Posts Tagged ‘cmd’
Saturday, April 28th, 2012
If there is necessity to look for a string in all hidden files with all sub-level subdirectories (be aware this will be time consuming and CPU stressing) use:
hipo@noah:~$ grep -rli 'PATH' .*
./.gftp/gftprc
./.gftp/cache/cache.OOqZVP
….
Sometimes its necessery to only grep for variables within the first-level directories (lets say you would like to grep a 'PATH' variable set, string within the $HOME directory, the command is:
hipo@noah:~$ grep PATH .[!.]*
.profile:PATH=/bin:/usr/bin/:${PATH}
.profile:export PATH
.profile:# set PATH so it includes user's private bin if it exists
.profile: PATH="$HOME/bin:$PATH"
.profile.language-env-bak:# set PATH so it includes user's private bin if it exists
.profile.language-env-bak: PATH="$HOME/bin:$PATH"
.viminfo:?/PATH.xcyrillic: XNLSPATH=/usr/X11R6/lib/X11/nls
.xcyrillic: export XNLSPATH
The regular expression .[!.]*, means exclude any file or directory name starting with '..', e.g. match only .* files
Note that to use the grep PATH .[!.]* on FreeBSD you will have to use this regular expression in bash shell, the default BSD csh or tsch shells will not recognize the regular expression, e.g.:
grep PATH '.[!.]*'
grep: .[!.]*: No such file or directory
Hence on BSD, if you need to look up for a string within the home directory, hidden files: .profile .bashrc .bash_profile .cshrc run it under bash shell:
freebsd# /usr/local/bin/bash
[root@freebsd:/home/hipo]# grep PATH .[!.]*
.bash_profile:# set PATH so it includes user's private bin if it exists
.bash_profile:# PATH=~/bin:"${PATH}"
.bash_profile:# do the same with …
Another easier to remember, alternative grep cmd is:
hipo@noah:~$ grep PATH .*
.profile:PATH=/bin:/usr/bin/:${PATH}
.profile:export PATH
.profile:# set PATH so it includes user's private bin if it exists
.profile: PATH="$HOME/bin:$PATH"
….
Note that grep 'string' .* is a bit different in meaning, as it will not prevent grep to match filenames with names ..filename1, ..filename2 etc.
Though grep 'string' .* will work note that it will sometimes output some unwanted matches if filenames with double dot in the beginning of file name are there …
That's all folks 
If you're new here, you may want to subscribe to my RSS feed. Thanks for visiting!
Tags: Auto, bash shell, bit, BSD, cache cache, cmd, consuming, csh, cshrc, directory name, Draft, export path, expression, file, freebsd, gftp, grep, hipo, home directory, level directories, Linux, MANPATHAnother, nbsp, necessery, noah, Path, path profile, private bin, profile path, quot, regular expression, rli, root, set path, Shell, shrc, text, text strings, time, time consuming, tsch, value, XNLSPATH, XNLSPATHThe, zcompdump
Posted in Linux, System Administration | 2 Comments »
Thursday, April 19th, 2012
I'm doing replication for one server. Its not the first time I do configure replication between two MySQL database nodes, however since I haven't done it for a few years, my "know how" has mostly vanished so I had some troubles in setting it up. Once I followed some steps to configure replication I had to check if the two MASTER / Slave MySQL db nodes communicate properly. Hence I decided to drop a short post on that just in case if someone has to do the same or if I myself forget how I did it so I can check later on:
1. Check if MASTER MySQL server node is configured properly
The standard way to check a MySQL master node status info is with:
mysql> show master status;
+——————+———-+———————————————————+——————+
| File | Position | Binlog_Do_DB | Binlog_Ignore_DB |
+——————+———-+———————————————————+——————+
| mysql-bin.000007 | 106 | database1,database2,database3 | |
+——————+———-+———————————————————+——————+
1 row in set (0.00 sec)
By putting \G some extra status info is provided:
mysql> show master status\G;
*************************** 1. row ***************************
File: mysql-bin.000007
Position: 106
Binlog_Do_DB: database1,database2,database3
Binlog_Ignore_DB:
1 row in set (0.00 sec)
ERROR:
No query specified
2. Check if Slave MySQL node is configured properly
To check status of the slave the cmd is:
mysql> show slave status;
The command returns an output like:
mysql> show slave status;+———————————-+————-+————-+————-+—————+——————+———————+————————-+—————+———————–+——————+——————-+——————————————————-+———————+——————–+————————+————————-+—————————–+————+————+————–+———————+—————–+—————–+—————-+—————+——————–+——————–+——————–+—————–+——————-+—————-+———————–+——————————-+—————+—————+—————-+—————-+| Slave_IO_State | Master_Host | Master_User | Master_Port | Connect_Retry | Master_Log_File | Read_Master_Log_Pos | Relay_Log_File | Relay_Log_Pos | Relay_Master_Log_File | Slave_IO_Running | Slave_SQL_Running | Replicate_Do_DB | Replicate_Ignore_DB | Replicate_Do_Table | Replicate_Ignore_Table | Replicate_Wild_Do_Table | Replicate_Wild_Ignore_Table | Last_Errno | Last_Error | Skip_Counter | Exec_Master_Log_Pos | Relay_Log_Space | Until_Condition | Until_Log_File | Until_Log_Pos | Master_SSL_Allowed | Master_SSL_CA_File | Master_SSL_CA_Path | Master_SSL_Cert | Master_SSL_Cipher | Master_SSL_Key | Seconds_Behind_Master | Master_SSL_Verify_Server_Cert | Last_IO_Errno | Last_IO_Error | Last_SQL_Errno | Last_SQL_Error |+———————————-+————-+————-+————-+—————+——————+———————+————————-+—————+———————–+——————+——————-+——————————————————-+———————+——————–+————————+————————-+—————————–+————+————+————–+———————+—————–+—————–+—————-+—————+——————–+——————–+——————–+—————–+——————-+—————-+———————–+——————————-+—————+—————+—————-+—————-+| Waiting for master to send event | HOST_NAME.COM | slave_user | 3306 | 10 | mysql-bin.000007 | 106 | mysqld-relay-bin.000002 | 251 | mysql-bin.000007 | Yes | Yes | database1,database2,database3 | | | | | | 0 | | 0 | 106 | 407 | None | | 0 | No | | | | | | 0 | No | 0 | | 0 | |+———————————-+————-+————-+————-+—————+——————+———————+————————-+—————+———————–+——————+——————-+——————————————————-+———————+——————–+————————+————————-+—————————–+————+————+————–+———————+—————–+—————–+—————-+—————+——————–+——————–+——————–+—————–+——————-+—————-+———————–+——————————-+—————+—————+—————-+—————-+
As you can see the output is not too readable, as there are too many columns and data to be displayed and this doesn't fit neither a text console nor a graphical terminal emulator.
To get more readable (more verbose) status for the SQL SLAVE, its better to use command:
mysql> show slave status\G;
Here is a sample returned output:
mysql> show slave status\G;*************************** 1. row *************************** Slave_IO_State: Waiting for master to send event Master_Host: HOST_NAME.COM Master_User: slave_user Master_Port: 3306 Connect_Retry: 10 Master_Log_File: mysql-bin.000007 Read_Master_Log_Pos: 106 Relay_Log_File: mysqld-relay-bin.000002 Relay_Log_Pos: 251 Relay_Master_Log_File: mysql-bin.000007 Slave_IO_Running: Yes Slave_SQL_Running: Yes Replicate_Do_DB: database1,database2,database3 Replicate_Ignore_DB: Replicate_Do_Table: Replicate_Ignore_Table: Replicate_Wild_Do_Table: Replicate_Wild_Ignore_Table: Last_Errno: 0 Last_Error: Skip_Counter: 0 Exec_Master_Log_Pos: 106 Relay_Log_Space: 407 Until_Condition: None Until_Log_File: Until_Log_Pos: 0 Master_SSL_Allowed: No Master_SSL_CA_File: Master_SSL_CA_Path: Master_SSL_Cert: Master_SSL_Cipher: Master_SSL_Key: Seconds_Behind_Master: 0Master_SSL_Verify_Server_Cert: No Last_IO_Errno: 0 Last_IO_Error: Last_SQL_Errno: 0 Last_SQL_Error: 1 row in set (0.00 sec)ERROR: No query specified
If show master status or shwo slave status commands didn't reveal replication issue, one needs to stare at the mysql log for more info.
