在Linux中刪除超大(100-200GB)文件的實(shí)現(xiàn)方式
一、IO調(diào)度策略
1.1、idle
空閑磁盤調(diào)度;
該調(diào)度策略是在當(dāng)前系統(tǒng)沒有其他進(jìn)程需要進(jìn)行磁盤IO時,才能進(jìn)行磁盤;
因此該策略對當(dāng)前系統(tǒng)的影響基本為0;
當(dāng)然,該調(diào)度策略不能帶有任何優(yōu)先級參數(shù);
1.2、Best effort
是缺省的磁盤IO調(diào)度策略;
(1)該調(diào)度策略可以指定優(yōu)先級參數(shù)(范圍是0~7,數(shù)值越小,優(yōu)先級越高);
(2)針對處于同一優(yōu)先級的程序?qū)⒉蓃ound-robin方式;
(3)對于best effort調(diào)度策略,8個優(yōu)先級等級可以說明在給定的一個調(diào)度窗口中時間片的大??;
(4)目前,普調(diào)用戶(非root用戶)是可以使用該調(diào)度策略;
(5)在內(nèi)核2.6.26之前,沒有設(shè)置IO優(yōu)先級的進(jìn)程會使用“none”作為調(diào)度策略,但是這種策略使得進(jìn)程看起來像是采用了best effort調(diào)度策略,因為其優(yōu)先級是通過關(guān)于 cpu nice有關(guān)的公式計算得到的:io_priority = (cpu_nice + 20) /5
(6)在內(nèi)核2.6.26之后,如果當(dāng)前系統(tǒng)使用的是CFQ調(diào)度器,那么如果進(jìn)程沒有設(shè)置IO優(yōu)先級級別,將采用與內(nèi)核2.6.26之前版本同樣的方式,推到出io優(yōu)先級級別;
1.3、Real time
實(shí)時調(diào)度策略,如果設(shè)置了該磁盤IO調(diào)度策略,則立即訪問磁盤,不管系統(tǒng)中其他進(jìn)程是否有IO。
因此使用實(shí)時調(diào)度策略,需要注意的是,該訪問策略可能會使得其他進(jìn)程處于等待狀態(tài);
二、ionice命令-設(shè)置排程優(yōu)先級
2.1、安裝
[root@docker][19:43:12][OK] ~ #yum provides */ionice Loaded plugins: fastestmirror Loading mirror speeds from cached hostfile * base: mirrors.aliyun.com * epel: mirror.01link.hk * extras: mirrors.aliyun.com * updates: mirrors.aliyun.com epel/x86_64/filelists_db | 12 MB 00:00:08 extras/7/x86_64/filelists_db | 276 kB 00:00:01 updates/7/x86_64/filelists_db | 10 MB 00:00:42 util-linux-2.23.2-65.el7.i686 : A collection of basic system utilities Repo : base Matched from: Filename : /usr/bin/ionice Filename : /usr/share/bash-completion/completions/ionice util-linux-2.23.2-65.el7.x86_64 : A collection of basic system utilities Repo : base Matched from: Filename : /usr/bin/ionice Filename : /usr/share/bash-completion/completions/ionice util-linux-2.23.2-65.el7_9.1.i686 : A collection of basic system utilities Repo : updates Matched from: Filename : /usr/bin/ionice Filename : /usr/share/bash-completion/completions/ionice util-linux-2.23.2-65.el7_9.1.x86_64 : A collection of basic system utilities Repo : updates Matched from: Filename : /usr/bin/ionice Filename : /usr/share/bash-completion/completions/ionice util-linux-2.23.2-65.el7.x86_64 : A collection of basic system utilities Repo : @anaconda Matched from: Filename : /usr/bin/ionice Filename : /usr/share/bash-completion/completions/ionice
安裝
yum install util-linux-2.23.2-65.el7_9.1.x86_64 -y
2.2、用法
[root@docker][19:35:01][OK] ~
#ionice --help
ionice - sets or gets process io scheduling class and priority.
