Skip to content

Instantly share code, notes, and snippets.

@dpryan79
Last active February 16, 2017 21:15
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save dpryan79/e30fb98513b8da5d5140d3eb7fe49501 to your computer and use it in GitHub Desktop.
Save dpryan79/e30fb98513b8da5d5140d3eb7fe49501 to your computer and use it in GitHub Desktop.
This is an only partially tested script to check that various NFS servers are still working and post messages to slack accordingly.
#!/usr/bin/env python3
import urllib.request
import json
from subprocess import Popen
from time import sleep
def checkUp(path, timeout):
p = Popen(['ls', path])
try:
rc = p.wait(timeout)
except:
return False
if rc is None:
# Still running! Kill it!
p.terminate()
return False
if rc < 0:
return False
return True
def sendMessage(content):
# CHANGE THIS TO SOMETHING APPROPRIATE!
url = "https://hooks.slack.com/services/T000000000/B000000000/XXXXXXXXXXXXXXXXXX"
data = json.dumps({"text": content,
"username": "nfs-bot",
"channel": "#general"})
opener = urllib.request.build_opener()
req = urllib.request.Request(url, data=bytes(data, "ascii"), headers={'Content-Type': 'application/json'})
resp = opener.open(req)
def iterateStatus(sleepTime, timeout):
"""
A list of NFS servers with a mount point associated with each.
The 3rd field is a placeholder for the state, which is
initially 'None', but can change to 'up' or 'down'.
"""
l = [["Void", "/data/processing3", None],
["Valid", "/data/galaxy2", None],
["solsys4", "/data/processing", None]]
while True:
for _ in l:
if checkUp(_[1], timeout) is False:
# NFS is down, only send a message if it was up/unknown
if _[2] != "down":
_[2] = "down"
sendMessage("{} is down, time for beer!".format(_[0]))
else:
# NFS is up, don't post if this was already the case
if _[2] is None:
_[2] = "up"
sendMessage("Monitoring NFS server {} according to accessibility of {}.".format(_[0], _[1]))
elif _[2] == "down":
_[2] = "up"
sendMessage("{} is risen!".format(_[0]))
sleep(sleepTime)
# Check every 5 minutes (60*5), wait up to 15 seconds for a mount point to respond
iterateStatus(60*5, 15)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment