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
|
# This file is part of cloud-init. See LICENSE file for license information.
import calendar
import sys
from datetime import datetime, timezone
from typing import IO, Any, Dict, List, Optional, TextIO, Tuple
from cloudinit import atomic_helper, subp, util
stage_to_description: Dict[str, str] = {
"finished": "finished running cloud-init",
"init-local": "starting search for local datasources",
"init-network": "searching for network datasources",
"init": "searching for network datasources",
"modules-config": "running config modules",
"modules-final": "finalizing modules",
"modules": "running modules for",
"single": "running single module ",
}
# logger's asctime format
CLOUD_INIT_ASCTIME_FMT = "%Y-%m-%d %H:%M:%S,%f"
# journctl -o short-precise
CLOUD_INIT_JOURNALCTL_FMT = "%b %d %H:%M:%S.%f %Y"
# other
DEFAULT_FMT = "%b %d %H:%M:%S %Y"
def parse_timestamp(timestampstr: str) -> float:
# default syslog time does not include the current year
months = [calendar.month_abbr[m] for m in range(1, 13)]
if timestampstr.split()[0] in months:
# Aug 29 22:55:26
FMT = DEFAULT_FMT
if "." in timestampstr:
FMT = CLOUD_INIT_JOURNALCTL_FMT
dt = datetime.strptime(
timestampstr + " " + str(datetime.now().year),
FMT,
).replace(tzinfo=timezone.utc)
timestamp = dt.timestamp()
elif "," in timestampstr:
# 2016-09-12 14:39:20,839
dt = datetime.strptime(timestampstr, CLOUD_INIT_ASCTIME_FMT).replace(
tzinfo=timezone.utc
)
timestamp = dt.timestamp()
else:
# allow GNU date(1) to handle other formats we don't expect
# This may throw a ValueError if no GNU date can be found
timestamp = parse_timestamp_from_date(timestampstr)
return float(timestamp)
def has_gnu_date() -> bool:
"""GNU date includes a string containing the word GNU in it in
help output. Posix date does not. Use this to indicate on Linux
systems without GNU date that the extended parsing is not
available.
"""
return "GNU" in subp.subp(["date", "--help"]).stdout
def parse_timestamp_from_date(timestampstr: str) -> float:
if not util.is_Linux() and subp.which("gdate"):
date = "gdate"
elif has_gnu_date():
date = "date"
else:
raise ValueError(
f"Unable to parse timestamp without GNU date: [{timestampstr}]"
)
return float(
subp.subp([date, "-u", "+%s.%3N", "-d", timestampstr]).stdout.strip()
)
def parse_ci_logline(line: str) -> Optional[Dict[str, Any]]:
# Stage Starts:
# Cloud-init v. 0.7.7 running 'init-local' at \
# Fri, 02 Sep 2016 19:28:07 +0000. Up 1.0 seconds.
# Cloud-init v. 0.7.7 running 'init' at \
# Fri, 02 Sep 2016 19:28:08 +0000. Up 2.0 seconds.
# Cloud-init v. 0.7.7 finished at
# Aug 29 22:55:26 test1 [CLOUDINIT] handlers.py[DEBUG]: \
# finish: modules-final: SUCCESS: running modules for final
# 2016-08-30T21:53:25.972325+00:00 y1 [CLOUDINIT] handlers.py[DEBUG]: \
# finish: modules-final: SUCCESS: running modules for final
#
# Nov 03 06:51:06.074410 x2 cloud-init[106]: [CLOUDINIT] util.py[DEBUG]: \
# Cloud-init v. 0.7.8 running 'init-local' at \
# Thu, 03 Nov 2016 06:51:06 +0000. Up 1.0 seconds.
#
# 2017-05-22 18:02:01,088 - util.py[DEBUG]: Cloud-init v. 0.7.9 running \
# 'init-local' at Mon, 22 May 2017 18:02:01 +0000. Up 2.0 seconds.
#
# Apr 30 19:39:11 cloud-init[2673]: handlers.py[DEBUG]: start: \
# init-local/check-cache: attempting to read from cache [check]
amazon_linux_2_sep = " cloud-init["
separators = [" - ", " [CLOUDINIT] ", amazon_linux_2_sep]
found = False
for sep in separators:
if sep in line:
found = True
break
if not found:
return None
(timehost, eventstr) = line.split(sep)
# journalctl -o short-precise
if timehost.endswith(":"):
timehost = " ".join(timehost.split()[0:-1])
if "," in timehost:
timestampstr, extra = timehost.split(",")
timestampstr += ",%s" % extra.split()[0]
if " " in extra:
hostname = extra.split()[-1]
else:
hostname = timehost.split()[-1]
if sep == amazon_linux_2_sep:
# This is an Amazon Linux style line, with no hostname and a PID.
# Use the whole of timehost as timestampstr, and strip off the PID
# from the start of eventstr.
timestampstr = timehost.strip()
eventstr = eventstr.split(maxsplit=1)[1]
else:
timestampstr = timehost.split(hostname)[0].strip()
if "Cloud-init v." in eventstr:
event_type = "start"
if "running" in eventstr:
stage_and_timestamp = eventstr.split("running")[1].lstrip()
event_name, _ = stage_and_timestamp.split(" at ")
event_name = event_name.replace("'", "").replace(":", "-")
if event_name == "init":
event_name = "init-network"
else:
# don't generate a start for the 'finished at' banner
return None
event_description = stage_to_description[event_name]
else:
(_pymodloglvl, event_type, event_name) = eventstr.split()[0:3]
event_description = eventstr.split(event_name)[1].strip()
event = {
"name": event_name.rstrip(":"),
"description": event_description,
"timestamp": parse_timestamp(timestampstr),
"origin": "cloudinit",
"event_type": event_type.rstrip(":"),
}
if event["event_type"] == "finish":
result = event_description.split(":")[0]
desc = event_description.split(result)[1].lstrip(":").strip()
event["result"] = result
event["description"] = desc.strip()
return event
def dump_events(
cisource: Optional[IO[str]] = None,
rawdata: Optional[str] = None,
) -> Tuple[List[Dict[str, Any]], List[str]]:
events = []
event = None
CI_EVENT_MATCHES = ["start:", "finish:", "Cloud-init v."]
if not any([cisource, rawdata]):
raise ValueError("Either cisource or rawdata parameters are required")
if rawdata is not None:
data = rawdata.splitlines()
elif cisource is not None:
data = cisource.readlines()
else:
raise ValueError(
"Either cisource or rawdata parameters must have a value"
)
for line in data:
for match in CI_EVENT_MATCHES:
if match in line:
try:
event = parse_ci_logline(line)
except ValueError:
sys.stderr.write("Skipping invalid entry\n")
if event:
events.append(event)
return events, data
def main() -> str:
if len(sys.argv) > 1:
cisource: TextIO = open(sys.argv[1])
else:
cisource = sys.stdin
return atomic_helper.json_dumps(dump_events(cisource))
if __name__ == "__main__":
print(main())
|
