mirror of
https://github.com/LCTT/TranslateProject.git
synced 2024-12-26 21:30:55 +08:00
submit tech/20180510 Get more done at the Linux command line with GNU Parallel.md
This commit is contained in:
songshunqiang
parent
20d665d839
commit
fb48ad7ecd
@ -1,262 +0,0 @@
|
||||
pinewall translating
|
||||
|
||||
Get more done at the Linux command line with GNU Parallel
|
||||
======
|
||||
|
||||
|
||||
|
||||
Do you ever get the funny feeling that your computer isn't quite as fast as it should be? I used to feel that way, and then I found GNU Parallel.
|
||||
|
||||
GNU Parallel is a shell utility for executing jobs in parallel. It can parse multiple inputs, thereby running your script or command against sets of data at the same time. You can use all your CPU at last!
|
||||
|
||||
If you've ever used `xargs`, you already know how to use Parallel. If you don't, then this article teaches you, along with many other use cases.
|
||||
|
||||
### Installing GNU Parallel
|
||||
|
||||
GNU Parallel may not come pre-installed on your Linux or BSD computer. Install it from your repository or ports collection. For example, on Fedora:
|
||||
```
|
||||
$ sudo dnf install parallel
|
||||
|
||||
```
|
||||
|
||||
Or on NetBSD:
|
||||
```
|
||||
# pkg_add parallel
|
||||
|
||||
```
|
||||
|
||||
If all else fails, refer to the [project homepage][1].
|
||||
|
||||
### From serial to parallel
|
||||
|
||||
As its name suggests, Parallel's strength is that it runs jobs in parallel rather than, as many of us still do, sequentially.
|
||||
|
||||
When you run one command against many objects, you're inherently creating a queue. Some number of objects can be processed by the command, and all the other objects just stand around and wait their turn. It's inefficient. Given enough data, there's always going to be a queue, but instead of having just one queue, why not have lots of small queues?
|
||||
|
||||
Imagine you have a folder full of images you want to convert from JPEG to PNG. There are many ways to do this. There's the manual way of opening each image in GIMP and exporting it to the new format. That's usually the worst possible way. It's not only time-intensive, it's labor-intensive.
|
||||
|
||||
A pretty neat variation on this theme is the shell-based solution:
|
||||
```
|
||||
$ convert 001.jpeg 001.png
|
||||
|
||||
$ convert 002.jpeg 002.png
|
||||
|
||||
$ convert 003.jpeg 003.png
|
||||
|
||||
... and so on ...
|
||||
|
||||
```
|
||||
|
||||
It's a great trick when you first learn it, and at first it's a vast improvement. No need for a GUI and constant clicking. But it's still labor-intensive.
|
||||
|
||||
Better still:
|
||||
```
|
||||
$ for i in *jpeg; do convert $i $i.png ; done
|
||||
|
||||
```
|
||||
|
||||
This, at least, sets the job(s) in motion and frees you up to do more productive things. The problem is, it's still a serial process. One image gets converted, and then the next one in the queue steps up for conversion, and so on until the queue has been emptied.
|
||||
|
||||
With Parallel:
|
||||
```
|
||||
$ find . -name "*jpeg" | parallel -I% --max-args 1 convert % %.png
|
||||
|
||||
```
|
||||
|
||||
This is a combination of two commands: the `find` command, which gathers the objects you want to operate on, and the `parallel` command, which sorts through the objects and makes sure everything gets processed as required.
|
||||
|
||||
* `find . -name "*jpeg"` finds all files in the current directory that end in `jpeg`.
|
||||
* `parallel` invokes GNU Parallel.
|
||||
* `-I%` creates a placeholder, called `%`, to stand in for whatever `find` hands over to Parallel. You use this because otherwise you'd have to manually write a new command for each result of `find`, and that's exactly what you're trying to avoid.
|
||||
* `--max-args 1` limits the rate at which Parallel requests a new object from the queue. Since the command Parallel is running requires only one file, you limit the rate to 1. Were you doing a more complex command that required two files (such as `cat 001.txt 002.txt > new.txt`), you would limit the rate to 2.
|
||||
* `convert % %.png` is the command you want to run in Parallel.