Tags: Auto, Cipher, cmd, condition, configure, Connect, Draft, event, exec, host, info, key, Master, master node, master slave, mysql server, name, Path, quot, Replicate, replication, Retry, server node, show, slave status, someone, Space, SQL, time, User
Posted in MySQL, System Administration, Web and CMS | 4 Comments »
Tuesday, February 7th, 2012
Many younger people, might not know lpr command, historically it was heavily used for printing in the early GNU / Linux days.
lpr ships the text to be printed to the printer which is physically attached on LPT (Line Print Terminal) parallel port . Those who lived the DOS era surely know in those "ancient" days, everyone who wanted to print has to use the LPT parallel port
Present time, everyone knows there is almost no modern printer that is attached to the PC via LPT port but rather the USB port is used for communication between the printer the computer.
Nevertheless The USB printers on Linux are managed by CUPS, the lpr command is still functional shipping the text to be printed via CUPS (cups-lpd daemon).
Before cups-lpd was introduced the service managing the print jobs was lpd
Hence lpr is still functional.
To print a plain text file of one page with lpr on Linux:
linux:~# cat text-file-to-print.txt | lpr
For multiple printers to switch between multiple printers there is the PRINTER shell variable:
linux:~# export PRINTER=printer-Name-and-Type
To print a really long text file (a book in TXT) the pr command comes handy. As you can read in the cmd manual pr – converts text files for printing
Lets say you would like to have a 60 lines of text per printed page, the cmd to issue is:
linux:~# pr -l60 text-file-to-print.txt | lrp
All queued printing jobs can be reviewed with the lpq, if you have a printer attached try:
linux:~# lpq
lp is ready and printing
Rank Owner Job Files Total Size
active hipo 1 text-file-to-print.txt 62045 bytes
Since some years it is pretty rare for people to use lpq, since most of the parallel printing is managed by CUPS server, what most people use nowdays to check the printer queue is lpstat : e.g.
linux:~# lpstat
...
Printing status and all things related to queued jobs for printing gets logged in /var/log/lpr.log
There is even more simplistic way to print directly to the printer (if the printer is attached via a LPT port) through the kernel /dev/lp, for example:
linux:~# cat text-file-to-print.txt >> /dev/lp
For more than one printer attached the naming of /dev/lp, might probably be /dev/lp0, /dev/lp1 etc.
The lprm command also exists in case if you would like to cancel a printjob in the queue. Lets say I want to cancel a job in the queue with Job ID 5:
linux:~# lrpm 5
...
To cancel a current running job in the middle the /usr/bin/cancel command exists.
An interesting historic fact is that nowdays opening lpr, lpq or any of the other tools for simple text mode printing one sees on top of the page Apple Inc.
Lets clear this up CUPS (Common Unix Printing System) (open source) printing platform is not owned by Apple, since it is licensed under GPL2 and LGPL. The reason why the Apple Inc. shows up in man pages is because in year 2007, the founder of CUPS printing server Michael Sweet hired him to work for Apple Inc. "purchasing" the CUPS source. However as we know they did not really purchased the code, because the code was already belonging to the community (licensed under GPL2). Apple however as a marketing trick used the fact that Sweet worked for them and as probably as a matter of marketing asked him to place the Apple Inc. in the copyright source and manual areas. Obviously this is not true, since Apple Inc. does not hold copyright for CUPS as CUPS can be copied by anyone (its open source) 
Most of the people will never print using this commands, since printing is now, ages ahead, anyways for simple people (like me), who just need to print a text with no special fonts or graphics text printing is just great.
Text printing is also a good learning experience for Linux novices and is good to be known just as a piece of UNIX history.
Tags: Auto, cmd, code, Copyright, Draft, everyone, file, Files, gnu linux, job, jobs, Linux, linux pr, lpd, lpd daemon, lpq, lpstat, page, parallel port, parallel printing, present time, Print, printer name, printer queue, printing, printingRank, queue, Shell, shipping, ships, Sizeactive, terminal, text, time, usb port, usb printers
Posted in Linux, System Administration | 2 Comments »
Saturday, January 14th, 2012
Mounting ISO files in Linux is easy with mount cmd, however remembering the exact command one has to issue is a hard task because mounting ISO files is not a common task.
