Luc's questions

I'd like to try out coreos on our little cluster (4 machines). I’d like to install coreos on each machine’s disk.

I’ve got a trusty64 live cd that I will use to run the coreos-install script. Is there a default cloud-config file I could use to set this up ?

I want to give each machine the private ip 192.168.5.i

Would this file be enough to init my cluster when installing coreos on disk on my first machine (192.168.5.1) ? http://pastie.org/private/njub0e4nahb10enh9nnw

#cloud-config
ssh_authorized_keys:
  - ssh-rsa MYKEY
coreos:
  etcd:
    discovery: https://discovery.etcd.io/MYTOKEN
    addr: 192.168.5.1:4001
    peer-addr: 192.168.5.1:7001
  fleet:
    public-ip: 192.168.5.1
  units:
    - name: etcd.service
      command: start
    - name: fleet.service
      command: start

Seems like when installing on disk $private_ipv4 is not interpreted (do I have to enter the ip then ?)

I need to configure 3 networks on my Ubuntu machine (OpenStack Network node)

The /etc/network/interfaces I have is:

# interfaces(5) file used by ifup(8) and ifdown(8)
auto lo
iface lo inet loopback

# OpenStack management interface
auto eth1
iface eth1 inet static
  address 192.168.1.13
  netmask 255.255.255.0
  gateway 192.168.1.1
  broadcast 192.168.1.255
  dns-nameservers 8.8.8.8

# OpenStack instance tunnels interface
auto eth1:0
iface eth1:0 inet static
  address 192.168.3.1
  netmask 255.255.255.0

# OpenStack The external network interface
auto eth2
iface eth2 inet manual
  up ip link set dev $IFACE up
  down ip link set dev $IFACE down

eth1 and eth1:0 are working fine. Also, when I issue ifconfig, only eth1 and eth1:1 appear in the list. but no eth2.

I also tested to defined the 3rd interface on eth1:1 (3 networks on the same eth1 interface) with:

auto eth1:1
iface eth1:1 inet manual
  up ip link set dev $IFACE up
  down ip link set dev $IFACE down

or on eth0, but in both cases this 3rd interfaces does not appear with ifconfig, how could this be possible ?

The machine has 2 NICs:

lspci | grep Ethernet
00:19.0 Ethernet controller: Intel Corporation 82579LM Gigabit Network Connection (rev 04)
02:00.0 Ethernet controller: Intel Corporation 82574L Gigabit Network Connection

How can I know the name of the interfaces linked to them ? It seems eth1 is working but how can I make sure eth0 is working ?

** UPDATE **

Using eth2 was a stupid thing. Instead of trying to use one single interface to map the 3 networks, I have used eth0 and plugged the NIC correctly.

auto eth0
iface eth0 inet manual
  up ip link set dev $IFACE up
  down ip link set dev $IFACE down

I have a fresh Debian 6.0 install. I have installed build-essential and added:

deb http://backports.debian.org/debian-backports squeeze-backports main

in /etc/apt/sources.list

I now try to install Postgres 9.1 but I have the following error:

The following packages have unmet dependencies:
 postgresql-9.1 : Depends: libpq5 (>= 9.1~) but 8.4.12-0squeeze1 is to be installed
                  Depends: postgresql-client-9.1 but it is not going to be installed
                  Depends: postgresql-common (>= 115~) but 113 is to be installed

Is there any additional stuff to install prior the postgres installation ?

I have a ubuntu 11.04 box running 24/7. It started to crash any couple of days recently.
I checked in kern.log, dmesg... but did not find anything there. Is there a specific place where shutdown reason (temperature, or whatever) are logged ?

I have an app running with:

• one instance of nginx as the frontend (serving static file)
• a cluster of node.js application for the backend (using cluster and expressjs modules)
• one instance of Postgres as the DB

Is this architecture sufficient if the application needs scalability (this is only for HTTP / REST requests) for:

• 500 request per seconds (each requests only fetches data from the DB, those data could be several ko, and with no big computation needed after the fetch).

• 20000 users connected at the same time

Where could be the bottlenecks ?

I have an iphone app using ASIHttpRequest. The server code is on heroku in node.js

From time to time, a single request is sent from the iphone (only one trace) app but it is received twice on herokuapp (I can see twice the same request in the heroku logs).

I though at the beginning the request was requested twice because of an error in the first attempt but it's not the case as both request (the one I need and the second one I don't need) are performed on server side.

