译自：
吐槽：为啥简书的 Markdown 不支持解析 html 呀，好忧伤...
OS
[Doc]TTY[Doc]OS (Operating System)[Doc]Command Line Options[Basic]Load[Point]CheckList[Basic]Index
TTY
"TTY" means "teletype", a typewriter, and "pty" is "pseudo-teletype", a pseudo typewriter. In Unix, /dev/tty* refers to any device that acts as a typewriter, such as the terminal.
You can view the currently logged in user through the w command, and you'll find a new tty every time you login to a window.
$ w 11:49:43 up 482 days, 19:38, 3 users, load average: 0.03, 0.08, 0.07USER TTY FROM LOGIN@ IDLE JCPU PCPU WHATdev pts/0 10.0.128.252 10:44 1:01m 0.09s 0.07s -bashdev pts/2 10.0.128.252 11:08 2:07 0.17s 0.14s toproot pts/3 10.0.240.2 11:43 7.00s 0.04s 0.00s w复制代码
Using the ps command to see process information, there is also information about tty:
$ ps -x PID TTY STAT TIME COMMAND 5530 ? S 0:00 sshd: dev@pts/3 5531 pts/3 Ss+ 0:00 -bash11296 ? S 0:00 sshd: dev@pts/411297 pts/4 Ss 0:00 -bash13318 pts/4 R+ 0:00 ps -x23733 ? Ssl 2:53 PM2 v1.1.2: God Daemon复制代码
If there is a ? means the process is not dependent on TTY, this process is called .
In Node.js, you can use stdio's isTTY to determine whether the current process is in a TTY (such as terminal) environment.
$ node -p -e "Boolean(process.stdout.isTTY)"true$ node -p -e "Boolean(process.stdout.isTTY)" | catfalse复制代码
OS
Through the OS module, you can get some auxiliary functions to the basic information of the current system.
| Attribute | Description |
|---|---|
| os.EOL | Return the current system's End Of Line, based on the current system |
| os.arch() | Return the CPU architecture of the current system, such as 'x86' or 'x64' |
| os.constants | Return system constants |
| os.cpus() | Return the information for each kernel of the CPU |
| os.endianness() | Return byte order of CPU, return BE if it is big endian, return LE if it is little endian. |
| os.freemem() | Return the size of the system's free memory, in bytes |
| os.homedir() | Return the root directory of the current user |
| os.hostname() | Return the hostname of the current system |
| os.loadavg() | Return load information |
| os.networkInterfaces() | Return the NIC information (similar to ifconfig) |
| os.platform() | Return the platform information specified at compile time, such as win32, linux, same as process.platform() |
| os.release() | Return the distribution version number of the operating system |
| os.tmpdir() | Return the default temporary folder of the system |
| os.totalmem() | Return the total memory size (witsame as the memory bar size) |
| os.type() | Return the name of the system according to |
| os.uptime() | Return the running time of the system, in seconds |
| os.userInfo([options]) | Return the current user information |
What is the difference between line breaks (EOL) in different operating systems?
End of line (EOL) is the same as newline, line ending and line break.
Usually composed of line feed (LF, \n) and carriage return (CR, \r). Common cases:
| Symbol | System |
|---|---|
| LF | In Unix or Unix compatible systems (GNU/Linux, AIX, Xenix, Mac OS X, ...)、BeOS、Amiga、RISC OS |
| CR+LF | MS-DOS、Microsoft Windows、Most non Unix systems |
| CR | Apple II family, Mac OS to version 9 |
If you don't understand the cross-system compatibility of EOL, you might have problems dealing with the line segmentation/row statistics of the file.
OS Constants
- Signal Constants, such as
SIGHUP,SIGKILL, etc. - POSIX Error Constants, such as
EACCES,EADDRINUSE, etc. - Windows Specific Error Constants, such as
WSAEACCES,WSAEBADF, etc. - libuv Constants, only
UV_UDP_REUSEADDR.
Path
The built-in path in Node.js is a module for handling path problems, but as we all know, the paths are irreconcilable in different operating systems.