|
||||
|
||||
|
||||
|
||||
The result of this command is that `find` gathers all relevant files and hands them over to `parallel`, which launches a job and immediately requests the next in line. Parallel continues to do this for as long as it is safe to launch new jobs without crippling your computer. As old jobs are completed, it replaces them with new ones, until all the data provided to it has been processed. What took 10 minutes before might take only 5 or 3 with Parallel.
|
||||
|
||||
### Multiple inputs
|
||||
|
||||
The `find` command is an excellent gateway to Parallel as long as you're familiar with `find` and `xargs` (collectively called GNU Find Utilities, or `findutils`). It provides a flexible interface that many Linux users are already comfortable with and is pretty easy to learn if you're a newcomer.
|
||||
|
||||
The `find` command is fairly straightforward: you provide `find` with a path to a directory you want to search and some portion of the file name you want to search for. Use wildcard characters to cast your net wider; in this example, the asterisk indicates anything, so `find` locates all files that end with the string `searchterm`:
|
||||
```
|
||||
$ find /path/to/directory -name "*searchterm"
|
||||
|
||||
```
|
||||
|
||||
By default, `find` returns the results of its search one item at a time, with one item per line:
|
||||
```
|
||||
$ find ~/graphics -name "*jpg"
|
||||
|
||||
/home/seth/graphics/001.jpg
|
||||
|
||||
/home/seth/graphics/cat.jpg
|
||||
|
||||
/home/seth/graphics/penguin.jpg
|
||||
|
||||
/home/seth/graphics/IMG_0135.jpg
|
||||
|
||||
```
|
||||
|
||||
When you pipe the results of `find` to `parallel`, each item on each line is treated as one argument to the command that `parallel` is arbitrating. If, on the other hand, you need to process more than one argument in one command, you can split up the way the data in the queue is handed over to `parallel`.
|
||||
|
||||
Here's a simple, unrealistic example, which I'll later turn into something more useful. You can follow along with this example, as long as you have GNU Parallel installed.
|
||||
|
||||
Assume you have four files. List them, one per line, to see exactly what you have:
|
||||
```
|
||||
$ echo ada > ada ; echo lovelace > lovelace
|
||||
|
||||
$ echo richard > richard ; echo stallman > stallman
|
||||
|
||||
$ ls -1
|
||||
|
||||
ada
|
||||
|
||||
lovelace
|
||||
|
||||
richard
|
||||
|
||||
stallman
|
||||
|
||||
```
|
||||
|
||||
You want to combine two files into a third that contains the contents of both files. This requires that Parallel has access to two files, so the `-I%` variable won't work in this case.
|
||||
|
||||
Parallel's default behavior is basically invisible:
|
||||
```
|
||||
$ ls -1 | parallel echo
|
||||
|
||||
ada
|
||||
|
||||
lovelace
|
||||
|
||||
richard
|
||||
|
||||
stallman
|
||||
|
||||
```
|
||||
|
||||
Now tell Parallel you want to get two objects per job:
|
||||
```
|
||||
$ ls -1 | parallel --max-args=2 echo
|
||||
|
||||
ada lovelace
|
||||
|
||||
richard stallman
|
||||
|
||||
```
|
||||
|
||||
Now the lines have been combined. Specifically, two results from `ls -1` are passed to Parallel all at once. That's the right number of arguments for this task, but they're effectively one argument right now: "ada lovelace" and "richard stallman." What you actually want is two distinct arguments per job.
|
||||
|
||||
Luckily, that technicality is parsed by Parallel itself. If you set `--max-args` to `2`, you get two variables, `{1}` and `{2}`, representing the first and second parts of the argument:
|
||||
```
|
||||
$ ls -1 | parallel --max-args=2 cat {1} {2} ">" {1}_{2}.person
|
||||
|
||||
```
|
||||
|
||||
In this command, the variable `{1}` is ada or richard (depending on which job you look at) and `{2}` is either `lovelace` or `stallman`. The contents of the files are redirected with a redirect symbol in quotes (the quotes grab the redirect symbol from Bash so Parallel can use it) and placed into new files called `ada_lovelace.person` and `richard_stallman.person`.
