我正在托管一些简单的docker容器。我想知道容器的大小会随着时间的推移而迅速增加，我不知道如何找出问题。
Docker报告的大小：
me@somewhere:~$ sudo docker ps -s
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES SIZE
02b30add1cb3 my-service "npm start" 23 hours ago Up 23 hours 3001/tcp, 0.0.0.0:9017->9017/tcp my-service-frontend 0 B (virtual 776.4 MB)
20a2be4931e7 my-service "phantomjs src/sites/" 23 hours ago Up 23 hours 0.0.0.0:3007->3001/tcp my-service-5 6.144 kB (virtual 776.4 MB)
ba340ba08941 my-service "phantomjs src/sites/" 23 hours ago Up 23 hours 0.0.0.0:3006->3001/tcp my-service-4 6.144 kB (virtual 776.4 MB)
7b5411d8a171 my-service "phantomjs src/sites/" 23 hours ago Up 23 hours 0.0.0.0:3003->3001/tcp my-service-1 6.144 kB (virtual 776.4 MB)
b583a544b37d my-service "phantomjs src/sites/" 23 hours ago Up 23 hours 0.0.0.0:3001->3001/tcp my-service-0 6.144 kB (virtual 776.4 MB)
91373086e06e foo_bar "/bin/sh -c 'git pull" 47 hours ago Up 47 hours 0.0.0.0:12776->8080/tcp kickass_murdock 11.26 MB (virtual 1.081 GB)
du在主持人上报告的尺寸：
me@somewhere:~$ sudo du -h -d 1 /var/lib/docker/containers
14G /var/lib/docker/containers/20a2be4931e7a10b2e29260b541e3c4d6581462650e47d59682f84626843752b
1,6G /var/lib/docker/containers/7b5411d8a171a35a3c937d62dbdea141fc0a9f3c4de25a2da3a0b94ea71a8f3d
9,6M /var/lib/docker/containers/02b30add1cb3ba6d5be1c36b2c9dd141d8d70cb88a021d2363af5684ef3c220f
480K /var/lib/docker/containers/91373086e06ea83269465e0b026cfe7ca0158a1315b0df04da9a1d1b4ee52823
13G /var/lib/docker/containers/b583a544b37db6144f17a4819ca2f636126b11d668caab3dcdbf4c3a33dedc65
13G /var/lib/docker/containers/ba340ba08941d47af45230be328ef7289c19b6bb6a0d120cf2098cbdd9983f65
40G /var/lib/docker/containers`
du为容器报告的大小（所有其他容器的类似输出）：
me@somewhere:~$ sudo docker exec -it my-service-4 du -h -d1 -c /
58M /root
0 /dev
3.0M /etc
706M /usr
1.4M /tmp
14M /var
9.0M /bin
32M /lib
4.0K /home
8.0K /run
4.0K /mnt
4.0K /boot
0 /sys
4.0K /opt
4.0K /srv
4.0K /lib64
3.9M /sbin
du: cannot access '/proc/12642/task/12642/fd/3': No such file or directory
du: cannot access '/proc/12642/task/12642/fdinfo/3': No such file or directory
du: cannot access '/proc/12642/fd/3': No such file or directory
du: cannot access '/proc/12642/fdinfo/3': No such file or directory
0 /proc
4.0K /media
825M /
825M total
所以：容器和docker ps都报告磁盘使用率低于1G，但实际的容器文件大小超过10 GB（至少对于某些容器而言）。任何人都可以帮助我并告诉我们发生了什么吗？我想在我的容器内发生了一些麻烦，虽然我不知道现在在哪里看。谁知道我要做什么？
答案 0 :(得分：1)
只要将更改存储到图像，Docker就会在其洋葱文件系统中将一层放在另一层上。当您删除容器internal class Timer
{
private readonly TimeSpan _disposalTimeout;
private readonly System.Threading.Timer _timer;
private bool _disposeEnded;
public Timer(TimeSpan disposalTimeout)
{
_disposalTimeout = disposalTimeout;
_timer = new System.Threading.Timer(HandleTimerElapsed);
}
public event Signal Elapsed;
public void TriggerOnceIn(TimeSpan time)
{
try
{
_timer.Change(time, Timeout.InfiniteTimeSpan);
}
catch (ObjectDisposedException)
{
// race condition with Dispose can cause trigger to be called when underlying
// timer is being disposed - and a change will fail in this case.
