Many programs allow to be run either in daemon mode (which is usually the default), or explicitly run it in foreground with a flag such as -f.
Now when configuring upstart to run programs like this, I can either do
exec foo -f # run in foreground
or
expect fork
exec foo # automatically runs as a daemon
Is there a reason to choose one over the other? It feels that it's just easier to run everything in foreground and let upstart handle the processes, but is there a downside to that?
I have an app (in the form of a single binary) that I need to run on one of my servers, and I'd like to be able to control its lifecycle properly. What I'd like is the following:
What I'm looking for is a lightweight solution that ideally works with Ubuntu Server out of the box, without much configuration.
I've installed my nginx via the ppa package
add-apt-repository ppa:nginx
apt-get install nginx
which works just fine. But now I need to add support for websockets for my socket.io app. I found some tutorials, but generally the process is to download the source and recompile nginx with the tcp_proxy module.
Is there any way I can do this with the package manager, so that I don't have to compile it manualy? I'm running Ubuntu 10.04 if that's of any relevance.
I have a .pid file, and I need to check if the process is running. So far I found two options
kill -0 `cat something.pid`
which prints out an error if the pid isn't running. I know this can be redirected to /dev/null, but it makes me think that this isn't the best solution.
The second solution would be to use ps, which however also prints on the STDOUT
ps -ef `cat something.pid`
Is it normal to redirect the output to /dev/null and just use the status code returned, or is it a sign that I'm doing something wrong and I need a different command?
It often happens to me, that I just used a command and I need to copy/paste it either to a different terminal window/IM or whatever.
On a Mac, I can use the pbcopy command to get something in clipboard, but I'm not sure how should I get to the command. What I've came up is
history | tail -n 1 | perl -ple 's/\d+\s+//' | pbcopy
which works, but it doesn't really seem like clean solution. Is there a better way to get to the last command? I'm using zsh, but I guess if there's a solution for bash it will work in zsh too.
Within the next couple of days, I'm going to need two Linux VPS instances (probably Ubuntu 10.04), one running PHP and one for Ruby on Rails. Both are most likely going to use nginx and MySQL and have about 256-512MB RAM.
Now the question is, should I go with a 32bit or 64bit system? There is a related question already posted, but without any specific information on how to decide.
How does 32bit vs 64bit system behave from your experience in these conditions? (Ruby/PHP, nginx, MySQL)
Say that I have an application running on one PC that is sending commands via SSH to another PC on the network (both machines running Linux).
For example every time something happens on #1, I want to run a task on #2. In this setup, I have to create SSH connection on every single command.
Is there any simple way to do this with basic unix tools without programming custom client/server application? Basically all I want is to establish a connection over SSH and then send one command after another.
I just installed Calibre for ebook management via apt-get on Ubuntu 10.04, however I found out that it's one major version behind the current release, so I decided to reinstall it directly from sources.
When I uninstalled the packaged version, apt added bunch of dependencies to the autoremove queue, and as I installed newer version of Calibre from sources, it has no knowledge of it being dependent on those packages.
Now I basically have all libraries that I want, but they are still in the autoremove queue.
The following packages were automatically installed and are no longer required:
libqt4-script libqt4-designer libqt4-dbus python-lxml python-cherrypy3
python-encutils libqt4-xmlpatterns libqt4-help python-qt4 python-clientform
python-sip python-django python-mechanize libqt4-svg python-django-tagging
libphonon4 libqt4-xml libqt4-assistant libqt4-webkit libqt4-scripttools
python-beautifulsoup python-pypdf python-dateutil python-cssutils
Use 'apt-get autoremove' to remove them.
How do I tell apt that I want to keep these packages installed, without reinstalling them manually?
Say that I have a PHP site running on that hosts larger images. Each page is very simple and takes virtually no time to process in PHP, but it might contain couple MB large image.
The question is, if I have Apache set up to handle 100 connections, and the page itself takes couple ms to generate, does it keep the connection blocked for the image transfer, or is this handled on some other connection pool?
Because in the first case, it might take couple seconds to download the image, so I would be able to serve only 100 clients at a time.
What is the simplest way to enable ANSI escape codes (eg. colored output) in Windows cmd?
I have this problem with server running CakePHP application. The server is insanely slow, I first thought that it's application problem, but then I found constant 5-6MB/s write to disk.
What is the easiest way to find cause of such a heavy write?
The server is running Gentoo.
In past few weeks I found out, that my Linux sysadmin skills are pretty low, so I bought bigger HDD (1.5TB) and started playing with virtual machines. I'm experimenting pretty hard way, so I end up reinstalling a lot. My problem is, I don't really know, what is the best way to do initial setup.