Windows vs. POSIX
| POSIX | Value | Windows | Value |
|---|---|---|---|
| path.posix.sep | '/' | path.win32.sep | '\\' |
| path.posix.normalize('/foo/bar//baz/asdf/quux/..') | '/foo/bar/baz/asdf' | path.win32.normalize('C:\temp\\foo\bar\..\') | 'C:\\temp\\foo\\' |
| path.posix.basename('/tmp/myfile.html') | 'myfile.html' | path.win32.basename('C:\temp\myfile.html') | 'myfile.html' |
| path.posix.join('/asdf', '/test.html') | '/asdf/test.html' | path.win32.join('/asdf', '/test.html') | '\\asdf\\test.html' |
| path.posix.relative('/root/a', '/root/b') | '../b' | path.win32.relative('C:\a', 'c:\b') | '..\\b' |
| path.posix.isAbsolute('/baz/..') | true | path.win32.isAbsolute('C:\foo\..') | true |
| path.posix.delimiter | ':' | path.win32.delimiter | ',' |
| process.env.PATH | '/usr/bin:/bin' | process.env.PATH | C:\Windows\system32;C:\Program Files\node\' |
| PATH.split(path.posix.delimiter) | ['/usr/bin', '/bin'] | PATH.split(path.win32.delimiter) | ['C:\\Windows\\system32', 'C:\\Program Files\\node\\'] |
After looking at the table, you should realize that when you are under a certain platform, the path module is actually the method of the corresponding platform. For example, I uses Mac here, so:
const path = require('path');console.log(path.basename === path.posix.basename); // true复制代码 If you are on one of these platforms, but you need to deal with the path of another platform, you need to be aware of this cross platform issue.
path Object
on POSIX:
path.parse('/home/user/dir/file.txt')// Returns:// { // root : "/",// dir : "/home/user/dir",// base : "file.txt",// ext : ".txt",// name : "file"// }复制代码 ┌─────────────────────┬────────────┐│ dir │ base │├──────┬ ├──────┬─────┤│ root │ │ name │ ext │" / home/user/dir / file .txt "└──────┴──────────────┴──────┴─────┘复制代码
on Windows:
path.parse('C:\\path\\dir\\file.txt')// Returns:// { // root : "C:\\",// dir : "C:\\path\\dir",// base : "file.txt",// ext : ".txt",// name : "file"// }复制代码 ┌─────────────────────┬────────────┐│ dir │ base │├──────┬ ├──────┬─────┤│ root │ │ name │ ext │" C:\ path\dir \ file .txt "└──────┴──────────────┴──────┴─────┘复制代码
path.extname(path)
| case | return |
|---|---|
| path.extname('index.html') | '.html' |
| path.extname('index.coffee.md') | '.md' |
| path.extname('index.') | '.' |
| path.extname('index') | '' |
| path.extname('.index') | '' |
Command Line Options
Command Line Options is some documentation on the use of CLI. There are 4 main ways of using CLI:
- node [options] [v8 options] [script.js | -e "script"] [arguments]
- node debug [script.js | -e "script" | :] …
- node --v8-options
- direct start REPL environment without parameters
Options
| Parameter | Introduction |
|---|---|
| -v, --version | View the current node version |
| -h, --help | View help documentation |
| -e, --eval "script" | The parameter string is executed as code |
| -p, --print "script" | Print the return value of -e |
| -c, --check | Checking syntax does not perform |
| -i, --interactive | Open REPL mode even if stdin is not terminal |
| -r, --require module | Specify the module require before startup |
| --no-deprecation | Close the scrap module warning |
| --trace-deprecation | Print stack trace information for an obsolete module |
| --throw-deprecation | Throw errors while executing an obsolete module |
| --no-warnings | Ignore warnings (including obsolete warnings) |
| --trace-warnings | Print warning stack (including discarded modules) |
| --trace-sync-io | As soon as the asynchronous I/O is detected at the beginning of the event loop, the stack trace will be printed |
