sheer-heart-attack

README

Sheer Heart Attack

sheer-heart-attack is a debugging tool that can execute any command on process/host metrics trigger.

Features

• Easy to use (just execute sheer-heart-attack launch).
• Track process and/or host metrics. and execute specified command when threshold is exceeded.
• Record the STDOUT and STDERR of the executed command in the structured log.
• Slack notification.

Quick Start

This is the recommended usage.

root@kilr_q:~# source <(curl -sL https://git.io/sheer-heart-attack)
You can use `sheer-heart-attack` command in this session.
root@kilr_q:~# sheer-heart-attack launch

In the case of fish 🐟

root@kilr_q:~# curl -sL https://git.io/sheer-heart-attack-fish | source
You can use `sheer-heart-attack` command in this session.
root@kilr_q:~# sheer-heart-attack launch

Install

manually:

Download binany from releases page

go get:

go get github.com/k1LoW/sheer-heart-attack

Usage

Just execute sheer-heart-attack launch.

Launch Options (Flags):

Option	Default	Purpose
pid		PID of the process.
threshold	cpu > 5 || mem > 10	Threshold conditions.
interval	5	Interval of checking if the threshold exceeded (seconds).
attempts	1	Maximum number of attempts continuously exceeding the threshold.
command		Command to execute when the maximum number of attempts is exceeded.
times	1	Maximum number of command executions. If times < 1, track and execute until timeout.
timeout	86400	Timeout of tracking (seconds).
slack-channel		Slack channel to notify.
slack-mention		Slack mention. (@here or user_id @UXXXXXXXXX)

Set threshold using operators

The following operators can be used to set the threshold:

+, -, *, /, ==, !=, <, >, <=, >=, not, and, or, !, &&, ||

For example, you can set the threshold as follows

• cpu > 10 and mem > 20
• (user + system) > 50 || iowait > 50
• load1 > 5 or load15 > 2

Slack Notification

sheer-heart-attack find and use Slack Incomming Webhook URL via envirionment variables ( SLACK_INCOMMING_WEBHOOK_URL, SLACK_WEBHOOK_URL, SLACK_URL )

Support Metrics

Metric
cpu	Percentage of the CPU time the process uses (percent).
mem	Percentage of the total RAM the process uses (percent).
rss	Non-swapped physical memory the process uses (bytes).
vms	Amount of virtual memory the process uses (bytes).
swap	Amount of memory that has been swapped out to disk the process uses (bytes).
connections	Amount of connections(TCP, UDP or UNIX) the process uses.
open_files	Amount of files and file discripters opend by the process. linux only
host_cpu	Percentage of cpu used.
host_mem	Percentage of RAM used.
host_swap	Amount of memory that has been swapped out to disk (bytes).
user	Percentage of CPU utilization that occurred while executing at the user level.
system	Percentage of CPU utilization that occurred while executing at the system level.
idle	Percentage of time that CPUs were idle and the system did not have an outstanding disk I/O request.
nice	Percentage of CPU utilization that occurred while executing at the user level with nice priority.
iowait	Percentage of time that CPUs were idle during which the system had an outstanding disk I/O request.
irq	Percentage of time spent by CPUs to service hardware interrupts.
softirq	Percentage of time spent by CPUs to service software interrupts.
steal	Percentage of time spent in involuntary wait by the virtual CPUs while the hypervisor was servicing another virtual processor.
guest	Percentage of time spent by CPUs to run a virtual processor.
guest_nice	Percentage of time spent by CPUs to run a virtual processor with nice priority.
load1	Load avarage for 1 minute.
load5	Load avarage for 5 minutes.
load15	Load avarage for 15 minutes.

Alternatives

• ProcDump for Linux - A Linux version of the ProcDump Sysinternals tool