My first idea was to create simple bash script, that would install all necessary packages, compile sources, set up accounts etc. But my problem was, that it usualy got stuck in the middle somewhere, because of some change or problem, so I had to manualy hack it anyway. Also such a script will outdate very fast.
Another idea, which I'm trying now, is to write a checklist with things I need to do, and then manually install everything and check things on the paper. This has advantage of having more generic steps (like install MySQL), however I have to do everything manualy, which isn't very convenient.
Last options is install a whole system manualy, configure it, and then create a image of partition, and restore it eventualy instead of clean instalation. While this seems to me like good solution for installing on say 20 servers at once, it will also be outdated soon, so I'd have to create new, up to date, clean installation and create it's image again.
I know that different situation applies different measures, so there probably isn't one fit all solution.
So my question is.
Do you have any kind of checklist, installation script, or any kind of that thing, when you are installing a new system?
I'm having problem with one of my Rails applications, that runs on 365MB VPS, which seems to be not very much. When there are more than 3-5 users at the same time it runs out of memory and starts to use swap, which is f****** slow.
As I look into top, I can see that there are many spawned instances of mysql (much more than Rails processes). Is this normal?
How do you recommend to scale down application for low memory enviromnent?
I'm using Ubuntu 9.04, Apache2.2 with mod passenger, MySQL 5.075 and Ruby 1.8.7
edit: After your suggestions, I upgraded VPS to 540MB, which for now seems to suffice. However, I'm just a poor student, so I can't pump too much money in some project I'm doing mostly for myself, untill it starts to pay back at least something.
Say I have a server and client. I need to create connection from client to a website through server like it was proxy.
Is it possible to do this using a SSH tunel, or do I have to install some proxy service to the server?
I have set up my main domain for Google Apps (intecro.com) and I've set alias domain in Google Apps for (intecro.cz).
I have administrator account [email protected] and it automaticaly gave me nickname [email protected], which is great, but I am unable to add other nickname @intecro.cz. It automaticaly suffixes every nickname with @intecro.com.
Is it even possible to set email nickname for an aliased domain?
I need to run script, that takes long time to execute, or I just want it to run forever. I can't just SSH to my machine, because when I disconnect it stops running.
Is there any way to run script that isn't dependent on the shell that started it?
I'm using Ubuntu 9.04.
I've commented out the bind-address line in my /etc/mysql/my.cnf
When I do nmap localhost, I see that mysql is correctly listening on 3306, but when I try to connect, I get error 1045.
Do I have to somehow specify whitelist of allowed ips (I've seen this on some shared hosts), or is it possible to allow access from any ip?
I'm using Ubuntu 9.04 x64.
What is the best way how to empty a bunch of files in bash? As far I've been doing this
echo "" > development.log
echo "" > production.log
I don't really want to delete those files, so rm is not possible. I've tried many things like
echo "" > *.log
but nothing worked.
I've created super simple rails app
rails test
cd test/
script/generate controller home index
rm public/index.html
# uncommented map.root :controller => "home" in config/routes.rb
When I run app in development mode, it works just fine and everything get logged
<VirtualHost *:80>
DocumentRoot /opt/rails/test/public
ErrorLog /opt/rails/test/log/error.log
CustomLog /opt/rails/testlog/access.log combined
RailsEnv development
</VirtualHost>
but when I change to RailsEnv production,
RailsEnv production
I get this ugly error page
We're sorry, but something went wrong.
We've been notified about this issue and we'll take a look at it shortly.
and there is absolutely nothing in either error.log or production.log, however, in access.log, I can see the request being logged.
I'm running Ubuntu x64 Jaunty.
I've installed passenger as ruby gem. I've added following to /etc/apache2/apache2.conf
LoadModule passenger_module /var/lib/gems/1.8/gems/passenger-2.2.4/ext/apache2/mod_passenger.so
PassengerRuby /usr/bin/ruby
PassengerRoot /var/lib/gems/1.8/gems/passenger-2.2.4
and set Vitual Host to rails app ~/public directory, but when I visit the site, I only see content of dispatch.fcgi, but nothing gets executed
.htaccess in ~/public directory is
AddHandler fcgid-script .fcgi
AddHandler cgi-script .cgi
Options +FollowSymLinks +ExecCGI
RewriteEngine On
RewriteRule ^$ index.html [QSA]
RewriteRule ^([^.]+)$ $1.html [QSA]
RewriteCond %{REQUEST_FILENAME} !-f
RewriteRule ^(.*)$ dispatch.fcgi [QSA,L]
However, if I run script/server, everything works fine, so there is no problem with the application itself.