Skip to content

Latest commit

 

History

History
104 lines (86 loc) · 4.39 KB

processes.md

File metadata and controls

104 lines (86 loc) · 4.39 KB
Raw

Processes

  • CTRL-z = Suspend current foreground job.
  • bg = Push most recently suspended job into background.
  • fg = Pull most recent background job into foreground.

  • ps -efH | less = View current running process.
    • -H = Hierarchy in tree structure.
    • -e = Everything.
    • -f = Full-format.

Signals

Name ID Hotkey Description
SIGHUP 1 Process' controlling terminal has been closed
SIGINT 2 Ctrl-c Nicely ask process to cleanup and terminate
SIGQUIT 3 Ctrl-\ Ask the process to perform a core dump
SIGKILL 9 Forcefully terminate process, cannot be ignored
SIGTERM 15 Identical to SIGINT
SIGSTP 20 Ctrl-z Ask the process to stop temporarily

Status Codes

  • D = Uninterruptible sleep (CPU waiting for I/O to complete).
  • S = Interruptible sleep (waiting for event).
  • T = Stopped by job control signal.
  • R = Running or in run queue.

Commands

  • top -u alice = Show user alice's currently running processes, use O to sort by column.
  • kill -s 9 7423 or kill -9 7423 = End process with PID 7423 by sending it a SIGKILL signal.

  • exec bash = Restart bash shell.
  • strace [command] = Trace system call.

performance-observation-tools

top Command

Upper Section

  • 15:39:37 = System time.
  • up 90 days, 15:26 = Uptime in days, hours:minutes.

  • load average: 0.00, 0.00, 0.00 = Average total system load over 1min, 5min, 15min.
  • (a value of 1 indicates one cpu core is fully occupied) (cat /proc/cpuinfo to find # of cores).
    • ex. for a single-core system -- 0.4 = Cpu at 40% capacity, 1.12 = Cpu 'overloaded' by 12% capacity.
    • ex. For a quad-core system - 1.0 = 3 cores idle, 1 core at full capacity, or all cores at 33% load (on average).
    • ex. 5.35 = System overloaded at 135% capacity, 1.35 processes were waiting for cpu time during the specified interval (1min, 5min or 15min.

  • %cpu(s): = Cpu time usage statistics, in % of total cpu time available.
    • us = % cpu time running userpace processes.
    • sy = % cpu time running kernel processes.
    • ni = % cpu time running processes with manually set nice value (lower nice value = higher priority).
    • id = % cpu time idle (likely in a power save mode).
    • wa = % time cpu waiting for I/O requests to complete (e.g. waiting for HDD to locate and read data.
    • hi = % cpu time handling hardware interrupts (keyboard & mouse events, peripherals, etc).
    • si = % cpu time handling software interrupts.
    • st = (virtualized environments) % time OS is waiting for cpu to finish executing processes on another VM (st for steal).

Lower Section

  • PID = Process ID.
  • USER = Process' 'effective' username.
  • PR & NI = Priority & nice value, a lower nice value correlates to higher priority.
  • VIRT = Total memory consumed (includes physical memory and swap).
  • RES = Physical memory consumed.
  • SHR = Memory shared with other processes.
  • S = Process state.

  • %CPU = % of non-idle cpu time spent on process.
  • %MEM = % of physical memory consumed.
  • TIME+ = Total cpu time used on process in format minutes:seconds:0.01 seconds.
  • COMMAND = Process name.

Hotkeys (case sensitive!)

  • P = Sort by %CPU column (default sort).
  • M = Sort by %MEM column.
  • N = Sort by PID column.
  • T = Sort by TIME+ column.
  • R = Reverse sort.

  • k = Specify pid to kill the specified process.
  • c = Show full process paths.
  • V = Toggle tree view.
  • O = Show search field.
  • (ex. COMMAND=audit = Filter processes with 'audit' in the COMMAND attribute)
  • (ex. !COMMAND=getty = Filter processes which do NOT have 'getty' in the COMMAND attribute)

  • Filters can be stacked via multiple searches, use = To clear all filters.

Hardware

  • lsof -u alice = List files currently open by processes (useful when unmounting a disk).
    • -u alice = Show files open by user alice.

  • lsmod = Show status of kernel modules.
  • lspci = List pci devices.
  • lsblk = List bock devices.