Mounting ISO files directly by clicking on the ISO file is very nice, especially for lazy people uninitiated with the command line
Besides that I'm sure many Windows users are curious if there is an equivallent program to DaemonTools for Linux / BSD*?
The answer to this question is YES!
There are two major programs which can be used as a DaemonTools substitute on Linux:
These are FuriousISOMount and AcetoneISO
AcetoneISO is more known and I've used it some long time ago and if I'm correct it used to be one of the first ISO Mount GUI programs for Linux. There is a project called GMount-ISO / (GMountISO) which of the time of writting this article seems to be dead (at least I couldn't find the source code).
Luckily FuriousISOMount and AcetoneISO are pretty easy to install and either one of the two is nowdays existing in most Linux distributions.
Probably the programs can also be easily run on BSD platform also quite easily using bsd linux emulation.
If someone has tried something to mount GUIs in Free/Net/OpenBSD, I'll be interesting to hear how?
1. Mount ISO files GUI in GNOME with Furius ISO Mount
FuriousISOMount is a simple Gtk+ interface to mount -t iso9660 -o loop command.
To start using the program on Debian / Ubuntu install with apt;
debian:~# apt-get install furiusisomount
The following extra packages will be installed:
fuseiso fuseiso9660 libumlib0
The following NEW packages will be installed:
furiusisomount fuseiso fuseiso9660 libumlib0
…
To access the program in GNOME after install use;
Applications -> Accessories -> Furious ISO Mount
When mounting it is important to choose Loop option to mount the iso instead of Fuse
After the program is installed to associate the (.iso) ISO files, to permanently be opened with furiusisomount roll over the .iso file and choose Open With -> Other Application -> (Use a custom command) -> furiusisomount
2. Mount ISO Files in KDE Graphical Environment with AcetoneISO
AcetoneISO is build on top of KDE's QT library and isway more feature rich than furiousisomount.
Installing AcetoneISO Ubuntu and Debian is done with:
debian:~# apt-get install acetoneiso
The following NEW packages will be installed:
acetoneiso gnupg-agent gnupg2 libksba8 pinentry-gtk2 pinentry-qt4
0 upgraded, 6 newly installed, 0 to remove and 35 not upgraded.
Need to get 3,963 kB of archives.
After this operation, 8,974 kB of additional disk space will be used.
...
AcetoneISO supports:
- conversion between different ISO formats
- burn images to disc
- split ISO image volumes
- encrypt images
- extract password protected files
Complete list of the rich functionality AcetoneISO offers is to be found on http://www.acetoneteam.org/viewpage.php?page_id=6
To start the program via the GNOME menus use;
Applications -> Accessories -> Sound & Video -> AcetoneISO
I personally don't like AcetoneISO as I'm not a KDE user and I see the functionality this program offers as to rich and mostly unnecessery for the simple purpose of mounting an ISO.
3. Mount ISO image files using the mount command
If you're a console guy and still prefer mounting ISO with the mount command instead of using fancy gui stuff use:
# mount -t iso9660 -o loop /home/binary/someiso.iso /home/username/Iso_Directory_Name
Tags: AcetoneISO, Auto, BSD, cmd, custom, debian gnu, Draft, emulation, exact command, file, Files, Free, GMount-ISO, Gnome, graphical environment, gtk interface, guis, image files, ISO, iso file, iso files, iso9660, libumlib, long time, loop command, most linux distributions, nbsp, Open, OpenBSD, option, platform, roll, someone, something, source code, substitute, time, Ubuntu, windows users, YES
Posted in Gnome, Linux, Linux and FreeBSD Desktop, Linux Audio & Video | No Comments »
Monday, January 9th, 2012
I've recently had to set up a backup system to synchronize backup archive files between two remote servers and as I do usually with this situation I just set up a crontab job to periodically execute rsync to copy data from source server to the destination server . Copying SRC to DEST is the default behaviour rsync uses, however in this case I had to copy from the destination server to the source server host (in other words sync files the reversely.
The usual way to copy with rsync via SSH (from SRC to DEST) is using a cmd line like:
debian:~$ /usr/bin/rsync -avz -e ssh backup-user@xxx.xxx.xxx.xxx:/home/backup-user/my-directory .
Where the xxx.xxx.xxx.xxx is my remote server IP with which files are synched.
