-
Notifications
You must be signed in to change notification settings - Fork 1
/
tasks.py
228 lines (197 loc) · 8.52 KB
/
tasks.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
"""
Celery Tasks for nodEvac
"""
import subprocess
import time
import re
from celery import Celery
from celery.decorators import task
import redis
from ganeti_utils import cluster_connection, get_node_info,\
get_cluster_info, GANETI_CLUSTER
from ipmi import get_ipmi_info
from ici import sched_downtime
celery_app = Celery('nodevac', broker="redis://localhost:6379/0")
celery_app.conf.result_backend = "redis://localhost:6379/0"
@task(bind=True)
def evacuate_node_task(self, node_name, cluster_name):
""" Handles the tasks needed to evacuate a Ganeti node."""
print("Begin node evacuation task for node " + node_name)
evac_status = {}
cluster_conn = cluster_connection(cluster_name)
node_info = get_node_info(node_name, cluster_name)
evac_status["role"] = node_info["role"]
evac_status["pinst_cnt"] = node_info["pinst_cnt"]
evac_status["inst_done"] = 0
print(evac_status)
self.update_state(state='E_STARTED',
meta={'status': evac_status})
maintenancetag_job = cluster_conn.AddNodeTags(node_name, ['maintenance'])
cluster_conn.WaitForJobCompletion(maintenancetag_job)
if evac_status["role"] != "Drained":
print('Current role is :' + evac_status["role"] + ". Draining...")
node_drain_job = cluster_conn.SetNodeRole(node_name, "drained")
node_drain_job_status = cluster_conn.WaitForJobCompletion(node_drain_job)
if node_drain_job_status:
evac_status["role"] = "Drained"
print("Node drained.")
self.update_state(state='E_DRAINED',
meta={'status': evac_status})
print("Spawning migrate jobs...")
print ("Running VMs:" + str(node_info["pinst_list"]))
# This dict will hold the task
# current status of each task
migration_progress = {}
for instance in node_info["pinst_list"]:
jobid = migrate_instance_task.apply_async(kwargs={
'instance_name': instance,
'cluster_name': cluster_name
})
migration_progress[instance] = {}
migration_progress[instance]["task_id"] = str(jobid)
self.update_state(state='E_MIGRATING',
meta={'status': evac_status,
'progress': migration_progress
}
)
while jobid.state in ["PENDING", "JSTARTED", "JPENDING"]:
time.sleep(10)
migration_progress[instance]["task_state"] = jobid.state
migration_progress[instance]["task_status"] = jobid.info
self.update_state(state='E_MIGRATING',
meta={'status': evac_status,
'progress': migration_progress
}
)
evac_status["inst_done"] += 1
print("VMs done:" + str(evac_status["inst_done"]) +
"/" + str(evac_status["pinst_cnt"]))
self.update_state(state='E_MIGRATING',
meta={'status': evac_status,
'progress': migration_progress
}
)
print("Node emptied. Continuing.")
node_offline_job = cluster_conn.SetNodeRole(node_name, "offline")
node_offline_job_status = cluster_conn.WaitForJobCompletion(node_offline_job)
if node_offline_job_status:
evac_status["role"] = "Offline"
print("Node drained.")
# Remove the redis key refering to this task, now that we are done.k
redis_conn = redis.StrictRedis()
redis_key = ('nodEvac:evacuate_node:' + cluster_name + ":" + node_name)
redis_conn.delete(redis_key)
self.update_state(state='E_OFFLINING',
meta={'status': evac_status,
'progress': migration_progress
}
)
return {'status': evac_status, 'progress': migration_progress}
@task(bind=True)
def migrate_instance_task(self, instance_name, cluster_name):
"""Migrate a Ganeti instance"""
message = "Connecting to cluster..."
self.update_state(state='JSTARTED',
meta={'percent': '10', 'status': message})
cluster_conn = cluster_connection(cluster_name)
migrate_job_id = cluster_conn.MigrateInstance(
instance_name, allow_failover=True)
migrate_job_details = cluster_conn.GetJobStatus(migrate_job_id)
message = "Job Submitted."