| --zero-fill-buffers | Zero-fill Buffer and SlowBuffer |
| --preserve-symlinks | When parsing and caching modules, instruct the module loader to save symbolic links |
| --track-heap-objects | Track the allocation of heap objects for heap snapshot |
| --prof-process | Using the V8 option --prof to generate the Profilling Report |
| --v8-options | Show the V8 command line options |
| --tls-cipher-list=list | Specifies the list of alternative default TLS encryption devices |
| --enable-fips | Turn on FIPS-compliant crypto at startup |
| --force-fips | Enforce FIPS-compliant at startup |
| --openssl-config=file | Load the OpenSSL configuration file at startup |
| --icu-data-dir=file | Specify the ICU data loading path |
Environment Variable
| Environment variable | Introduction |
|---|---|
NODE_DEBUG=module[,…] | Specifies the list of core modules to print debug information |
NODE_PATH=path[:…] | Specifies the list of search directory module path prefix |
NODE_DISABLE_COLORS=1 | Close the color display for REPL |
NODE_ICU_DATA=file | ICU (Intl, object) data path |
NODE_REPL_HISTORY=file | Path of persistent storage REPL history file |
NODE_TTY_UNSAFE_ASYNC=1 | When set to 1, The stdio operation will proceed synchronously (such as console.log becomes synchronous) |
NODE_EXTRA_CA_CERTS=file | Specify an extra certificate path for CA (such as VeriSign) |
Load
Load is an important concept to measure the running state of the server. Through the load situation, we can know whether the server is idle, good, busy or about crash.
Typically, the load we want to look at is the CPU load, for more information you can read this blog: .
To get the current system load, you can use uptime, top command in terminal or os.loadavg() in Node.js:
load average: 0.09, 0.05, 0.01复制代码
Here are the average load on the system of the last 1 minutes, 5 minutes, 15 minutes. When one of the CPU core work in full load, the value of load is 1, so the value represent how many CPU cores are in full load.
In Node.js, the CPU load of a single process can be viewed using the module.
In addition to the CPU load, the server (prefer maintain) needs to know about the network load, disk load, and so on.
CheckList
A police officer sees a drunken man intently searching the ground near a lamppost and asks him the goal of his quest. The inebriate replies that he is looking for his car keys, and the officer helps for a few minutes without success then he asks whether the man is certain that he dropped the keys near the lamppost.
“No,” is the reply, “I lost the keys somewhere across the street.” “Why look here?” asks the surprised and irritated officer. “The light is much better here,” the intoxicated man responds with aplomb.
When it comes to checking server status, many server-side friends only know how to use the top command. In fact, the situation is the same as the jokes above, because top is the brightest street lamp for them.
For server-side programmers, the full server-side checklist is the described in the second chapter of .
The USE Method provides a strategy for performing a complete check of system health, identifying common bottlenecks and errors. For each system resource, metrics for utilization, saturation and errors are identified and checked. Any issues discovered are then investigated using further strategies.
This is an example USE-based metric list for Linux operating systems (eg, Ubuntu, CentOS, Fedora). This is primarily intended for system administrators of the physical systems, who are using command line tools. Some of these metrics can be found in remote monitoring tools.
