |
+ metadata = remote_execution_pb2.ExecuteOperationMetadata()
|
|
76 |
+ # The metadata is expected to be an ExecuteOperationMetadata message:
|
|
77 |
+ assert operation.metadata.Is(metadata.DESCRIPTOR)
|
|
78 |
+ operation.metadata.Unpack(metadata)
|
|
79 |
+ |
|
80 |
+ stage = OperationStage(metadata.stage)
|
|
81 |
+ |
|
82 |
+ if not operation.done:
|
|
83 |
+ if stage == OperationStage.CACHE_CHECK:
|
|
84 |
+ click.echo('CacheCheck: {}: Querying action-cache (stage={})'
|
|
85 |
+ .format(operation.name, metadata.stage))
|
|
86 |
+ elif stage == OperationStage.QUEUED:
|
|
87 |
+ click.echo('Queued: {}: Waiting for execution (stage={})'
|
|
88 |
+ .format(operation.name, metadata.stage))
|
|
89 |
+ elif stage == OperationStage.EXECUTING:
|
|
90 |
+ click.echo('Executing: {}: Currently running (stage={})'
|
|
91 |
+ .format(operation.name, metadata.stage))
|
|
92 |
+ else:
|
|
93 |
+ click.echo('Error: {}: In an invalid state (stage={})'
|
|
94 |
+ .format(operation.name, metadata.stage), err=True)
|
|
95 |
+ return
|
|
96 |
+ |
|
97 |
+ assert stage == OperationStage.COMPLETED
|
|
98 |
+ |
|
99 |
+ response = remote_execution_pb2.ExecuteResponse()
|
|
100 |
+ # The response is expected to be an ExecutionResponse message:
|
|
101 |
+ assert operation.response.Is(response.DESCRIPTOR)
|
|
102 |
+ operation.response.Unpack(response)
|
|
103 |
+ |
|
104 |
+ if response.status.code != code_pb2.OK:
|
|
105 |
+ click.echo('Failure: {}: {} (code={})'
|
|
106 |
+ .format(operation.name, response.status.message, response.status.code))
|
|
107 |
+ else:
|
|
108 |
+ if response.result.exit_code != 0:
|
|
109 |
+ click.echo('Success: {}: Completed with failure (stage={}, exit_code={})'
|
|
110 |
+ .format(operation.name, metadata.stage, response.result.exit_code))
|
|
111 |
+ else:
|
|
112 |
+ click.echo('Success: {}: Completed succesfully (stage={}, exit_code={})'
|
|
113 |
+ .format(operation.name, metadata.stage, response.result.exit_code))
|
|
114 |
+ |
|
115 |
+ if print_details:
|
|
116 |
+ metadata = response.result.execution_metadata
|
|
117 |
+ click.echo(indent('worker={}'.format(metadata.worker), ' '))
|
|
118 |
+ |
|
119 |
+ queued = metadata.queued_timestamp.ToDatetime()
|
|
120 |
+ click.echo(indent('queued_at={}'.format(queued), ' '))
|
|
121 |
+ |
|
122 |
+ worker_start = metadata.worker_start_timestamp.ToDatetime()
|
|
123 |
+ worker_completed = metadata.worker_completed_timestamp.ToDatetime()
|
|
124 |
+ click.echo(indent('work_duration={}'.format(worker_completed - worker_start), ' '))
|
|
125 |
+ |
|
126 |
+ fetch_start = metadata.input_fetch_start_timestamp.ToDatetime()
|
|
127 |
+ fetch_completed = metadata.input_fetch_completed_timestamp.ToDatetime()
|
|
128 |
+ click.echo(indent('fetch_duration={}'.format(fetch_completed - fetch_start), ' '))
|
|
129 |
+ |
|
130 |
+ execution_start = metadata.execution_start_timestamp.ToDatetime()
|
|
131 |
+ execution_completed = metadata.execution_completed_timestamp.ToDatetime()
|
|
132 |
+ click.echo(indent('exection_duration={}'.format(execution_completed - execution_start), ' '))
|
|
133 |
+ |
|
134 |
+ upload_start = metadata.output_upload_start_timestamp.ToDatetime()
|
|
135 |
+ upload_completed = metadata.output_upload_completed_timestamp.ToDatetime()
|
|
136 |
+ click.echo(indent('upload_duration={}'.format(upload_completed - upload_start), ' '))
|
|
137 |
+ |
|
138 |
+ click.echo(indent('total_duration={}'.format(worker_completed - queued), ' '))
|
|
139 |
+ |
|
140 |
+ |
|
68 | 141 |
@cli.command('status', short_help="Get the status of an operation.")
