u123's questions

I would like to create a few VMs with different max IOPS - e.g. 100, 200 and 500 - for various bench-marking tests.

I have tried VMWare Workstation Player for Linux but it does not have options to configure this. I have also tried to get a free trial of VMWare vSphere ESXi but so far no luck with that.

Are there any easy/simple free alternative for linux where I can create multiple VMs with a specified max IOPS?

Looks like its possible in Hyper-V for Windows: https://www.youtube.com/watch?v=qbWVFS4UyH0

but I need a Linux solution

In Ubuntu I am trying to create a docker network with a specified subnet when on VPN but I get the below error:

$ docker network create my-network --subnet 172.17.24.0/18
Error response from daemon: Pool overlaps with other one on this address space

Looking at the route table:

$ route -n
Kernel IP routing table
Destination     Gateway         Genmask         Flags Metric Ref    Use Iface
0.0.0.0         0.0.0.0         0.0.0.0         U     0      0        0 cscotun0
0.0.0.0         192.168.0.1     0.0.0.0         UG    20600  0        0 wlp0s20f3
10.174.33.0     0.0.0.0         255.255.255.0   U     0      0        0 cscotun0
169.254.0.0     0.0.0.0         255.255.0.0     U     0      0        0 cscotun0
169.254.0.0     0.0.0.0         255.255.0.0     U     1000   0        0 wlp0s20f3
172.17.0.0      0.0.0.0         255.255.0.0     U     0      0        0 docker0
188.64.152.45   192.168.0.1     255.255.255.255 UGH   0      0        0 wlp0s20f3
192.168.0.0     0.0.0.0         255.255.255.0   U     0      0        0 cscotun0
192.168.0.0     0.0.0.0         255.255.255.0   U     600    0        0 wlp0s20f3
192.168.0.1     0.0.0.0         255.255.255.255 UH    0      0        0 wlp0s20f3

I don't see that my VPN (cscotun0) is occupying that range so what is going on?

Unless 172.17.24.0/18 is overlapping with docker0:

172.17.0.0      0.0.0.0         255.255.0.0     U     0      0        0 docker0

but I assume both docker0 and 172.17.24.0/18 should be able to exist in the same range?

I have this working ldapsearch command for connecting anonymously to an internal ldap server:

$ ldapsearch -h my.host -p 389 -x -LLL -s base -b "" namingContexts
dn:
namingContexts: DC=org,DC=sample

Now I would like to list the same using curl:

$ curl -v "ldap://my.host:389/dc=org,dc=sample namingContexts"
*   Trying ...
* Connected to my.host (....) port 389 (#0)
* LDAP remote: search failed Success 000004DC: LdapErr: DSID-0C090A5C, comment: In order to perform this operation a successful bind must be completed on the connection., data 0, v4563
* Closing connection 0
curl: (39) LDAP remote: search failed Success 000004DC: LdapErr: DSID-0C090A5C, comment: In order to perform this operation a successful bind must be completed on the connection., data 0, v4563

So I get a connection to the server but how do I list the namingContext without providing credentials/anonymously with curl.

I am running a apache2 docker container on my ubuntu box but it keeps writing logs to my host os. Details below:

samples/docker/apache2/Dockerfile

FROM php:5.6.34-apache
COPY ./sites-available/mysite.conf /etc/apache2/sites-available/mysite.conf
RUN a2dissite 000-default.conf
#RUN a2dissite default-ssl.conf
RUN a2ensite mysite.conf

with:

samples/docker/apache2/sites-available/mysite.conf:

<VirtualHost *:80>
    ServerAdmin [email protected]
    ServerName localhost
    ServerAlias localhost
    #DocumentRoot /var/www/html
    DocumentRoot /var/www


    ErrorLog /dev/stderr
    CustomLog /dev/stdout combined

    #ErrorLog ${APACHE_LOG_DIR}/error.log
    #CustomLog ${APACHE_LOG_DIR}/access.log combined
    