// see
// https://msdn.microsoft.com/en-us/library/b97tkt95(v=vs.110).aspx#Anchor_2
if (_disposeEnded)
{
// we still want to throw the exception in case someone really tries
// to change the timer after disposal has finished
// of course there's a slight race condition here where we might not
// throw even though disposal is already done.
// since the offending code would most likely already be "failing"
// unreliably i personally can live with increasing the
// "unreliable failure" time-window slightly
throw;
}
}
}
private void HandleTimerElapsed(object state)
{
Elapsed.SafeInvoke();
}
public void Dispose()
{
using (var waitHandle = new ManualResetEvent(false))
{
// returns false on second dispose
if (_timer.Dispose(waitHandle))
{
if (!waitHandle.WaitOne(_disposalTimeout))
{
throw new TimeoutException(
"Timeout waiting for timer to stop. (...)");
}
_disposeEnded = true;
}
}
}
}
时，您应该看到/ var / lib / docker / containers占用的空间更少。
构建图像时会存储对图像的更改。如果您只使用图像 - 运行容器 - 更改的数据只是“保持”。因此，您应该调查容器中发生的情况。 ESP。看看你的幻像服务器在文件系统中产生了什么。 ncdu将是一个很好的工具。
从ncdu开始，然后存储在hosts文件系统的附加文件夹中生成的文件。 docker rmi -f CONTAINERID
<强>测试
运行一个简单的容器，只包含os（在我的情况下是alpine）
docker run -it -v FULLPATHATHOST:FULLPATHWITHINCONTAINER CONTAINERID在主持人转到docker run -it stk/alpine:base sh
并列出包含ncdu的目录。 （当然，你可以使用你喜欢的任何其他程序来确定指令大小）
/var/lib/docker/aufs/diff/在容器内执行类似
的操作ncdu 1.10 ~ Use the arrow keys to navigate, press ? for help
--- /var/lib/docker/aufs/diff --------------------------------------------------
/..
286,4MiB [##########] /39f3e2ea0dfe17366b8cd7b0...bf3681b99a1081e33ad62a509f28
214,3MiB [####### ] /2de39307b9361cae12f0116e...d28056e4699b21b9a4d34f374461
207,7MiB [####### ] /92ec6d044cb3e39ae0050012...78d0591675f2231daafbf0877778
154,0MiB [##### ] /9f3806e6bedc8fb01929131b...e01aa1980aadba914fdd9d2f96ae
149,5MiB [##### ] /5f0ca2331640639507d85b83...693659438367311abb0c792b8a62
136,8MiB [#### ] /902b87aaaec929e805414868...1f529ad7f37ab300d4ef9f3a0dbf
136,2MiB [#### ] /222ba86561913d299deb9e0e...6b5f5ec117b01386a4156d092687
132,6MiB [#### ] /8b3a9a9eeaf8ed59f24f21a2...dfe8d033890a2fa44b445deb2e3c
128,5MiB [#### ] /72b3edf317a8d682466c1500...a5e2cad31c8305ed42c41cd61149
117,0MiB [#### ] /818e3763e72ef82b28b0552e...b9f163dc601d266e94e46fd26bb0
57,4MiB [## ] /eeffdfafed9f60771b5bf87a...e8bbd16b572f77899c8e689d174d
56,1MiB [# ] /6976ce3ed5fab37382d90467...37578332417ffcf35a1d499eba52
51,3MiB [# ] /a5a6e0549d247f1c8b81a350...c5071f46d17afe2f8988817360b3
Total disk usage: 2,7GiB Apparent size: 2,7GiB Items: 168658
将创建一个名为tr -dc A-Za-z0-9 </dev/urandom | head -c 409600000 > a.txt && ls a.txt -all -h
的随机数据的文件我选择大小a.txt大于286,4MiB - 409600000中最大的文件夹 - 以便ncdu显示它在顶部
/var/lib/docker/aufs/diff/现在我知道这是以ncdu 1.10 ~ Use the arrow keys to navigate, press ? for help
--- /var/lib/docker/aufs/diff --------------------------------------------------
/..