Any idea ?

I need to redirect the traffic to one backend or another according to the user-agent. Is that the right thing to do ?

server {
    listen      80;
    server_name my_domain.com;

    if ($http_user_agent ~ iPhone ) {
        rewrite     ^(.*)   https://m.domain1.com$1 permanent;
    }
    if ($http_user_agent ~ Android ) {
        rewrite     ^(.*)   https://m.domain1.com$1 permanent;
    }
    if ($http_user_agent ~ MSIE ) {
        rewrite     ^(.*)   https://domain2.com$1 permanent;
    }
    if ($http_user_agent ~ Mozilla ) {
        rewrite     ^(.*)   https://domain2.com$1 permanent;
    }
}

I use nginx to proxy https request to my application server (currently running on 8443).
This app server serves dynamic pages, some of them including jquery minified.
The pages are in error as it seems jquery is truncated... is there a file size limit or something ?

My nginx conf is the following one:

server {
listen      443;
server_name my_serv.com;
ssl                 on;
ssl_certificate     certificate.pem;
ssl_certificate_key privatekey.pem;

keepalive_timeout 70;

location / {
  proxy_pass https://localhost:8443;
  proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
  proxy_set_header Host $http_host;
  proxy_buffering off;
  proxy_buffers 8 8k;
}
}

I use nginx with proxy path directive.
When the application to which the request is proxied return a response, it seems nginx add some header containing the Content-Length. Is that possible to remove this additional header ?

UPDATE

I have re-installed nginx with the more_headers module but I still have the same result. My config is:

upstream my_sock {
   server unix:/tmp/test.sock
   fail_timeout=0;
}

server {
   listen 11111;
   client_max_body_size 4G;
   server_name localhost;

   keepalive_timeout 5;

   location / {
      more_clear_headers 'Content-Length';
      proxy_pass http://my_sock;
      proxy_redirect off;
      proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
      proxy_set_header Host $http_host;
    }
}

I am running ruby1.9 on Debian 5, and did not manage to install 'unicorn' with rubygems. I got this error and do not really know how to solve it.

Do you have any idea of the possible root cause ?

> gem install unicorn


Building native extensions.  This could take a while...
ERROR:  Error installing unicorn:
ERROR: Failed to build gem native extension.

    /usr/bin/ruby1.9 extconf.rb
checking for CLOCK_MONOTONIC in time.h... yes
checking for clockid_t in time.h... yes
checking for clock_gettime() in -lrt... yes
checking for t_open() in -lnsl... no
checking for socket() in -lsocket... no
checking for poll() in poll.h... yes
checking for getaddrinfo() in sys/types.h,sys/socket.h,netdb.h... yes
checking for getnameinfo() in sys/types.h,sys/socket.h,netdb.h... yes
checking for struct sockaddr_storage in sys/types.h,sys/socket.h... yes
checking for accept4() in sys/socket.h... no
checking for sys/select.h... yes
checking for ruby/io.h... yes
checking for rb_io_t.fd in ruby.h,ruby/io.h... yes
checking for rb_io_t.mode in ruby.h,ruby/io.h... yes
checking for rb_io_t.pathv in ruby.h,ruby/io.h... no
checking for struct RFile in ruby.h,ruby/io.h... yes
checking size of struct RFile in ruby.h,ruby/io.h... 24
checking for struct RObject... no
checking size of int... 4
checking for rb_io_ascii8bit_binmode()... no
checking for rb_thread_blocking_region()... yes
checking for rb_thread_io_blocking_region()... no
checking for rb_str_set_len()... yes
checking for rb_time_interval()... yes
checking for rb_wait_for_single_fd()... no
creating Makefile