|
||||
```
|
||||
$ ls -1
|
||||
|
||||
ada
|
||||
|
||||
ada_lovelace.person
|
||||
|
||||
lovelace
|
||||
|
||||
richard
|
||||
|
||||
richard_stallman.person
|
||||
|
||||
stallman
|
||||
|
||||
|
||||
|
||||
$ cat ada_*person
|
||||
|
||||
ada lovelace
|
||||
|
||||
$ cat ri*person
|
||||
|
||||
richard stallman
|
||||
|
||||
```
|
||||
|
||||
If you spend all day parsing log files that are hundreds of megabytes in size, you might see how parallelized text parsing could be useful to you; otherwise, this is mostly a demonstrative exercise.
|
||||
|
||||
However, this kind of processing is invaluable for more than just text parsing. Here's a real-life example from the film world. Consider a directory of video files and audio files that need to be joined together.
|
||||
```
|
||||
$ ls -1
|
||||
|
||||
12_LS_establishing-manor.avi
|
||||
|
||||
12_wildsound.flac
|
||||
|
||||
14_butler-dialogue-mixed.flac
|
||||
|
||||
14_MS_butler.avi
|
||||
|
||||
...and so on...
|
||||
|
||||
```
|
||||
|
||||
Using the same principles, a simple command can be created so that the files are combined in parallel:
|
||||
```
|
||||
$ ls -1 | parallel --max-args=2 ffmpeg -i {1} -i {2} -vcodec copy -acodec copy {1}.mkv
|
||||
|
||||
```
|
||||
|
||||
### Brute. Force.
|
||||
|
||||
All this fancy input and output parsing isn't to everyone's taste. If you prefer a more direct approach, you can throw commands at Parallel and walk away.
|
||||
|
||||
First, create a text file with one command on each line:
|
||||
```
|
||||
$ cat jobs2run
|
||||
|
||||
bzip2 oldstuff.tar
|
||||
|
||||
oggenc music.flac
|
||||
|
||||
opusenc ambiance.wav
|
||||
|
||||
convert bigfile.tiff small.jpeg
|
||||
|
||||
ffmepg -i foo.avi -v:b 12000k foo.mp4
|
||||
|
||||
xsltproc --output build/tmp.fo style/dm.xsl src/tmp.xml
|
||||
|
||||
bzip2 archive.tar
|
||||
|
||||
```
|
||||
|
||||
Then hand the file over to Parallel:
|
||||
```
|
||||
$ parallel --jobs 6 < jobs2run
|
||||
|
||||
```
|
||||
|
||||
And now all jobs in your file are run in Parallel. If more jobs exist than jobs allowed, a queue is formed and maintained by Parallel until all jobs have run.
|
||||
|
||||
### Much, much more
|
||||
|
||||
GNU Parallel is a powerful and flexible tool, with far more use cases than can fit into this article. Its man page provides examples of really cool things you can do with it, from remote execution over SSH to incorporating Bash functions into your Parallel commands. There's even an extensive demonstration series on [YouTube][2], so you can learn from the GNU Parallel team directly. The GNU Parallel lead maintainer has also just released the command's official guide, available from [Lulu.com][3].
|
||||
|
||||
GNU Parallel has the power to change the way you compute, and if doesn't do that, it will at the very least change the time your computer spends computing. Try it today!