|
69 | 142 |
@click.argument('operation-name', nargs=1, type=click.STRING, required=True)
|
143 |
+@click.option('--json', is_flag=True, show_default=True,
|
|
144 |
+ help="Print operations status in JSON format.")
|
|
70 | 145 |
@pass_context
|
71 |
-def status(context, operation_name):
|
|
72 |
- context.logger.info("Getting operation status...")
|
|
146 |
+def status(context, operation_name, json):
|
|
73 | 147 |
stub = operations_pb2_grpc.OperationsStub(context.channel)
|
74 |
- |
|
75 | 148 |
request = operations_pb2.GetOperationRequest(name=operation_name)
|
76 | 149 |
|
77 |
- response = stub.GetOperation(request)
|
|
78 |
- context.logger.info(response)
|
|
150 |
+ operation = stub.GetOperation(request)
|
|
151 |
+ |
|
152 |
+ if not json:
|
|
153 |
+ _print_operation_status(operation, print_details=True)
|
|
154 |
+ else:
|
|
155 |
+ click.echo(json_format.MessageToJson(operation))
|
|
79 | 156 |
|
80 | 157 |
|
81 | 158 |
@cli.command('list', short_help="List operations.")
|
159 |
+@click.option('--json', is_flag=True, show_default=True,
|
|
160 |
+ help="Print operations list in JSON format.")
|
|
82 | 161 |
@pass_context
|
83 |
-def lists(context):
|
|
84 |
- context.logger.info("Getting list of operations")
|
|
162 |
+def lists(context, json):
|
|
85 | 163 |
stub = operations_pb2_grpc.OperationsStub(context.channel)
|
86 |
- |
|
87 | 164 |
request = operations_pb2.ListOperationsRequest(name=context.instance_name)
|
88 | 165 |
|
89 | 166 |
response = stub.ListOperations(request)
|
90 | 167 |
|
91 | 168 |
if not response.operations:
|
92 |
- context.logger.warning("No operations to list")
|
|
169 |
+ click.echo('Error: No operations to list.', err=True)
|
|
93 | 170 |
return
|
94 | 171 |
|
95 |
- for op in response.operations:
|
|
96 |
- context.logger.info(op)
|
|
172 |
+ operations_map = OrderedDict([
|
|
173 |
+ (OperationStage.CACHE_CHECK, []),
|
|
174 |
+ (OperationStage.QUEUED, []),
|
|
175 |
+ (OperationStage.EXECUTING, []),
|
|
176 |
+ (OperationStage.COMPLETED, [])
|
|
177 |
+ ])
|
|
178 |
+ |
|
179 |
+ for operation in response.operations:
|
|
180 |
+ metadata = remote_execution_pb2.ExecuteOperationMetadata()
|
|
181 |
+ # The metadata is expected to be an ExecuteOperationMetadata message:
|
|
182 |
+ assert operation.metadata.Is(metadata.DESCRIPTOR)
|
|
183 |
+ operation.metadata.Unpack(metadata)
|
|
184 |
+ |
|
185 |
+ stage = OperationStage(metadata.stage)
|
|
186 |
+ |
|
187 |
+ operations_map[stage].append(operation)
|
|
188 |
+ |
|
189 |
+ for operations in operations_map.values():
|
|
190 |
+ operations.sort(key=attrgetter('name'))
|
|
191 |
+ for operation in operations:
|
|
192 |
+ if not json:
|
|
193 |
+ _print_operation_status(operation)
|
|
194 |
+ else:
|
|
195 |
+ click.echo(json_format.MessageToJson(operation))
|
|
97 | 196 |
|
98 | 197 |
|
99 | 198 |
@cli.command('wait', short_help="Streams an operation until it is complete.")