According to rsync manual, the proposed docs SYNOPSIS is in the format;
Local: rsync [OPTION...] SRC… [DEST
Obviusly the default way to use rsync is to copy source to destination which I used until now, but in this case I had to the other way around and copy files from a destination host to the source server. It was logical that swapping the SRC and DEST would complete my required task. Anyways I consulted with some rsync gurus in irc.freenode.net , just to make sure it is proper to just swap the SRC, DEST arguments.
I was told this is possible, so I swapped args;
debian:~$ /usr/bin/rsync -avz -e ssh . backup-user@xxx.xxx.xxx.xxx:/home/backup-user/my-directory
...
Surprisingly this worked Anyways I was adviced by by a good guy nick named scheel , that putting -e ssh to command line is generally unnecessery except if there is no some uncommon used SSH port over which the data is transferred. An example case in which -e 'ssh is necessery would be if transferring via lets say SSH port 1234;
rsync -avz -e 'ssh -p1234' /source user@host:/dest
In all other cases omitting '-e ssh' is better as '-e ssh' is rsync default. Therefore my final swapped line I put in cron to copy from a destinatio to source host with rsync looked like so:
05 03 2 * * /usr/bin/ionice -c 3 /usr/bin/rsync -avz my-directory backup-user@xxx.xxx.xxx.xxx:/home/backup-user/ >/dev/null 2>&1
Tags: archive files, Auto, avz, backup system, case, cmd, cmd line, copy, copying, cron, default, default behaviour, destIn, destinatio, destination, destination host, destination server, docs, Draft, example, example case, few words, format, gurus, host, job, nbsp, necessery, nick, option, port 1234, remote server, rsync, scheel, server host, server ip, source host, source server, ssh port, sync files, system, usr bin
Posted in FreeBSD, Linux, System Administration | No Comments »
Thursday, December 15th, 2011
I'm writting some game reviews and movie reviews, every now and then and therefore being able to capture a fullscreen running program like let's say mplayer or vlc or some full screen running game is something I really need.
The usual PrtScr button which normally works to prepare screenshots in GNOME or KDE, however is not working if the root window handler is being passed to a different program than the Window Manager and pressing it while inside of many older programs or applicationsdoes not produce a print screen of the current screen.
Anyways I found a hack to this using the good old ImageMagick – import command line screenshotting program.
To take a screenshot of a certain program run from gnome-terminal or konsole using import cmd its possible to use a quick one liner which will take a snapshot of the root Window the started program will use.
Let's say you want to make a screenshot of the entry screen of the FreeDOOM (DooM 3d shooter classical game arcade free Software Alternative).
Launch gnome-terminal or konsole , xterm , depending on the GUI environment you use and issue the commands:
debian:~$ ( sleep 15; import -window root my_desired_screenshot_name.png ) &
debian:~$ freedoom
The first command will launch import after a sleep of 15 secs and therefore will screenshot the active window which will be at focus after 15 seconds, where the & sign will background it and the second one will launch FreeDooM . You will have to wait for a certain secs and switch to the exact screen you will want to screenshot.
If you want to screenshot some game scene that will appear in 20 minutes change above sleep 15 cmd to be to something like sleep 180
That method can be used for screenshotting any other program running on fullscreen, the method is a bit inflexible as you will have to adjust a timing but it works fine
Tags: bit, Button, classical game, cmd, Desktop, desktop environment, exact screen, first command, FreeDooM, fullscreen, game arcade, game reviews, game scene, Gnome, GNU, gnu linux, gui environment, how to take a screenshot, ImageMagick, import, import command, import window, launch, Linux, mplayer, PrtScr, root, root window, running game, screen, screenshot, shooter, sign, sleep, software, something, timing, VLC, xterm
Posted in FreeBSD, Linux, Linux and FreeBSD Desktop, Various | 1 Comment »
Tuesday, November 29th, 2011
The classical way to keep a process running in background after log out of a shell is using screen
Anyways using screen is not the only way to detach a running process , GNU / Linux and BSDs (Free, Open, Net BSDs) had a command nohup which aim is to run a command immune to hangups, with output to a non-tty
Let's say one wants to keep track (log) constantly ICMP traffic to a certain host with ping command on a UNIX server / desktop which doesn't have the screen manager … terminal emulation program installed. Achieving this task is possible with nohup cmd by backgrounding the ping process, like so:
guoi@host:~$ nohup ping google.com >ping.log &[1] 45931hipo@host:~$ nohup: ignoring input and redirecting stderr to stdout
Afterwards even after closing up the opened ssh session or console (tty) / terminal (pts) on which the ping process is background nohup prevents the ping to be sent kill SIGNAL so the process continues running in the background.