make
cc -I. -I/usr/include/ruby-1.9.0/x86_64-linux -I/usr/include/ruby-1.9.0 -I. -DHAVE_TYPE_CLOCKID_T -DHAVE_POLL -DHAVE_GETADDRINFO -DHAVE_GETNAMEINFO -DHAVE_TYPE_STRUCT_SOCKADDR_STORAGE -DHAVE_SYS_SELECT_H -DHAVE_RUBY_IO_H -DHAVE_RB_IO_T_FD -DHAVE_ST_FD -DHAVE_RB_IO_T_MODE -DHAVE_ST_MODE -DHAVE_TYPE_STRUCT_RFILE -DSIZEOF_STRUCT_RFILE=24 -DSIZEOF_INT=4 -DHAVE_RB_THREAD_BLOCKING_REGION -DHAVE_RB_STR_SET_LEN -DHAVE_RB_TIME_INTERVAL    -D_GNU_SOURCE -DPOSIX_C_SOURCE=1-D_POSIX_C_SOURCE=200112L -fPIC -fno-strict-aliasing -g -g -O2 -O2 -g -Wall -Wno-parentheses  -fPIC  -o kgio_ext.o -c kgio_ext.c
cc -I. -I/usr/include/ruby-1.9.0/x86_64-linux -I/usr/include/ruby-1.9.0 -I. -DHAVE_TYPE_CLOCKID_T -DHAVE_POLL -DHAVE_GETADDRINFO -DHAVE_GETNAMEINFO -DHAVE_TYPE_STRUCT_SOCKADDR_STORAGE -DHAVE_SYS_SELECT_H -DHAVE_RUBY_IO_H -DHAVE_RB_IO_T_FD -DHAVE_ST_FD -DHAVE_RB_IO_T_MODE -DHAVE_ST_MODE -DHAVE_TYPE_STRUCT_RFILE -DSIZEOF_STRUCT_RFILE=24 -DSIZEOF_INT=4 -DHAVE_RB_THREAD_BLOCKING_REGION -DHAVE_RB_STR_SET_LEN -DHAVE_RB_TIME_INTERVAL    -D_GNU_SOURCE -DPOSIX_C_SOURCE=1-D_POSIX_C_SOURCE=200112L -fPIC -fno-strict-aliasing -g -g -O2 -O2 -g -Wall -Wno-parentheses  -fPIC  -o autopush.o -c autopush.c
cc -I. -I/usr/include/ruby-1.9.0/x86_64-linux -I/usr/include/ruby-1.9.0 -I. -DHAVE_TYPE_CLOCKID_T -DHAVE_POLL -DHAVE_GETADDRINFO -DHAVE_GETNAMEINFO -DHAVE_TYPE_STRUCT_SOCKADDR_STORAGE -DHAVE_SYS_SELECT_H -DHAVE_RUBY_IO_H -DHAVE_RB_IO_T_FD -DHAVE_ST_FD -DHAVE_RB_IO_T_MODE -DHAVE_ST_MODE -DHAVE_TYPE_STRUCT_RFILE -DSIZEOF_STRUCT_RFILE=24 -DSIZEOF_INT=4 -DHAVE_RB_THREAD_BLOCKING_REGION -DHAVE_RB_STR_SET_LEN -DHAVE_RB_TIME_INTERVAL    -D_GNU_SOURCE -DPOSIX_C_SOURCE=1-D_POSIX_C_SOURCE=200112L -fPIC -fno-strict-aliasing -g -g -O2 -O2 -g -Wall -Wno-parentheses  -fPIC  -o wait.o -c wait.c
cc -I. -I/usr/include/ruby-1.9.0/x86_64-linux -I/usr/include/ruby-1.9.0 -I. -DHAVE_TYPE_CLOCKID_T -DHAVE_POLL -DHAVE_GETADDRINFO -DHAVE_GETNAMEINFO -DHAVE_TYPE_STRUCT_SOCKADDR_STORAGE -DHAVE_SYS_SELECT_H -DHAVE_RUBY_IO_H -DHAVE_RB_IO_T_FD -DHAVE_ST_FD -DHAVE_RB_IO_T_MODE -DHAVE_ST_MODE -DHAVE_TYPE_STRUCT_RFILE -DSIZEOF_STRUCT_RFILE=24 -DSIZEOF_INT=4 -DHAVE_RB_THREAD_BLOCKING_REGION -DHAVE_RB_STR_SET_LEN -DHAVE_RB_TIME_INTERVAL    -D_GNU_SOURCE -DPOSIX_C_SOURCE=1-D_POSIX_C_SOURCE=200112L -fPIC -fno-strict-aliasing -g -g -O2 -O2 -g -Wall -Wno-parentheses  -fPIC  -o connect.o -c connect.c
cc -I. -I/usr/include/ruby-1.9.0/x86_64-linux -I/usr/include/ruby-1.9.0 -I. -DHAVE_TYPE_CLOCKID_T -DHAVE_POLL -DHAVE_GETADDRINFO -DHAVE_GETNAMEINFO -DHAVE_TYPE_STRUCT_SOCKADDR_STORAGE -DHAVE_SYS_SELECT_H -DHAVE_RUBY_IO_H -DHAVE_RB_IO_T_FD -DHAVE_ST_FD -DHAVE_RB_IO_T_MODE -DHAVE_ST_MODE -DHAVE_TYPE_STRUCT_RFILE -DSIZEOF_STRUCT_RFILE=24 -DSIZEOF_INT=4 -DHAVE_RB_THREAD_BLOCKING_REGION -DHAVE_RB_STR_SET_LEN -DHAVE_RB_TIME_INTERVAL    -D_GNU_SOURCE -DPOSIX_C_SOURCE=1-D_POSIX_C_SOURCE=200112L -fPIC -fno-strict-aliasing -g -g -O2 -O2 -g -Wall -Wno-parentheses  -fPIC  -o poll.o -c poll.c

