I just want to run some things past you guys, to make sure I'm doing it right.
I'm using Python to parse disk metrics out of iostat output. The device lines
look like this:
Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz
avgqu-sz await svctm %util
sda 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00
My goal is JSON output for each metric that look like the below (this is for
InfluxDB):
{
"measurement": "read_requests",
"tags": {
"project": "SOME_PROJECT",
"hostname": "server1.newyork.com",
},
"time": timestamp.isoformat(),
"fields": {
"value": 0.00
}
}
To create the above, I am using the following code:
disk_stat_headers = ['device', 'read_requests_merged',
'write_requests_merged', 'read_requests', 'write_requests', 'read_sectors',
'write_sectors', 'average_request_size', 'average_queue_length',
'average_wait', 'average_service_time', 'utilisation']
...<code to iterate over lines>...
elif i >= 5 and line:
disk_stats = {}
device = line.split()[0]
disk_stats[device] = dict(zip(disk_stat_headers, line.split()[1:]))
json_points = []
for disk_name, metrics in disk_stats.items():
print(disk_name)
print(metrics)
for key, value in metrics.items():
json_points.append({
"measurement": key,
"tags": {
"project": project,
"hostname": hostname,
},
"time": timestamp.isoformat(),
"fields": {
"value": value
}
})
Is there any issue with the above? Or can you see a better way to do this?
(I'm calling split() twice, not sure if that's worse than storing it in a
variable)
Second question - the timestamps in isotat are timezone-naive. I'm using the
below to add the right timezone (EDT in this case) to them, which depends on
the pytz library:
from pytz import timezone
eastern = timezone('US/Eastern')
timestamp = eastern.localize(line)
Is the above an appropriate way of doing this? Or is there an easier way just
using the Python stdlib's?
--
https://mail.python.org/mailman/listinfo/python-list