Later on to check in real time the statistics of the continuous ICMP ping requests tail, less, or cat can be used for example watching the tail:
hipo@host:~$ tail -f ping.log
64 bytes from fx-in-f106.1e100.net (74.125.39.106): icmp_req=562 ttl=51 time=44.0 ms
64 bytes from fx-in-f106.1e100.net (74.125.39.106): icmp_req=563 ttl=51 time=43.8 ms
64 bytes from fx-in-f106.1e100.net (74.125.39.106): icmp_req=564 ttl=51 time=43.3 ms
64 bytes from fx-in-f106.1e100.net (74.125.39.106): icmp_req=565 ttl=51 time=43.1 ms
64 bytes from fx-in-f106.1e100.net (74.125.39.106): icmp_req=566 ttl=51 time=43.4 ms
64 bytes from fx-in-f106.1e100.net (74.125.39.106): icmp_req=567 ttl=51 time=43.6 ms
I'm using Linux / BSD for quite a lot of time and never before put in use the nohup cmd I guess there are more ppl who never heard of this handy UNIX basic command. Hope I'm not the only one who never heard about it and its useful knowledge to someone out. Cheers
Tags: aim, BSDs, cheer, Cheers, cmd, com, Desktop, detach, Free, fx, gnu linux, google, Hope, host, ICMP, icmp ping, log amp, manager, msi, net, nohup, Open, ping, ping command, ppl, real time, req, screen, server desktop, Shell, someone, ssh, ssh session, stderr, terminal, terminal emulation program, time, tty, unix server, useful knowledge
Posted in FreeBSD, Linux, System Administration | No Comments »
Wednesday, November 23rd, 2011
I needed to show the Administrator user on one Windows 7 install.
Achieving this is done through command prompt – cmd.exe where the command prompt with the exclusive option of Run as Administrator .
The exact command that unhides the Administrator user so further on on next windows login screen one sees Administrator user ready for use is:
C:\> net user administrator /active:yes
Unhiding Administrator user is always handy whether one needs to do some bunch of operations with Super User. After finishing all my required tasks with administrator I reverted back and hid the Administrator user once again like so:
C:> net user administrator /active:no
This commands also works fine on Vista and presumably on Windows XP.
Tags: Achieving, administrator, bunch, cmd, command, command prompt, exact command, exe, fine, login, option, run, screen, show, Super, unhide, Unhiding, use, User, Vista, Windows, windows xp
Posted in System Administration, Various, Windows | No Comments »
Wednesday, October 19th, 2011
I've been using Debian GNU / Linux on my Thinkpad laptop for almost 3 years and half. Initially the Debian version which I had installed was a stable Debian Lenny. As I was mostly dissatisfied of the old versions of the programs, I migrated to testing / unstable
Testing / unstables shipped program versions were a bit better but still back in the day I wanted to get advantage of the latest program versions so for a while I switched to unstable .
Later I regretted for this bad idea, after the migration to Unstable, it was too buggy to run on a notebook one uses for everyday work.
Then to revert back to a bit stable I downgraded to testing unstable again.
When Debian launched Debian Squeeze I set in my /etc/apt/sources.list file software repositories to be the one for the stable Debian Squeeze.
As you can see, I've done quite a lot of "experiments" and "excersises". Many packages were installed, then removed, some became obsolete with time others I just temporary installed out of curiosity. Anyways as a result I ended up with many packages uninstalled / removed , which still kept some of their directory structres and configurations on the machine.
Today, I decided to check how many of these obsolete packages are still present in dpkg database and I was shocked to find out 412 debs were still in my package database! To check the number I used cmd:
root@noah:~# dpkg -l | grep -i '^rc\s.*$'|wc -l
Considering the tremendous number of packs waiting to be purged, I decided to get rid of this old and already unnecessery files for the sake of clarity, besides that removing the old already uninstalled packages removes old configuration files, readmes, directories and frees some little space and therefore frees some inodes
Before proceeding to remove them, I carefully reviewed and all the package names which I was about to completely purge in order to make sure there is no package with a configuration files I might need in future:
root@noah:~# dpkg -l |grep -i '^rc\s.*$'
...