poll.c:11:18: error: st.h: No such file or directory
poll.c: In function 'do_poll':
poll.c:148: error: 'RUBY_UBF_IO' undeclared (first use in this function)
poll.c:148: error: (Each undeclared identifier is reported only once
poll.c:148: error: for each function it appears in.)
make: *** [poll.o] Error 1


Gem files will remain installed in /usr/lib/ruby/gems/1.9.0/gems/kgio-2.5.0 for inspection.
Results logged to /usr/lib/ruby/gems/1.9.0/gems/kgio-2.5.0/ext/kgio/gem_make.out

I have 2 aplications, each one running on unicorn.
I'd like to use nginx as a front end for each of those application.
For each nginx / unicorn pair I would then use a dedicated unix socket.
To do so, do I need to have to instance of nginx or can I use 2 virtual hosts ?

I have installed nginx 1.0.3 using the following procedure:

apt-get install libpcre3-dev build-essential libssl-dev
cd /opt/
wget http://nginx.org/download/nginx-1.0.3.tar.gz
tar -zxvf nginx-1.0.3.tar.gz
cd /opt/nginx-1.0.3/
./configure --prefix=/opt/nginx --user=nginx --group=nginx --with-http_ssl_module
make
make install
adduser --system --no-create-home --disabled-login --disabled-password --group nginx

The installation seems ok. I then need to create 2 virtual hosts but I cannot find any "nginx/sites-available" folder in /etc. nginx is only located in /opt (where I installed it) and so sites-available/enabled anywhere.

I will end up adding the virtualhost in the /opt/nginx/conf/nginx.conf but this is not what I want to do.

Can somebody help ?

I have uncommented the following line to have php beeing taken into account by apache:

LoadModule php5_module        libexec/apache2/libphp5.so

I then created a test.php file with

<?php phpinfo(); ?>

and restarted apache

sudo apachectl graceful

But the php file is not taken into account, I just get the plain html.

Any idea ?

I have installed kannel in a virtual machine (ubuntu 10.04). I did not change any of the default config and tested the fakesmsc. This work fine. When I plug my iPhone in USB, I do not see any /dev/ttyUSB showing up (I though this was how modem was mounted). Could you please help ?

Thanks a lot,

Luc

I have just setup a Lucid Lynx desktop machine. I have installed VirtualBox 3.2.6 on this machine to run several server. I have created a new Lucid Lynx server in one VM, the VM creation is fine but when I try to take a snapshot, it seems the action starts but never ends... The desktop machine is in ext4.

Is this a know problem ?

Thanks a lot,

Luc

I'm working on an app (iPhone) that makes several Http Request to a RESTFULL web server. When I lauch the app, the request is performed and very often it timeouts. If I perform the request once again (a dedicated button is in the UI for this purpose) it's quite fast (no time out). Do you have any hints on the location of this problem ? Could it be on server side (I running a rack application on nginx / passenger) ?

Thanks a lot for you help. Regards, Luc

I have a rack application that I want to migrate from Ruby 1.8.7 + Apache + passenger to Ruby 1.9.1 + Nginx + passenger.

I have made up the following script for a quick install all in one, and it raises an error...