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://opensource.com/article/18/5/gnu-parallel
|
||||
|
||||
作者：[Seth Kenlon][a]
|
||||
选题：[lujun9972](https://github.com/lujun9972)
|
||||
译者：[译者ID](https://github.com/译者ID)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
[a]: https://opensource.com/users/seth
|
||||
[1]:https://www.gnu.org/software/parallel
|
||||
[2]:https://www.youtube.com/watch?v=OpaiGYxkSuQ&list=PL284C9FF2488BC6D1
|
||||
[3]:http://www.lulu.com/shop/ole-tange/gnu-parallel-2018/paperback/product-23558902.html
|
||||
@ -0,0 +1,216 @@
|
||||
使用 GNU Parallel 提高 Linux 命令行执行效率
|
||||
======
|
||||
|
||||
|
||||
|
||||
你是否有过这种感觉，你的主机运行速度没有预期的那么快？我也曾经有过这种感觉，直到我发现了 GNU Parallel。
|
||||
|
||||
GNU Parallel 是一个 shell 工具，可以并行执行任务。它可以解析多种输入，让你可以同时在多份数据上运行脚本或命令。你终于可以使用全部的 CPU 了！
|
||||
|
||||
如果你用过 `xargs`，上手 Parallel 几乎没有难度。如果没有用过，这篇教程会告诉你如何使用，同时给出一些其它的用例。
|
||||
|
||||
### 安装 GNU Parallel
|
||||
|
||||
GNU Parallel 很可能没有预装在你的 Linux 或 BSD 主机上，你可以从软件源（Linux 对应 repository，BSD 对应 ports collection）中安装。以 Fedora 为例：
|
||||
```
|
||||
$ sudo dnf install parallel
|
||||
|
||||
```
|
||||
|
||||
对于 NetBSD：
|
||||
```
|
||||
# pkg_add parallel
|
||||
|
||||
```
|
||||
|
||||
如果各种方式都不成功，请参考[项目主页][1]。
|
||||
|
||||
### 从串行到并行
|
||||
|
||||
正如其名称所示，Parallel 的强大之处是以并行方式执行任务；而我们中不少人平时仍然以串行方式运行任务。
|
||||
|
||||
当你对多个对象执行某个命令时，你实际上创建了一个任务队列。一部分对象可以被命令处理，剩余的对象需要等待，直到命令处理它们。这种方式是低效的。只要数据够多，总会形成任务队列；但与其只使用一个任务队列，为何不使用多个更小规模的任务队列呢？
|
||||
|
||||
假设你有一个图片目录，你希望将目录中的图片从 JEEG 格式转换为 PNG 格式。有多种方法可以完成这个任务。可以手动用 GIMP 打开每个图片，输出成新格式，但这基本是最差的选择，费时费力。
|
||||
|
||||
上述方法有一个漂亮且简洁的变种，即基于 shell 的方案：
|
||||
```
|
||||
$ convert 001.jpeg 001.png
|
||||
$ convert 002.jpeg 002.png
|
||||
$ convert 003.jpeg 003.png
|
||||
... 略 ...
|
||||
|
||||
```
|
||||
|
||||
对于初学者而言，这是一个不小的转变，而且看起来是个不小的改进。不再需要图像界面和不断的鼠标点击，但仍然是费力的。
|
||||
|
||||
进一步改进：
|
||||
```
|
||||
$ for i in *jpeg; do convert $i $i.png ; done
|
||||
|
||||
```
|
||||
|
||||
至少，这一步设置好任务执行，让你节省时间去做更有价值的事情。但问题来了，这仍然是串行操作；一张图片转换完成后，队列中的下一张进行转换，依此类推直到全部完成。
|
||||
|
||||
使用 Parallel:
|
||||
```
|
||||
$ find . -name "*jpeg" | parallel -I% --max-args 1 convert % %.png
|
||||
|
||||
```
|
||||
|
||||
这是两条命令的组合：`find` 命令，用于收集需要操作的对象；`parallel` 命令，用于对象排序并确保每个对象按需处理。
|
||||
|
||||
* `find . -name "*jpeg"` 查找当前目录下以 `jpeg` 结尾的所有文件。
|
||||
* `parallel` 调用 GNU Parallel。
|
||||
* `-I%` 创建了一个占位符 `%`，代表 `find` 传递给 Parallel 的内容。如果不使用占位符，你需要对 `find` 命令的每一个结果手动编写一个命令，而这恰恰是你想要避免的。
|
||||
* `--max-args 1` 给出 Parallel 从队列获取新对象的速率限制。考虑到 Parallel 运行的命令只需要一个文件输入，这里将速率限制设置为 1。假如你需要执行更复杂的命令，需要两个文件输入（例如 `cat 001.txt 002.txt > new.txt`），你需要将速率限制设置为 2。
|
||||
* `convert % %.png` 是你希望 Parallel 执行的命令。
|
||||
|
||||
|
||||
组合命令的执行效果如下：`find` 命令收集所有相关的文件信息并传递给 `parallel`，后者（使用当前参数）启动一个任务，（无需等待任务完成）立即获取参数行中的下一个参数（注：管道输出的每一行对应 `parallel` 的一个参数，所有参数构成参数行）；只要你的主机没有瘫痪，Parallel 会不断做这样的操作。旧任务完成后，Parallel 会为分配新任务，直到所有数据都处理完成。不使用 Parallel 完成任务大约需要 10 分钟，使用后仅需 3 至 5 分钟。