After reviewing all the deb packages possessing the rc – (remove candidate) flag, I used the following bash one liners to remove the obsolete deb packages:
root@noah:~# for i in $(dpkg -l |grep -i '^rc\s.*$'|awk '{ print $2 }'); do echo dpkg --purge $i done...
root@noah:~# for i in $(dpkg -l |grep -i '^rc\s.*$'|awk '{ print $2 }'); do dpkg --purge $i done
First line will just print out what will be purged with dpkg , so after I checked it out I used the second one to purge all the RC packs.
Tags: 3 years, bad idea, bit, buggy, candidate, clarity, cmd, configuration files, curiosity, debian version, debs, dpkg, everyday work, file, half, inodes, laptop, Linux, little space, machine, need, noah, number, old versions, package, package database, package names, program versions, quot, root, software, software repositories, squeeze, system, system directories, time, Ubuntu, work
Posted in Linux, Linux and FreeBSD Desktop, Linux Audio & Video, System Administration, Various | 3 Comments »
Wednesday, September 28th, 2011
To find out which programs are causing the most read/write overhead on a Linux server one can use iotop
Here is the description of iotop – simple top-like I/O monitor, taken from its manpage.
iotop does precisely the same as the classic linux top but for hard disk IN/OUT operations.
To check the overhead caused by some daemon on the system or some random processes launching iotop without any arguments is enough;
debian:~# iotop
The main overview of iostat statistics, are the:
Total DISK READ: xx.xx MB/s | Total DISK WRITE: xx.xx K/s
If launching iotop, shows a huge numbers and the server is facing performance drop downs, its a symptom for hdd i/o overheads.
iotop is available for Debian and Ubuntu as a standard package part of the distros repositories. On RHEL based Linuxes unfortunately, its not available as RPM.
While talking about keeping an eye on hard disk utilization and disk i/o's as bottleneck and a possible pitfall to cause a server performance down, it's worthy to mention about another really great tool, which I use on every single server I administrate. For all those unfamiliar I'm talking about dstat
dstat is a – versatile tool for generating system resource statistics as the description on top of the manual states. dstat is great for people who want to have iostat, vmstat and ifstat in one single program.
dstat is nowdays available on most Linux distributions ready to be installed from the respective distro package manager. I've used it and I can confirm tt is installable via a deb/rpm package on Fedora, CentOS, Debian and Ubuntu linuces.
Here is how the tool in action looks like:
The most interesting things from all the dstat cmd output are read, writ and recv, send , they give a good general overview on hard drive performance and if tracked can reveal if the hdd disk/writes are a bottleneck to create server performance issues.
Another handy tool in tracking hdd i/o problems is iostat its a tool however more suitable for the hard core admins as the tool statistics output is not easily readable.
In case if you need to periodically grasp data about disks read/write operations you will definitely want to look at collectl i/o benchmarking tool .Unfortunately collect is not included as a packaget for most linux distributions except in Fedora. Besides its capabilities to report on servers disk usage, collect is also capable to show brief stats on cpu, network.
Collectl looks really promosing and even seems to be in active development the latest tool release is from May 2011. It even supports NVidia's GPU monitoring In short what collectl does is very similar to sysstat which by the way also has some possibilities to track disk reads in time. collectl's website praises the tool, much and says that in most machines the extra load the tool would add to a system to generate reports on cpu, disk and disk io is < 0.1%. I couldn't find any data online on how much sysstat (sar) extra loads a system. It will be interesting if some of someone concluded some testing and can tell which of the two puts less load on a system.
Tags: action, bottleneck, CentOS, cmd, description, Disk, disk utilization, drive, drive performance, drop, eye, generating system, GNU, gnu linux, handy tool, interesting things, Linux, linux server, linux top, linuxes, manager, manual states, most linux distributions, overhead, package, performance drop, performance issues, pitfall, program, random processes, recv, repositories, resource statistics, server performance, sIf, system, system resource, tool, Ubuntu, versatile, versatile tool, vmstat
Posted in Linux, System Administration, Various | 6 Comments »
on 