390,6MiB [##########] /0720a07653a57d938c861cf3...e61c81c29f12289759f0560aa38f
286,4MiB [####### ] /39f3e2ea0dfe17366b8cd7b0...bf3681b99a1081e33ad62a509f28
214,3MiB [##### ] /2de39307b9361cae12f0116e...d28056e4699b21b9a4d34f374461
207,7MiB [##### ] /92ec6d044cb3e39ae0050012...78d0591675f2231daafbf0877778
154,0MiB [### ] /9f3806e6bedc8fb01929131b...e01aa1980aadba914fdd9d2f96ae
149,5MiB [### ] /5f0ca2331640639507d85b83...693659438367311abb0c792b8a62
136,8MiB [### ] /902b87aaaec929e805414868...1f529ad7f37ab300d4ef9f3a0dbf
136,2MiB [### ] /222ba86561913d299deb9e0e...6b5f5ec117b01386a4156d092687
132,6MiB [### ] /8b3a9a9eeaf8ed59f24f21a2...dfe8d033890a2fa44b445deb2e3c
128,5MiB [### ] /72b3edf317a8d682466c1500...a5e2cad31c8305ed42c41cd61149
117,0MiB [## ] /818e3763e72ef82b28b0552e...b9f163dc601d266e94e46fd26bb0
57,4MiB [# ] /eeffdfafed9f60771b5bf87a...e8bbd16b572f77899c8e689d174d
56,1MiB [# ] /6976ce3ed5fab37382d90467...37578332417ffcf35a1d499eba52
Total disk usage: 3,0GiB Apparent size: 3,0GiB Items: 168678
开头的目录，我必须寻找。进入它并列出内容
0720a07653a57d9...如您所见，列出了root@T520:/var/lib/docker/aufs/diff# cd 0720a07653a57d938c861cf32f4bee87fa4be61c81c29f12289759f0560aa38f
root@T520:/var/lib/docker/aufs/diff/0720a07653a57d938c861cf32f4bee87fa4be61c81c29f12289759f0560aa38f# ls -all -h
insgesamt 391M
drwxr-xr-x 5 root root 4,0K Feb 23 10:55 .
drwxr-xr-x 674 root root 80K Feb 23 10:55 ..
-rw-r--r-- 1 root root 391M Feb 23 10:57 a.txt
drwx------ 2 root root 4,0K Feb 23 10:55 root
-r--r--r-- 1 root root 0 Feb 23 10:55 .wh..wh.aufs
drwx------ 2 root root 4,0K Feb 23 10:55 .wh..wh.orph
drwx------ 2 root root 4,0K Feb 23 10:55 .wh..wh.plnk
文件。
现在重新运行程序，创建随机文件并重新执行ncdu（只需在ncdu中点击a.txt。）
r应该向您显示，ncdu应该这样做，目录大小没有改变。所以docker fs中的数据只是被覆盖了。如果你选择一个小尺寸它会变小。
那么这对你有什么帮助？ 如上所示，文件中的数据更改没有文件增长。你可以找到包含文件系统的女巫目录，并查看容器中添加/更改文件的普通文件结构。
希望这有助于查找容器中的文件。
如果您退出容器并再次使用相同的命令重新启动它，则会创建一个新实例，并创建自己的fs图层。
您可以使用
找到已停止容器的IDls要在删除之前查看找到的内容...
docker ps -a | grep Exited | grep stk/alpine:base | awk '{print $1 }'
附加docker ps -a | grep Exited | grep stk/alpine:base
7b7d3f6e857a stk/alpine:base "sh" 22 minutes ago Exited (0) 2 minutes ago gigantic_swartz
2f51ea988a28 stk/alpine:base "sh" 23 minutes ago Exited (0) 22 minutes ago cranky_euler
4bfafbb034fe stk/alpine:base "sh" 34 minutes ago Exited (0) 25 minutes ago sick_williams
80cd5687fcd7 stk/alpine:base "sh" 44 minutes ago Exited (137) 37 minutes ago determined_panini
a2179a8dd543 stk/alpine:base "sh" 58 minutes ago Exited (130) 44 minutes ago agitated_shockley
8596cd310292 stk/alpine:base "sh" 3 days ago Exited (137) 3 days ago dreamy_murdock
33db61a7830b stk/alpine:base "sh" 3 days ago Exited (0) 3 days ago desperate_hodgkin
2f96c15dc8a1 stk/alpine:base "sh" 2 weeks ago Exited (0) 2 weeks ago determined_babbage
以删除它们
一线解决方案
| xargs -r docker rm Docker将检查其他图片是否未使用这些图片，并在docker ps -a | grep Exited | grep stk/alpine:base | awk '{print $1 }' | xargs -r docker rm
执行docker rmi时无法删除。但在这种情况下，您希望删除容器而不是图像。因此，请使用rm代替rmi（我更新了答案）
享受