|
100 | 199 |
@click.argument('operation-name', nargs=1, type=click.STRING, required=True)
|
200 |
+@click.option('--json', is_flag=True, show_default=True,
|
|
201 |
+ help="Print operations statuses in JSON format.")
|
|
101 | 202 |
@pass_context
|
102 |
-def wait(context, operation_name):
|
|
203 |
+def wait(context, operation_name, json):
|
|
103 | 204 |
stub = remote_execution_pb2_grpc.ExecutionStub(context.channel)
|
104 | 205 |
request = remote_execution_pb2.WaitExecutionRequest(name=operation_name)
|
105 | 206 |
|
106 |
- response = stub.WaitExecution(request)
|
|
207 |
+ operation_iterator = stub.WaitExecution(request)
|
|
107 | 208 |
|
108 |
- for stream in response:
|
|
109 |
- context.logger.info(stream)
|
|
209 |
+ for operation in operation_iterator:
|
|
210 |
+ if not json and operation.done:
|
|
211 |
+ _print_operation_status(operation, print_details=True)
|
|
212 |
+ elif not json:
|
|
213 |
+ _print_operation_status(operation)
|
|
214 |
+ else:
|
|
215 |
+ click.echo(json_format.MessageToJson(operation))
|
1 |
+# Copyright (C) 2018 Bloomberg LP
|
|
2 |
+#
|
|
3 |
+# Licensed under the Apache License, Version 2.0 (the "License");
|
|
4 |
+# you may not use this file except in compliance with the License.
|
|
5 |
+# You may obtain a copy of the License at
|
|
6 |
+#
|
|
7 |
+# <http://www.apache.org/licenses/LICENSE-2.0>
|
|
8 |
+#
|
|
9 |
+# Unless required by applicable law or agreed to in writing, software
|
|
10 |
+# distributed under the License is distributed on an "AS IS" BASIS,
|
|
11 |
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
12 |
+# See the License for the specific language governing permissions and
|
|
13 |
+# limitations under the License.
|
|
14 |
+ |
|
15 |
+ |
|
16 |
+from enum import Enum
|
|
17 |
+ |
|
18 |
+from buildgrid._protos.build.bazel.remote.execution.v2 import remote_execution_pb2
|
|
19 |
+from buildgrid._protos.google.devtools.remoteworkers.v1test2 import bots_pb2
|
|
20 |
+ |
|
21 |
+ |
|
22 |
+class BotStatus(Enum):
|
|
23 |
+ # Initially unknown state.
|
|
24 |
+ BOT_STATUS_UNSPECIFIED = bots_pb2.BotStatus.Value('BOT_STATUS_UNSPECIFIED')
|
|
25 |
+ # The bot is healthy, and will accept leases as normal.
|
|
26 |
+ OK = bots_pb2.BotStatus.Value('OK')
|
|
27 |
+ # The bot is unhealthy and will not accept new leases.
|
|
28 |
+ UNHEALTHY = bots_pb2.BotStatus.Value('UNHEALTHY')
|
|
29 |
+ # The bot has been asked to reboot the host.
|
|
30 |
+ HOST_REBOOTING = bots_pb2.BotStatus.Value('HOST_REBOOTING')
|
|
31 |
+ # The bot has been asked to shut down.
|
|
32 |
+ BOT_TERMINATING = bots_pb2.BotStatus.Value('BOT_TERMINATING')
|
|
33 |
+ |
|
34 |
+ |
|
35 |
+class LeaseState(Enum):
|
|
36 |
+ # Initially unknown state.
