Exit status code 1 but work fine in commandline

, , ,
1

Hi,

I’ve recently come across a problem with one of my telegraf input.exec:

2024-01-29T16:16:30Z E! [inputs.exec] Error in plugin: exec: exit status 1 for command '/etc/telegraf/scripts/netapp/OMCL_RMN_NetApp_VOl_nfs_forecasting.py': Traceback (most recent call last):...

If I run it in root commandline, it run well:

[root@qpr-sd-qosdb02 netapp]# python OMCL_RMN_NetApp_VOl_nfs_forecasting.py
predicted_capacity,week_number=6 value=70.9999961853
predicted_capacity,week_number=7 value=70.7199954987
predicted_capacity,week_number=8 value=70.439994812
predicted_capacity,week_number=9 value=70.1599941254
predicted_capacity,week_number=10 value=69.8799934387
predicted_capacity,week_number=11 value=69.5999927521
[root@qpr-sd-qosdb02 netapp]# echo $?
0

If I run it as the telegraf user, nothing happen and I get the following output:

[root@qpr-sd-qosdb02 netapp]# telegraf OMCL_RMN_NetApp_VOl_nfs_forecasting.py --test
2024-01-29T16:26:13Z I! Starting Telegraf 1.16.3
2024-01-29T16:26:13Z I! Using config file: /etc/telegraf/telegraf.conf

what is really odd is that if I run telegraf with --test or --once, I can see the output correctly, but on the second batch, I get the previous error.

Here is the complete script that I use:

#!/usr/bin/python
import requests

host = 'http://localhost:8086'
database = 'influxdb_script'

query = 'SELECT TOP("date",5),* FROM NetApp_VOl_nfs WHERE time > now() - 10m'

url = '{}/query?db={}&q={}'.format(host, database, query)
response = requests.get(url)

if response.status_code == 200:
    data = response.json()

    if 'results' in data and 'series' in data['results'][0]:
        series = data['results'][0]['series'][0]

        week_index = series['columns'].index('week')
        used_capacity_index = series['columns'].index('used_data')

        wk = []
        uc = []

        for point in series['values']:
            wk.append(point[week_index])
            uc.append(point[used_capacity_index])


    else:
        print("Aucun resultat trouve dans la reponse.")
else:
    print("Erreur lors de la requete. Code de statut:", response.status_code)


data = {
    "week_number": [int(wk[0]), int(wk[1]), int(wk[2]), int(wk[3])],
    "used_capacity": [float(uc[0]), float(uc[1]), float(uc[2]), float(uc[3])]
}

def linear_regression(x, y):

    n = len(x)
    x_mean = sum(x) / n
    y_mean = sum(y) / n

    numerator = sum((x[i] - x_mean) * (y[i] - y_mean) for i in range(n))
    denominator = sum((x[i] - x_mean) ** 2 for i in range(n))

    slope = numerator / denominator
    intercept = y_mean - slope * x_mean

    return slope, intercept

week_numbers = sorted(data["week_number"])
used_capacity_values = data["used_capacity"]
slope, intercept = linear_regression(week_numbers, used_capacity_values)

prediction_weeks = [ week_numbers[-1]+i for i in range(1, 7)]
predicted_capacity = [slope * week + intercept for week in prediction_weeks]


predicted_lines = [
    "predicted_capacity,week_number={0} value={1}".format(prediction_weeks[i], predicted_capacity[i]) for i in range(len(prediction_weeks))
]

all_lines = predicted_lines

body = "\n".join(all_lines)

print(body)

Thank you !

2

Hello @Dmh991.2,
You can print to stdout to after the python script runs to make sure that you’re formatting the data correctly. Although its odd if you’re running --test and you see the right thing being printed.

 print(line.rstrip())
       sys.stdout.flush()

Can you run a cpu input telegraf config?