Here is the installation script: (basic one with all the steps I need to install everything on a Ubuntu 10.04 Lucid Lynx fresh box)

Nginx sources

cd /tmp
wget http://nginx.org/download/nginx-0.7.66.tar.gz
tar xzf nginx-0.7.66.tar.gz
cd nginx-0.7.66

openssl required for SSL/TLS

sudo apt-get install openssl
sudo apt-get install libssl-dev

Compilation stuff

sudo apt-get zlib1g-dev

Ruby interpreter 1.9.1

sudo apt-get install ruby1.9.1 ruby1.9.1-dev rubygems1.9.1 irb1.9.1 \
ri1.9.1 rdoc1.9.1 build-essential nginx libopenssl-ruby1.9.1

Make sure default ruby uses version 1.9.1

sudo update-alternatives --install /usr/bin/ruby ruby /usr/bin/ruby1.9.1 400 --slave /usr/share/man/man1/ruby.1.gz ruby.1.gz /usr/share/man/man1/ruby1.9.1.1.gz --slave /usr/bin/ri ri /usr/bin/ri1.9.1 --slave /usr/bin/irb irb /usr/bin/irb1.9.1 --slave /usr/bin/rdoc rdoc /usr/bin/rdoc1.9.1
sudo update-alternatives --config ruby

Passenger (rake-0.8.7, fastthread-1.0.7, rack-1.1.0, passenger-2.2.14)

sudo gem install passenger

Activate Passenger in nginx, select option 2 to use nginx sources donwloaded above

cd /var/lib/gems/1.9.1/gems/passenger-2.2.14/bin
sudo ./passenger-install-nginx-module

And this is the error message I got:

/var/lib/gems/1.9.1/gems/passenger-2.2.14/ext/nginx/ContentHandler.c
gcc -c -pipe  -O -W -Wall -Wpointer-arith -Wno-unused-parameter -Wunused-function -Wunused-variable -Wunused-value -Werror -g   -I src/core -I src/event -I src/event/modules -I src/os/unix -I /tmp/pcre-8.00 -I objs -I src/http -I src/http/modules -I src/mail \
    -o objs/addon/nginx/StaticContentHandler.o \
/var/lib/gems/1.9.1/gems/passenger-2.2.14/ext/nginx/StaticContentHandler.c
/var/lib/gems/1.9.1/gems/passenger-2.2.14/ext/nginx/StaticContentHandler.c: In function ‘passenger_static_content_handler’:
/var/lib/gems/1.9.1/gems/passenger-2.2.14/ext/nginx/StaticContentHandler.c:71: error: ‘ngx_http_request_t’ has no member named ‘zero_in_uri’
make[1]: *** [objs/addon/nginx/StaticContentHandler.o] Error 1
make[1]: Leaving directory `/tmp/nginx-0.7.66'
make: *** [build] Error 2
--------------------------------------------

It looks like something went wrong
Please read our Users guide for troubleshooting tips:
/var/lib/gems/1.9.1/gems/passenger-2.2.14/doc/Users guide Nginx.html

I do not understand the reason of this error. Is this a compatibility problem?
Hope you have any clues :)

I have several machines on my lan. One is used as a http proxy to target web sites located on the others (that's working fine now thanks to ServerFault). On my router, Port 22 is NATed to this proxy machine. I would like to be able to access the other machines, within internet, with something like:

ssh user@first_machine.my_domain.tld ssh user@second_machine.my_domain.tld

Could I use the proxy machine to 'filter' the incoming ssh request and to route them to the correct machine ? (in the same way it's possible to do so for web sites using a mix of mod_proxy and namevirtualhost in Apache)

Thanks a lot, Luc

I have several machines on my lan. On of them is running 2 web sites, first_web_site and second_web_site (each one in a dedicated NameVirtualHost). Another machine is running another site third_web_site. I would like to be able to access each one, within internet, with the url:

• first_web_site.domain.tld
• second_web_site.domain.tld
• third_web_site.domain.tld

knowing that 2 sites are on the same machine. Can Apache help me to do this ?

I have a machine that will have a apache server to be used for proxy purposes. I was talk to set up virtualhost on this one and use proxy server but I do not know how to do this.

Could you please give me hints ? Thanks a lot, Luc

I have an ip like username.dyndns.org, this is the external IP of my router. On my lan, I have several machine (m1, m2, ...) , each running a dedicated web server. Is it possible to reach each machine from the outside with something like: http://m1.username.dyndns.org http://m2.username.dyndns.org ? Do you know what needs to be configured in my router for NAT ? Also, is there a special directive in Apache to do so ? Thanks a lot, Regards, Luc