job_status = "Pending"
self.update_state(state='JPENDING',
meta={
'percent': '20',
'job_id': migrate_job_id,
'job_status': job_status,
'status': message,
'job_details': migrate_job_details})
migrate_job_success = cluster_conn.WaitForJobCompletion(migrate_job_id)
migrate_job_details = cluster_conn.GetJobStatus(migrate_job_id)
if migrate_job_success:
message = "Migration Complete!"
job_status = "Success"
else:
message = "Migration Failed!"
job_status = "Failure"
return {
'percent': 100,
'job_id': migrate_job_id,
'job_status': job_status,
'status': message,
'job_details': migrate_job_details
}
@task(bind=True)
def shutdown_node_task(self, node_name, cluster_name):
"""
After doing the nessecary checks, sends an ipmi command to shutdown
and loops until it's off
"""
cluster_conn = cluster_connection(cluster_name)
node_info = cluster_conn.GetNode(node_name)
if node_info["pinst_cnt"] != 0:
return 'Node is not empty. Refusing to proceed.'
ipmi_info = get_ipmi_info(node_name, cluster_name)
node_drain_job = cluster_conn.SetNodeRole(node_name, "offline")
cluster_conn.WaitForJobCompletion(node_drain_job)
ipmi_status_cmd = ("ipmitool -H " + ipmi_info["host"] + " -U " +
ipmi_info["username"] + " -I lanplus -P " +
ipmi_info["password"] + " power status")
ipmi_shut_cmd = ("ipmitool -H " + ipmi_info["host"] + " -U " +
ipmi_info["username"] + " -I lanplus -P " +
ipmi_info["password"] + " power soft")
ipmi_status_regex = "Chassis Power is ([a-z]*)\\n"
print("Sending downtime command for host.")
# Send downtime command for 3h
sched_downtime(node_name, '10800')
message = "Sending shutdown signal to host " + ipmi_info["host"] + "."
try:
subprocess.check_output(ipmi_shut_cmd, shell=True)
self.update_state(state='SHUT_SENT', meta={'message': message})
except subprocess.CalledProcessError as ipmitool_error:
print(ipmitool_error)
while True:
ipmi_status_out = subprocess.check_output(ipmi_status_cmd, shell=True)
ipmi_status = re.match(ipmi_status_regex, ipmi_status_out).group(1)
if ipmi_status == "on" or ipmi_status == "soft":
print("IPMI is still powered on." + ipmi_status_out)
time.sleep(6)
else:
print("IPMI power if off.")
break
return "IPMI host is down."
@task()
def startup_node_task(node_name, cluster_name):
"""
Send ipmi command to start a hardware node and do a ping check until we
can readd the node to the cluster
"""
cluster_conn = cluster_connection(cluster_name)
node_info = cluster_conn.GetNode(node_name)
ipmi_info = get_ipmi_info(node_name, cluster_name)
ping_cmd = "ping -c 1 -w 1 " + node_name
print("Sending wakeup signal to host " + ipmi_info["host"] + ".")
ipmi_up_cmd = ("ipmitool -H " + ipmi_info["host"] + " -U " +
ipmi_info["username"] + " -I lanplus -P " +
ipmi_info["password"] + " power up")
print(ipmi_up_cmd)
ipmi_up_out = subprocess.check_output(ipmi_up_cmd, shell=True)
print(ipmi_up_out)
while True:
try:
time.sleep(20)
subprocess.check_output(ping_cmd, shell=True)
except subprocess.CalledProcessError:
print("No response yet. Wating 20 sec and retrying.")
else:
print("Ping check sucessful. Host is online.")
break
time.sleep(20)
node_readd_job = cluster_conn.SetNodeRole(node_name, "regular")
node_readd_job_status = cluster_conn.WaitForJobCompletion(node_readd_job)
maintenancetag_job = cluster_conn.DeleteNodeTags(node_name, ['maintenance'])
cluster_conn.WaitForJobCompletion(maintenancetag_job)
if node_readd_job_status:
return "Host is up."
else:
return "Host is up but coundn't read to cluster."