python 使用Boto自动标记的AWS EC2竞价请求
Posted
tags:
篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了python 使用Boto自动标记的AWS EC2竞价请求相关的知识,希望对你有一定的参考价值。
import argparse
import json
import time
import boto.ec2
# Set up argument parser
parser = argparse.ArgumentParser(
description='Request AWS EC2 spot instance and tag instance and volumes.',
formatter_class=argparse.RawTextHelpFormatter)
parser.add_argument(
'region',
help='EC2 region to connect to (e.g. us-west-2).')
parser.add_argument(
'configfile', type=argparse.FileType(),
help='JSON file containing configuration for spot request. It \n'
'should contain one dictionary with the following\n'
'elements:\n'
'tags: dictionary of key-value pairs\n'
'spot-request: dictionary of arguments to\n'
' boto.ec2.connection.EC2Connection.\n'
' request_spot_instances()\n'
'block-device-mapping: dictionary with block device names\n'
" (e.g. '/dev/sda1') as keys and\n"
' dictionaries of arguments to\n'
' boto.ec2.blockdevicemapping.\n'
' BlockDeviceType as values.')
parser.add_argument(
'-p', '--profile', default=None,
help='Use profile PROFILE specified ~/.boto or\n'
'~/.aws/credentials.\n'
"(Default: Boto will look in the usual places for the\n"
'default credentials.)')
parser.add_argument(
'-s', '--sleep', default=3, type=int,
help='How often to check for the status, in seconds.\n'
'(Default: %(default)s)')
args = parser.parse_args()
# Load config from json file
config = json.load(args.configfile)
args.configfile.close()
# Open EC2 connection
ec2_conn = boto.ec2.connect_to_region(args.region,
profile_name=args.profile)
# Configure block device mapping
if 'block-device-mapping' in config:
bdm = boto.ec2.blockdevicemapping.BlockDeviceMapping()
for name, bd in config['block-device-mapping'].iteritems():
bdm[name] = boto.ec2.blockdevicemapping.BlockDeviceType(**bd)
else:
bdm = None
# Request spot instance
spot_req = ec2_conn.request_spot_instances(block_device_map=bdm,
**config['spot-request'])[0]
# Tag the request, once we can get a valid request ID.
print('Tagging spot request.')
while True:
try:
ec2_conn.create_tags(spot_req.id, config['tags'])
except:
pass
else:
break
# Wait while the spot request remains open
state = 'open'
while state == 'open':
time.sleep(args.sleep)
spot = ec2_conn.get_all_spot_instance_requests(spot_req.id)[0]
state = spot.state
print('Spot request ' + spot.id + ' status: ' + spot.status.code + ': ' +
spot.status.message)
# Exit if there is an error
if (state != 'active'):
exit(1)
# List of resources to tag
ids_to_tag = [spot.instance_id]
# Get the instance that was just launched, waiting until the number of block
# devices attached to the instance is at least as many as requested.
bd_count = 0
while bd_count < len(bdm):
time.sleep(args.sleep)
instance = ec2_conn.get_only_instances(spot.instance_id)[0]
bd_count = len(instance.block_device_mapping)
# Get block devices to tag
for bd in instance.block_device_mapping.itervalues():
ids_to_tag.append(bd.volume_id)
# Tag resources
print('Tagging instance and attached volumes.')
ec2_conn.create_tags(ids_to_tag, config['tags'])
# Wait till instance is out of pending state
while instance.state == 'pending':
time.sleep(args.sleep)
instance = ec2_conn.get_only_instances(spot.instance_id)[0]
print('Instance ' + instance.id + ' state: ' + instance.state)
以上是关于python 使用Boto自动标记的AWS EC2竞价请求的主要内容,如果未能解决你的问题,请参考以下文章
使用boto3 Python SDK返回AWS EC2可用性区域属性
使用 python boto 创建和下载 AWS ec2 密钥对
使用 boto3 和 python 将 AWS EC2 详细信息导出到 xlsx/csv