Usage:
ionice [OPTION] -p PID [PID...]
ionice [OPTION] COMMAND
Options:
-c, --class <class> scheduling class name or number
0: none, 1: realtime, 2: best-effort, 3: idle
-n, --classdata <num> scheduling class data
0-7 for realtime and best-effort classes
-p, --pid=PID view or modify already running process
-t, --ignore ignore failures
-V, --version output version information and exit
-h, --help display this help and exit2.3、示例
—- Deleting Huge Files in Linux —– # ionice -c 3 rm /var/logs/syslog # ionice -c 3 rm -rf /var/log/apache
注意:要使用安全的方法刪除大檔案,我們可以使 shred、wipe 以及 secure-deletion 工具包中的不同工具,而不是 rm 命令。
三、Shred -覆蓋文件以隱藏內(nèi)容
3.1、安裝軟件包
yum install coreutils-8.22-24.el7_9.2.x86_64 -y
3.2、命令及參數(shù)選項
[root@192_168_209_128 ~]# shred --help
Usage: shred [OPTION]... FILE...
Overwrite the specified FILE(s) repeatedly, in order to make it harder
for even very expensive hardware probing to recover the data.
Mandatory arguments to long options are mandatory for short options too.
-f, --force change permissions to allow writing if necessary
-n, --iterations=N overwrite N times instead of the default (3)
--random-source=FILE get random bytes from FILE
-s, --size=N shred this many bytes (suffixes like K, M, G accepted)
-u, --remove[=HOW] truncate and remove file after overwriting; See below
-v, --verbose show progress
-x, --exact do not round file sizes up to the next full block;
this is the default for non-regular files
-z, --zero add a final overwrite with zeros to hide shredding
--help display this help and exit
--version output version information and exit選項說明:
- -z- 添加一個帶有零的最終覆蓋以隱藏切碎。
- -v– 可以顯示操作進(jìn)度。
- -u- 覆蓋后截斷和刪除文件。
- -n– 指定覆蓋文件內(nèi)容的次數(shù)(默認(rèn)為 3)。
3.3、示例
shred覆蓋文件以隱藏其內(nèi)容,也可以選擇刪除它。
[root@docker][19:51:03][OK] ~ #shred -zvu -n 5 cuttingLog_bak shred: cuttingLog_bak: pass 1/6 (random)... shred: cuttingLog_bak: pass 2/6 (ffffff)... shred: cuttingLog_bak: pass 3/6 (random)... shred: cuttingLog_bak: pass 4/6 (000000)... shred: cuttingLog_bak: pass 5/6 (random)... shred: cuttingLog_bak: pass 6/6 (000000)... shred: cuttingLog_bak: removing shred: cuttingLog_bak: renamed to 00000000000000 shred: 00000000000000: renamed to 0000000000000 shred: 0000000000000: renamed to 000000000000 shred: 000000000000: renamed to 00000000000 shred: 00000000000: renamed to 0000000000 shred: 0000000000: renamed to 000000000 shred: 000000000: renamed to 00000000 shred: 00000000: renamed to 0000000 shred: 0000000: renamed to 000000 shred: 000000: renamed to 00000 shred: 00000: renamed to 0000 shred: 0000: renamed to 000 shred: 000: renamed to 00 shred: 00: renamed to 0 shred: cuttingLog_bak: removed
四、Wipe —在 Linux 中安全擦除文件
介紹:Linux wipe命令可以安全地從磁存儲器中擦除文件,從而無法恢復(fù)。
4.1、安裝軟件包
rpm -ivh wipe-2.2.0-1.2.el7.rf.x86_64.rpm
4.2、命令及參數(shù)選項
[root@docker][19:58:11][OK] ~
#wipe -h
Wipe v2.2.0 - released January 10, 2004
by Tom Vier <nester@users.sf.net>
Usage is wipe [options] [file-list]
Options: Default: wipe -ZdntVAkO -S512 -C4096 -l1 -x8 -p1
