Python lib that provides an easy way to check the health of an *nix OS
OS-health-check has been designed to provide a quick way for monitoring the health of a set of *nix based OS.
Doesn't need any agent on each node to be monitorized, just share pubkey based credentials thorugh SSH
Takes care about:
- CPU
- Mem
- Swap
- Disks (ocupation and inodes)
- Processes (creation)
- Network I/O, transfer stats
- All block dev
- Interrupts
Just call it
$ os-health-check.py -h HOST -c CHECKSlist
, and reach a status line for each check
check1 status detail
check2 status detail
...
checkn status detail
, where
-H define the hostname/IP address to check
-C the set of checks to be performed comma (,) separated (without spaces!)
OPTIONS: cpu mem disk net
-m set the metrics pair warning,critical to analyze the status of the tests comma (,) separated (without spaces!)
//-m 75,90 will set status = warning if range >75 and critial if >90
-d define a delay for the execution of the checks
//-d5 will wait 5sec and trigger al the checks
-D define a delay for the execution of **each** check
//-D5 will wait 5secs between the execution of each check
-c get the results from cache (last execution results)
$ os-health-check.py -h GNUites -c cpu,disk,mem,swap -m 75,90
cpu ok "la 1,0,0"
disk ok "all devices ok"
mem warning "85% used"
swap ok ""