|
|
37 |
+ LEASE_STATE_UNSPECIFIED = bots_pb2.LeaseState.Value('LEASE_STATE_UNSPECIFIED')
|
|
38 |
+ # The server expects the bot to accept this lease.
|
|
39 |
+ PENDING = bots_pb2.LeaseState.Value('PENDING')
|
|
40 |
+ # The bot has accepted this lease.
|
|
41 |
+ ACTIVE = bots_pb2.LeaseState.Value('ACTIVE')
|
|
42 |
+ # The bot is no longer leased.
|
|
43 |
+ COMPLETED = bots_pb2.LeaseState.Value('COMPLETED')
|
|
44 |
+ # The bot should immediately release all resources associated with the lease.
|
|
45 |
+ CANCELLED = bots_pb2.LeaseState.Value('CANCELLED')
|
|
46 |
+ |
|
47 |
+ |
|
48 |
+class OperationStage(Enum):
|
|
49 |
+ # Initially unknown stage.
|
|
50 |
+ UNKNOWN = remote_execution_pb2.ExecuteOperationMetadata.Stage.Value('UNKNOWN')
|
|
51 |
+ # Checking the result against the cache.
|
|
52 |
+ CACHE_CHECK = remote_execution_pb2.ExecuteOperationMetadata.Stage.Value('CACHE_CHECK')
|
|
53 |
+ # Currently idle, awaiting a free machine to execute.
|
|
54 |
+ QUEUED = remote_execution_pb2.ExecuteOperationMetadata.Stage.Value('QUEUED')
|
|
55 |
+ # Currently being executed by a worker.
|
|
56 |
+ EXECUTING = remote_execution_pb2.ExecuteOperationMetadata.Stage.Value('EXECUTING')
|
|
57 |
+ # Finished execution.
|
|
58 |
+ COMPLETED = remote_execution_pb2.ExecuteOperationMetadata.Stage.Value('COMPLETED')
|
... | ... | @@ -26,49 +26,15 @@ import asyncio |
26 | 26 |
import logging
|
27 | 27 |
import platform
|
28 | 28 |
import uuid
|
29 |
-from enum import Enum
|
|
30 | 29 |
|
31 | 30 |
import grpc
|
32 | 31 |
|
32 |
+from buildgrid._enums import BotStatus, LeaseState
|
|
33 | 33 |
from buildgrid._protos.google.rpc import code_pb2
|
34 | 34 |
from buildgrid._protos.google.devtools.remoteworkers.v1test2 import bots_pb2, worker_pb2
|
35 | 35 |
from buildgrid._exceptions import BotError
|
36 | 36 |
|
37 | 37 |
|
38 |
-class BotStatus(Enum):
|
|
39 |
- # Default value.
|
|
40 |
- BOT_STATUS_UNSPECIFIED = bots_pb2.BotStatus.Value('BOT_STATUS_UNSPECIFIED')
|
|
41 |
- |
|
42 |
- # The bot is healthy, and will accept leases as normal.
|
|
43 |
- OK = bots_pb2.BotStatus.Value('OK')
|
|
44 |
- |
|
45 |
- # The bot is unhealthy and will not accept new leases.
|
|
46 |
- UNHEALTHY = bots_pb2.BotStatus.Value('UNHEALTHY')
|
|
47 |
- |
|
48 |
- # The bot has been asked to reboot the host.
|
|
49 |
- HOST_REBOOTING = bots_pb2.BotStatus.Value('HOST_REBOOTING')
|
|
50 |
- |
|
51 |
- # The bot has been asked to shut down.
|
|
52 |
- BOT_TERMINATING = bots_pb2.BotStatus.Value('BOT_TERMINATING')
|
|
53 |
- |
|
54 |
- |
|
55 |
-class LeaseState(Enum):
|
|
56 |
- # Default value.
|
|
57 |
- LEASE_STATE_UNSPECIFIED = bots_pb2.LeaseState.Value('LEASE_STATE_UNSPECIFIED')
|
|
58 |
- |
|
59 |
- # The server expects the bot to accept this lease.
|
|
60 |
- PENDING = bots_pb2.LeaseState.Value('PENDING')
|
|
61 |
- |
|
62 |
- # The bot has accepted this lease.
|
|
63 |
- ACTIVE = bots_pb2.LeaseState.Value('ACTIVE')
|
|
64 |
- |
|
65 |
- # The bot is no longer leased.
|
|
66 |
- COMPLETED = bots_pb2.LeaseState.Value('COMPLETED')
|
|
67 |
- |
|
68 |
- # The bot should immediately release all resources associated with the lease.
|
|
69 |
- CANCELLED = bots_pb2.LeaseState.Value('CANCELLED')
|
|
70 |
- |
|
71 |
- |
|
72 | 38 |
class BotSession:
|
73 | 39 |
def __init__(self, parent, interface):
|
74 | 40 |
""" Unique bot ID within the farm used to identify this bot
|
... | ... | @@ -15,41 +15,15 @@ |
15 | 15 |
|
16 | 16 |
import logging
|
17 | 17 |
import uuid
|
18 |
-from enum import Enum
|
|
19 | 18 |
|
20 | 19 |
from google.protobuf import timestamp_pb2
|
21 | 20 |
|
21 |
+from buildgrid._enums import LeaseState, OperationStage
|
|
22 | 22 |
from buildgrid._protos.build.bazel.remote.execution.v2 import remote_execution_pb2
|
23 | 23 |
from buildgrid._protos.google.devtools.remoteworkers.v1test2 import bots_pb2
|
24 | 24 |
from buildgrid._protos.google.longrunning import operations_pb2
|
25 | 25 |
|
26 | 26 |
|
27 |
-class OperationStage(Enum):
|
|
28 |
- # Initially unknown stage.
|
|
29 |
- UNKNOWN = remote_execution_pb2.ExecuteOperationMetadata.Stage.Value('UNKNOWN')
|
|
30 |
- # Checking the result against the cache.
|
|
31 |
- CACHE_CHECK = remote_execution_pb2.ExecuteOperationMetadata.Stage.Value('CACHE_CHECK')
|
|
32 |
- # Currently idle, awaiting a free machine to execute.
|
|
33 |
- QUEUED = remote_execution_pb2.ExecuteOperationMetadata.Stage.Value('QUEUED')
|
|
34 |
- # Currently being executed by a worker.
|
|
35 |
- EXECUTING = remote_execution_pb2.ExecuteOperationMetadata.Stage.Value('EXECUTING')
|
|
36 |
- # Finished execution.
|
|
37 |
- COMPLETED = remote_execution_pb2.ExecuteOperationMetadata.Stage.Value('COMPLETED')
|
|
38 |
- |
|
39 |
- |
|
40 |
-class LeaseState(Enum):
|
|
41 |
- # Initially unknown state.
|
|
42 |
- LEASE_STATE_UNSPECIFIED = bots_pb2.LeaseState.Value('LEASE_STATE_UNSPECIFIED')
|
|
43 |
- # The server expects the bot to accept this lease.
|
|
44 |
- PENDING = bots_pb2.LeaseState.Value('PENDING')
|
|
45 |
- # The bot has accepted this lease.
|
|
46 |
- ACTIVE = bots_pb2.LeaseState.Value('ACTIVE')
|
|
47 |
- # The bot is no longer leased.
|
|
48 |
- COMPLETED = bots_pb2.LeaseState.Value('COMPLETED')
|
|
49 |
- # The bot should immediately release all resources associated with the lease.
|
|
50 |
- CANCELLED = bots_pb2.LeaseState.Value('CANCELLED')
|
|
51 |
- |
|
52 |
- |
|
53 | 27 |
class Job:
|
54 | 28 |
|
55 | 29 |
def __init__(self, action, action_digest):
|