</VirtualHost>

that I build/run with:

samples/docker-compose.yaml:

version: '3.8'
services:
  apache2-php:
    container_name: apache2-php-container
    image: apache2-php-image
    build:
      context: ./docker/apache2
    ports:
      - "8082:80"

But log, run and lock folders are created on my host under the root user one folder above my docker-compose.yaml file:

$ ll
total 60
drwxrwxr-x   3 user user  4096 Nov 15 09:22 samples/
drwxr-xr-x   3 root root  4096 Nov 15 10:40 lock/
drwxr-xr-x   3 root root  4096 Nov 15 10:40 log/
drwxr-xr-x   3 root root  4096 Nov 15 10:40 run/

with:

├── log
│   └── apache2
│       ├── error.log
│       └── other_vhosts_access.log

I have also tried suggestions here:

Writing Apache2 Logs to stdout/stderr?

but it has no effect.

Why are those folders/files created at that location on my host and how do I configure apache2 container to only write logs inside the container and not on my host?

Using this ldif file:

# LDIF Export for cn=sample,dc=example,dc=org
# Server: ldap-service (ldap-service)
# Search Scope: sub
# Search Filter: (objectClass=*)
# Total Entries: 3
#
# Generated by phpLDAPadmin (http://phpldapadmin.sourceforge.net) on February 8, 2018 7:38 pm
# Version: 1.2.3

version: 1

# Entry 1: cn=sample,dc=example,dc=org
dn: cn=sample,dc=example,dc=org
cn: sample
gidnumber: 500
objectclass: posixGroup
objectclass: top

# Entry 2: cn=John Doe,cn=sample,dc=example,dc=org
dn: cn=John Doe,cn=sample,dc=example,dc=org
cn: John Doe
gidnumber: 500
givenname: John
homedirectory: /home/users/jdoe
loginshell: /bin/sh
objectclass: inetOrgPerson
objectclass: posixAccount
objectclass: top
sn: Doe
uid: jdoe
uidnumber: 1000
userpassword: {MD5}X03MO1qnZdYdgyfeuILPmQ==

# Entry 3: cn=Sally Hudson,cn=sample,dc=example,dc=org
dn: cn=Sally Hudson,cn=sample,dc=example,dc=org
cn: Sally Hudson
gidnumber: 500
givenname: Sally
homedirectory: /home/users/shudson
loginshell: /bin/sh
objectclass: inetOrgPerson
objectclass: posixAccount
objectclass: top
sn: Hudson
uid: shudson
uidnumber: 1001
userpassword: {MD5}X03MO1qnZdYdgyfeuILPmQ==

I have populated an openldap server running on localhost on default port 389 and verified the connection/content:

Now I would like to test that the user: John Doe can login using ldapsearch. I have tried:

ldapsearch -x -h localhost -p 389 -x -D 'cn=sample,cn=John Doe,dc=example,dc=org' -w password
ldapsearch -x -h localhost -p 389 -x -D 'cn=John Doe,dc=example,dc=org' -w password
ldapsearch -x -h localhost -p 389 -x -D 'uid=jdoe,cn=sample,dc=example,dc=org' -w password

but they all fail with:

ldap_bind: Invalid credentials (49)

even after I have tried to update the password (using Apache Directory Studio).

Also tried updating with a plain text password: qwerty

$ ldapsearch -x -h localhost -p 389 -x -D 'uid=jdoe,cn=sample,dc=example,dc=org' -w 'qwerty'
ldap_bind: Invalid credentials (49)

Same error.

Any suggestions?

I am running apache2 2.4.10 in a debian docker container using below docker-compose file:

version: '3.8'
services:
  apache2-php:
    container_name: apache2-php-container
    image: apache2-php-image
    build:
      context: ./docker/apache2
    ports:
      - "8082:80"
    volumes:
      - ./docker/apache2/index.php:/var/www/index.php
 

with this site.