|
||||
|
||||
### 多个输入
|
||||
|
||||
只要你熟悉 `find` 和 `xargs` （整体被称为 GNU 查找工具，或 `findutils`）,`find` 命令是一个完美的 Parallel 数据提供者。它提供了灵活的接口，大多数 Linux 用户已经很习惯使用，即使对于初学者也很容易学习。
|
||||
|
||||
`find` 命令十分直截了当：你向 `find` 提供搜索路径和待查找文件的一部分信息。可以使用通配符完成模糊搜索；在下面的例子中，星号匹配任何字符，故 `find` 定位（文件名）以字符 `searchterm` 结尾的全部文件：
|
||||
```
|
||||
$ find /path/to/directory -name "*searchterm"
|
||||
|
||||
```
|
||||
|
||||
默认情况下，`find` 逐行返回搜索结果，每个结果对应 1 行：
|
||||
```
|
||||
$ find ~/graphics -name "*jpg"
|
||||
/home/seth/graphics/001.jpg
|
||||
/home/seth/graphics/cat.jpg
|
||||
/home/seth/graphics/penguin.jpg
|
||||
/home/seth/graphics/IMG_0135.jpg
|
||||
|
||||
```
|
||||
|
||||
当使用管道将 `find` 的结果传递给 `parallel` 时，每一行中的文件路径被视为 `parallel` 命令的一个参数。另一方面，如果你需要使用命令处理多个参数，你可以改变队列数据传递给 `parallel` 的方式。
|
||||
|
||||
下面先给出一个不那么真实的例子，后续会做一些修改使其更加有意义。如果你安装了 GNU Parallel，你可以跟着这个例子操作。
|
||||
|
||||
假设你有 4 个文件，按照每行一个文件的方式列出，具体如下：
|
||||
```
|
||||
$ echo ada > ada ; echo lovelace > lovelace
|
||||
$ echo richard > richard ; echo stallman > stallman
|
||||
$ ls -1
|
||||
ada
|
||||
lovelace
|
||||
richard
|
||||
stallman
|
||||
|
||||
```
|
||||
|
||||
你需要将两个文件合并成第三个文件，后者同时包含前两个文件的内容。这种情况下，Parallel 需要访问两个文件，使用 `-I%` 变量的方式不符合本例的预期。
|
||||
|
||||
Parallel 默认情况下读取 1 个队列对象：
|
||||
```
|
||||
$ ls -1 | parallel echo
|
||||
ada
|
||||
lovelace
|
||||
richard
|
||||
stallman
|
||||
|
||||
```
|
||||
现在让 Parallel 每个任务使用 2 个队列对象：
|
||||
```
|
||||
$ ls -1 | parallel --max-args=2 echo
|
||||
ada lovelace
|
||||
richard stallman
|
||||
|
||||
```
|
||||
|
||||
现在，我们看到行已经并合并；具体而言，`ls -1` 的两个查询结果会被同时传送给 Parallel。传送给 Parallel 的参数涉及了任务所需的 2 个文件，但目前还只是 1 个有效参数：（对于两个任务分别为）"ada lovelace" 和 "richard stallman"。你真正需要的是每个任务对应 2 个独立的参数。
|
||||
|
||||
值得庆幸的是，Parallel 本身提供了上述所需的解析功能。如果你将 `--max-args` 设置为 `2`，那么 `{1}` 和 `{2}` 这两个变量分别代表传入参数的第一和第二部分：
|
||||
```
|
||||
$ ls -1 | parallel --max-args=2 cat {1} {2} ">" {1}_{2}.person
|
||||
|
||||
```
|
||||
|
||||
在上面的命令中，变量 `{1}` 值为 `ada` 或 `richard` （取决于你选取的任务），变量 `{2}` 值为 `lovelace` 或 `stallman`。通过使用重定向符号（放到引号中，防止被 Bash 识别，以便 Parallel 使用），（两个）文件的内容被分别重定向至新文件 `ada_lovelace.person` 和 `richard_stallman.person`。
|
||||
```
|
||||
$ ls -1
|
||||
ada
|
||||
ada_lovelace.person
|
||||
lovelace
|
||||
richard
|
||||
richard_stallman.person
|
||||
stallman
|
||||
|
||||
$ cat ada_*person
|
||||
ada lovelace
|
||||
$ cat ri*person
|
||||
richard stallman
|
||||
|
||||
```
|
||||
|
||||
如果你整天处理大量几百 MB 大小的日志文件，那么（上述）并行处理文本的方法对你帮忙很大；否则，上述例子只是个用于上手的示例。
|
||||
|
||||
然而，这种处理方法对于很多文本处理之外的操作也有很大帮助。下面是来自电影产业的真实案例，其中需要将一个目录中的视频文件和（对应的）音频文件进行合并。
|
||||
```
|
||||
$ ls -1
|
||||
12_LS_establishing-manor.avi
|
||||
12_wildsound.flac
|
||||
14_butler-dialogue-mixed.flac
|
||||
14_MS_butler.avi
|
||||
...略...