Physical Resources
| component | type | metric |
|---|---|---|
| CPU | utilization | system-wide: vmstat 1, "us" + "sy" + "st"; sar -u, sum fields except "%idle" and "%iowait"; dstat -c, sum fields except "idl" and "wai"; per-cpu: mpstat -P ALL 1, sum fields except "%idle" and "%iowait"; sar -P ALL, same as mpstat; per-process: top, "%CPU"; htop, "CPU%"; ps -o pcpu; pidstat 1, "%CPU"; per-kernel-thread: top/htop ("K" to toggle), where VIRT == 0 (heuristic). [1] |
| CPU | saturation | system-wide: vmstat 1, "r" > CPU count [2]; sar -q, "runq-sz" > CPU count; dstat -p, "run" > CPU count; per-process: /proc/PID/schedstat 2nd field (sched_info.run_delay); perf sched latency (shows "Average" and "Maximum" delay per-schedule); dynamic tracing, eg, SystemTap schedtimes.stp "queued(us)" [3] |
| CPU | errors | perf (LPE) if processor specific error events (CPC) are available; eg, AMD64's "04Ah Single-bit ECC Errors Recorded by Scrubber" [4] |
| Memory capacity | utilization | system-wide: free -m, "Mem:" (main memory), "Swap:" (virtual memory); vmstat 1, "free" (main memory), "swap" (virtual memory); sar -r, "%memused"; dstat -m, "free"; slabtop -s c for kmem slab usage; per-process: top/htop, "RES" (resident main memory), "VIRT" (virtual memory), "Mem" for system-wide summary |
| Memory capacity | saturation | system-wide: vmstat 1, "si"/"so" (swapping); sar -B, "pgscank" + "pgscand" (scanning); sar -W; per-process: 10th field (min_flt) from /proc/PID/stat for minor-fault rate, or dynamic tracing [5]; OOM killer: dmesg | grep killed |
| Memory capacity | errors | dmesg for physical failures; dynamic tracing, eg, SystemTap uprobes for failed malloc()s |
| Network Interfaces | utilization | sar -n DEV 1, "rxKB/s"/max "txKB/s"/max; ip -s link, RX/TX tput / max bandwidth; /proc/net/dev, "bytes" RX/TX tput/max; nicstat "%Util" [6] |
| Network Interfaces | saturation | ifconfig, "overruns", "dropped"; netstat -s, "segments retransmited"; sar -n EDEV, *drop and *fifo metrics; /proc/net/dev, RX/TX "drop"; nicstat "Sat" [6]; dynamic tracing for other TCP/IP stack queueing [7] |
| Network Interfaces | errors | ifconfig, "errors", "dropped"; netstat -i, "RX-ERR"/"TX-ERR"; ip -s link, "errors"; sar -n EDEV, "rxerr/s" "txerr/s"; /proc/net/dev, "errs", "drop"; extra counters may be under /sys/class/net/...; dynamic tracing of driver function returns 76] |
| Storage device I/O | utilization | system-wide: iostat -xz 1, "%util"; sar -d, "%util"; per-process: iotop; pidstat -d; /proc/PID/sched "se.statistics.iowait_sum" |
| Storage device I/O | saturation | iostat -xnz 1, "avgqu-sz" > 1, or high "await"; sar -d same; LPE block probes for queue length/latency; dynamic/static tracing of I/O subsystem (incl. LPE block probes) |
| Storage device I/O | errors | /sys/devices/.../ioerr_cnt; smartctl; dynamic/static tracing of I/O subsystem response codes [8] |
| Storage capacity | utilization | swap: swapon -s; free; /proc/meminfo "SwapFree"/"SwapTotal"; file systems: "df -h" |
| Storage capacity | saturation | not sure this one makes sense - once it's full, ENOSPC |
| Storage capacity | errors | strace for ENOSPC; dynamic tracing for ENOSPC; /var/log/messages errs, depending on FS |
| Storage controller | utilization | iostat -xz 1, sum devices and compare to known IOPS/tput limits per-card |
| Storage controller | saturation | see storage device saturation, ... |
| Storage controller | errors | see storage device errors, ... |
| Network controller | utilization | infer from ip -s link (or /proc/net/dev) and known controller max tput for its interfaces |
| Network controller | saturation | see network interface saturation, ... |
| Network controller | errors | see network interface errors, ... |
| CPU interconnect | utilization | LPE (CPC) for CPU interconnect ports, tput / max |
| CPU interconnect | saturation | LPE (CPC) for stall cycles |
| CPU interconnect | errors | LPE (CPC) for whatever is available |
| Memory interconnect | utilization | LPE (CPC) for memory busses, tput / max; or CPI greater than, say, 5; CPC may also have local vs remote counters |