<VirtualHost *:80>
    ServerAdmin [email protected]
    ServerName localhost
    ServerAlias localhost
    DocumentRoot /var/www
    ErrorLog ${APACHE_LOG_DIR}/error.log
    CustomLog ${APACHE_LOG_DIR}/access.log combined
</VirtualHost>

Its up an running just fine but for some reason logs are written to stdout/stderr instead of above specified files. Also I see:

$ ls -la /var/log/apache2
total 8
drwxrwxrwx 2 www-data www-data 4096 Dec 29  2018 .
drwxr-xr-x 1 root     root     4096 Dec 29  2018 ..
lrwxrwxrwx 1 www-data www-data   11 Dec 29  2018 access.log -> /dev/stdout
lrwxrwxrwx 1 www-data www-data   11 Dec 29  2018 error.log -> /dev/stderr
lrwxrwxrwx 1 www-data www-data   11 Dec 29  2018 other_vhosts_access.log -> /dev/stdout

Do I need to add some option in my site to force apache2 to write to the actual files instead of to stdout/stderr?

I currently host content on the following paths:

http://oldsite/samples/content/
http://oldsite/samples/ql
http://oldsite/samples/api/content/plans/
http://oldsite/samples/api/streams/map.xml
...

I have a new site coming up which will eventually serve that content under:

http://newsite/content/
http://newsite/ql
http://newsite/api/content/plans/
http://newsite/api/streams/map.xml
...

Meaning the samples subpath will be removed.

So basically the inverted version of:

Nginx redirect one path to another

During this transition my plan is to rewrite those path using nginx. Currently I have this configured:

/etc/nginx.conf.d/sandbox.conf

server {

    #listen 80;
    #listen [::]:80;

    listen 8082 default_server;
    listen [::]:8082 default_server;


    location / {
         proxy_pass http://localhost:8080/;
         proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
         proxy_set_header X-Forwarded-Proto $scheme;
         proxy_set_header X-Forwarded-Port $server_port;
         
         location /content/ {
            rewrite 301 /samples/content/;
         }

         location /ql {
            rewrite 301 /samples/ql;
         }

         location /api/content/plans/ {
            rewrite 301 /samples/api/content/plans/;
         }

         location /api/streams/map.xml {
            rewrite 301 /samples/api/streams/map.xml;
         }
    }

}

But is there a more clean way to do this - using some regex/wildcard pattern so I don't have to write out each path?

Also if I instead do:

/etc/nginx.conf.d/sandbox.conf

server {

    #listen 80;
    #listen [::]:80;

    listen 8082 default_server;
    listen [::]:8082 default_server;

    location / {
         proxy_pass http://localhost:8080/;
         proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
         proxy_set_header X-Forwarded-Proto $scheme;
         proxy_set_header X-Forwarded-Port $server_port;
         
         location /samples/content/ {
            rewrite 301 /content/;
         }

         location /samples/ql {
            rewrite 301 /ql;
         }

         location /samples/api/content/plans/ {
            rewrite 301 /api/content/plans/;
         }

         location /samples/api/streams/map.xml {
            rewrite 301 /api/streams/map.xml;
         }
    }

}

I get:

$ curl localhost:8082/content
<html>
<head><title>404 Not Found</title></head>
<body>
<center><h1>404 Not Found</h1></center>
<hr><center>nginx/1.18.0 (Ubuntu)</center>
</body>
</html>

and access log:

, request: "GET /content HTTP/1.1", host: "localhost:8082"
2021/03/15 10:05:18 [error] 81448#81448: *4 open() "/usr/share/nginx/html/content" failed (2: No such file or directory), client: 127.0.0.1, server: , request: "GET /content HTTP/1.1", host: "localhost:8082"

Where bypassing nginx I get:

curl localhost:8080/samples/content
Hello!

I want to try out JFrog artifactory OSS using docker compose. I have found:

https://jfrog.com/download-jfrog-platform/

https://api.bintray.com/content/jfrog/artifactory-pro/org/artifactory/pro/docker/jfrog-artifactory-pro/$latest/jfrog-artifactory-pro-$latest-compose.tar.gz;bt_package=jfrog-artifactory-pro

But it seems its only for the PRO/Paid version.