|
||||
|
||||
```
|
||||
|
||||
使用同样的方法，使用下面这个简单命令即可并行地合并文件：
|
||||
```
|
||||
$ ls -1 | parallel --max-args=2 ffmpeg -i {1} -i {2} -vcodec copy -acodec copy {1}.mkv
|
||||
|
||||
```
|
||||
|
||||
### 简单粗暴的方式
|
||||
|
||||
上述花哨的输入输出处理不一定对所有人的口味。如果你希望更直接一些，可以将一堆命令甩给 Parallel，然后去干些其它事情。
|
||||
|
||||
首先，需要创建一个文本文件，每行包含一个命令：
|
||||
```
|
||||
$ cat jobs2run
|
||||
bzip2 oldstuff.tar
|
||||
oggenc music.flac
|
||||
opusenc ambiance.wav
|
||||
convert bigfile.tiff small.jpeg
|
||||
ffmepg -i foo.avi -v:b 12000k foo.mp4
|
||||
xsltproc --output build/tmp.fo style/dm.xsl src/tmp.xml
|
||||
bzip2 archive.tar
|
||||
|
||||
```
|
||||
|
||||
接着，将文件传递给 Parallel：
|
||||
```
|
||||
$ parallel --jobs 6 < jobs2run
|
||||
|
||||
```
|
||||
|
||||
现在文件中对应的全部任务都在被 Parallel 执行。如果任务数量超过允许的数目（译者注：应该是 --jobs 指定的数目或默认值），Parallel 会创建并维护一个队列，直到任务全部完成。
|
||||
|
||||
### 更多内容
|
||||
|
||||
GNU Parallel 是个强大而灵活的工具，还有很多很多用例无法在本文中讲述。工具的 man 页面提供很多非常酷的例子可供你参考，包括通过 SSH 远程执行和在 Parallel 命令中使用 Bash 函数等。[YouTube][2] 上甚至有一个系列，包含大量操作演示，让你可以直接从 GNU Parallel 团队学习。GNU Paralle 的主要维护者还发布了官方使用指导手册，可以从 [Lulu.com][3] 获取。
|
||||
|
||||
GNU Parallel 有可能改变你完成计算的方式；即使没有，也会至少改变你主机花在计算上的时间。马上上手试试吧！
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://opensource.com/article/18/5/gnu-parallel
|
||||
|
||||
作者：[Seth Kenlon][a]
|
||||
选题：[lujun9972](https://github.com/lujun9972)
|
||||
译者：[pinewall](https://github.com/pinewall)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
[a]: https://opensource.com/users/seth
|
||||
[1]:https://www.gnu.org/software/parallel
|
||||
[2]:https://www.youtube.com/watch?v=OpaiGYxkSuQ&list=PL284C9FF2488BC6D1
|
||||
[3]:http://www.lulu.com/shop/ole-tange/gnu-parallel-2018/paperback/product-23558902.html
|
||||
Loading…
Reference in New Issue
Block a user