-h -- help - display this screen
-u -- usage
-c -- show copyright and license
-w -- show warranty information
-i and -I -- enable (-i) or disable (-I) interaction - overrides force
-f -- force file wiping and override interaction
-r and -R -- recursion - traverse subdirectories
-s -- silent - disable percentage and error reporting
-v -- force verbose - always show percentage
-V -- verbose - show percentage if file is >= 25K
-d and -D -- delete (-d) or keep (-D) after wiping
-n and -N -- delete (-n) or skip (-N) special files
-k and -K -- lock (-k) or don't lock (-K) files
-z -- zero-out file - single pass of zeroes
-Z -- perform normal wipe passes
-t and -T -- enable (-t) or disable (-T) static passes
-a and -A -- write until out of space (-a) or don't (-A)
-o[size] -O -- write to stdout (-o) or use files (-O)
-B(count) -- block device sector count
-S(size) -- block device sector size - default 512 bytes
or stdout write length when used with -A
-C(size) -- chunk size - maximum file buffer size in kilobytes (2^10)
-l[0-2] -- sets wipe secure level
-x[1-32] -X -- sets number of random passes per wipe or disables
-p(1-32) -- wipe file x number of times
-b(0-255) -- overwrite file with this value byte4.3、示例
[root@docker][19:59:50][OK] ~ #wipe -fi cuttingLog_bak wipe: destroy file `cuttingLog_bak'? y
五、Linux 安全刪除工具包
Secure-delete是一個安全文件刪除工具的集合,其中包含srm (secure_deletion) 工具,用于安全刪除文件。
5.1、安裝軟件包
yum -y install srm-1.2.15-1.el7.x86_64
5.2、命令及參數(shù)選項
[root@docker][20:05:39][OK] ~ #srm --help Usage: srm [OPTION]... [FILE]... Overwrite and remove (unlink) the files. By default use the 35-pass Gutmann method to overwrite files. -d, --directory ignored (for compatability with rm(1)) -f, --force ignore nonexistant files, never prompt -i, --interactive prompt before any removal -x, --one-file-system do not cross file system boundaries -s, --simple overwrite with single pass using 0x00 (default) -P, --openbsd overwrite with three passes like OpenBSD rm -D, --dod overwrite with 7 US DoD compliant passes -E, --doe overwrite with 3 US DoE compliant passes -G, --gutmann overwrite with 35-pass Gutmann method -C, --rcmp overwrite with Royal Canadian Mounted Police passes -r, -R, --recursive remove the contents of directories -v, --verbose explain what is being done -h, --help display this help and exit -V, --version display version information and exit
5.3、示例
#srm -vf cuttingLog_bal srm: removing cuttingLog_bal
總結(jié)
以上為個人經(jīng)驗,希望能給大家一個參考,也希望大家多多支持腳本之家。
相關(guān)文章
在Linux系統(tǒng)中設(shè)置系統(tǒng)時間的操作方法
本文介紹了在CentOS系統(tǒng)中設(shè)置和同步系統(tǒng)時間的方法,首先,通過查看系統(tǒng)時間和時區(qū),發(fā)現(xiàn)與當(dāng)?shù)貢r間不一致,然后,通過設(shè)置正確的時區(qū),將系統(tǒng)時間調(diào)整為北京時間,最后,通過開啟NTP服務(wù),確保系統(tǒng)時間與網(wǎng)絡(luò)時間同步,避免誤差,需要的朋友可以參考下2026-02-02
使用 Linux seq 命令生成數(shù)字序列(推薦)
seq命令用于以指定增量從首數(shù)開始打印數(shù)字到尾數(shù),即產(chǎn)生從某個數(shù)到另外一個數(shù)之間的所有整數(shù),并且可以對整數(shù)的格式、寬度、分割符號進(jìn)行控制。這篇文章主要介紹了使用 Linux seq 命令生成數(shù)字序列,需要的朋友可以參考下2020-01-01