| Memory interconnect | saturation | LPE (CPC) for stall cycles |
| Memory interconnect | errors | LPE (CPC) for whatever is available |
| I/O interconnect | utilization | LPE (CPC) for tput / max if available; inference via known tput from iostat/ip/... |
| I/O interconnect | saturation | LPE (CPC) for stall cycles |
| I/O interconnect | errors | LPE (CPC) for whatever is available |
Software Resources
| component | type | metric |
|---|---|---|
| Kernel mutex | utilization | With CONFIG_LOCK_STATS=y, /proc/lock_stat "holdtime-totat" / "acquisitions" (also see "holdtime-min", "holdtime-max") [8]; dynamic tracing of lock functions or instructions (maybe) |
| Kernel mutex | saturation | With CONFIG_LOCK_STATS=y, /proc/lock_stat "waittime-total" / "contentions" (also see "waittime-min", "waittime-max"); dynamic tracing of lock functions or instructions (maybe); spinning shows up with profiling (perf record -a -g -F 997 ..., oprofile, dynamic tracing) |
| Kernel mutex | errors | dynamic tracing (eg, recusive mutex enter); other errors can cause kernel lockup/panic, debug with kdump/crash |
| User mutex | utilization | valgrind --tool=drd --exclusive-threshold=... (held time); dynamic tracing of lock to unlock function time |
| User mutex | saturation | valgrind --tool=drd to infer contention from held time; dynamic tracing of synchronization functions for wait time; profiling (oprofile, PEL, ...) user stacks for spins |
| User mutex | errors | valgrind --tool=drd various errors; dynamic tracing of pthread_mutex_lock() for EAGAIN, EINVAL, EPERM, EDEADLK, ENOMEM, EOWNERDEAD, ... |
| Task capacity | utilization | top/htop, "Tasks" (current); sysctl kernel.threads-max, /proc/sys/kernel/threads-max (max) |
| Task capacity | saturation | threads blocking on memory allocation; at this point the page scanner should be running (sar -B "pgscan*"), else examine using dynamic tracing |
| Task capacity | errors | "can't fork()" errors; user-level threads: pthread_create() failures with EAGAIN, EINVAL, ...; kernel: dynamic tracing of kernel_thread() ENOMEM |
| File descriptors | utilization | system-wide: sar -v, "file-nr" vs /proc/sys/fs/file-max; dstat --fs, "files"; or just /proc/sys/fs/file-nr; per-process: ls /proc/PID/fd | wc -l vs ulimit -n |
| File descriptors | saturation | does this make sense? I don't think there is any queueing or blocking, other than on memory allocation. |
| File descriptors | errors | strace errno == EMFILE on syscalls returning fds (eg, open(), accept(), ...). |
ulimit
ulimit is used to manage user access to system resources.
-a All current limits are reported-c The maximum size of core files created, take block as a unit-d The maximum size of a process's data segment, take KB as a unit-fThe maximum size of files written by the shell and its children, take block as a unit-H Set a hard limit to the resource, that is the limits set by the administrator-m The maximum resident set size, take KB as a unit-n The maximum number of open file descriptors at the same time-p The pipe size in 512-byte blocks, take 512-byte as a unit-s The maximum stack size, take KB as a unit-S Set flexible limits for resources-t The maximum amount of cpu time, in seconds-u The maximum number of processes available to a single user-v The maximum amount of virtual memory available to the shell, take KB as a unit复制代码
For example:
$ ulimit -acore file size (blocks, -c) 0data seg size (kbytes, -d) unlimitedscheduling priority (-e) 0file size (blocks, -f) unlimitedpending signals (-i) 127988max locked memory (kbytes, -l) 64max memory size (kbytes, -m) unlimitedopen files (-n) 655360pipe size (512 bytes, -p) 8POSIX message queues (bytes, -q) 819200real-time priority (-r) 0stack size (kbytes, -s) 8192cpu time (seconds, -t) unlimitedmax user processes (-u) 4096virtual memory (kbytes, -v) unlimitedfile locks (-x) unlimited复制代码
Note, open socket and other resources are also kinds of file descriptor, if ulimit -n is too small, in addition to the file can not open, it may not establish socket links.
如有任何知识产权、版权问题或理论错误，还请指正。
https://juejin.im/post/5a32164ef265da43062ac8c9 转载请注明原作者及以上信息。