I have found an old/deprecated OSS docker-compose file here:

https://github.com/jfrog/artifactory-docker-examples/blob/master/docker-compose/artifactory/artifactory-oss.yml

Does JFrog no longer support OSS when it comes to docker-compose installations?

Based on this guide I am trying to send a test email using telnet from linux

https://linuxconfig.org/send-an-email-using-telnet

but the connection immediately disconnects:

$ telnet smtp.gmail.com 465
Trying 108.177.126.108...
Connected to smtp.gmail.com.
Escape character is '^]'.
MAIL FROM: [email protected] closed by foreign host

How do I keep the connection open long enough to send my test mail?

I am running this python script on a remote ubuntu server from command line for speed testing:

https://github.com/sivel/speedtest-cli

with:

$ curl -s https://raw.githubusercontent.com/sivel/speedtest-cli/master/speedtest.py | python3 -
Retrieving speedtest.net configuration...
Testing from Microsoft Corporation (x.x.x.x)...
Retrieving speedtest.net server list...
Selecting best server based on ping...

But it just hangs. I have then tried ( 5463) Linkem SPA (Rome, Italy)):

$ python3 speedtest.py 5463
Retrieving speedtest.net configuration...
Testing from Microsoft Corporation (x.x.x.x)...
Retrieving speedtest.net server list...
Retrieving information for the selected server...
Hosted by Linkem SPA (Rome) [1297.45 km]: 1800000.0 ms
Testing download speed................................................................................
Download: 0.00 Mbit/s
Testing upload speed

So its not connected at all. Also running:

$ ping sl-03.wemacom.de

Or:

$ hping3 sl-03.wemacom.de
HPING sl-03.wemacom.de (eth0 46.30.119.166): NO FLAGS are set, 40 headers + 0 data bytes

Or:

$ hping3 sl-03.wemacom.de --syn -p 8080

just hangs. On my local machine it works fine:

$ ping sl-03.wemacom.de
PING sl-03.wemacom.de (46.30.119.166) 56(84) bytes of data.
64 bytes from sl-03.wemacom.de (46.30.119.166): icmp_seq=1 ttl=53 time=42.6 ms
64 bytes from sl-03.wemacom.de (46.30.119.166): icmp_seq=2 ttl=53 time=44.4 ms

Maybe I need to run the test one layer up - e.g. on TCP level - on this machine?

In Openshift I am used to expose my service outside the cluster through a route. That will automatically create a new "domain name" that I can use to access my application from e.g. my local PC on the local network.

That new doman name is not automatically made public to the internet but it works automatically on the local network so I assume there is some DNS server running - maybe as part of the OpenShift platform - where these routes/DNS names gets added/removed from automatically.

Now I am working on kubernetes and have created an Ingress object for my service/application:

apiVersion: extensions/v1beta1
kind: Ingress
metadata:
  annotations:
    kubernetes.io/ingress.class: nginx
  name: my-app
  namespace: my-ns
spec:
  rules:
  - host: ???
    http:
      paths:
      - backend:
          serviceName: my-app
          servicePort: 8080
        path: /hello

But how do I fill out the host?

Do I manually have to contact the DNS server admin each time I need to do this and get him to create a new DNS record with a uniqe name on the local network?

Or is there some standard/automated way to do this in Kubernetes - similar to OpenShift?

First a bit of background:

I currently use VMWare ESXi for my virtual machines. Some servers needs to be up 24/7 and others I tear down and re-create from templates when I need to.

After playing around with docker I see I can accomplish much of the above and also much faster (at least for my Linux machines).

But what I like about VMware ESXi is a good graphical overview of the global resources for the host, all running/stopped servers, their hardware resources and the ability with a few clicks to add more RAM, CPU or disk space to individual servers.

Regarding tweaking memory/CPU seems you can do the same in docker:

https://docs.docker.com/engine/admin/resource_constraints/#configure-the-realtime-scheduler

just a bit more low-level.

So at the moment I don't really see a path where I could completely move from a VMWare/ESXi managed infrastructure to a pure docker infrastructure and based on:

https://www.upguard.com/articles/docker-vs.-vmware-how-do-they-stack-up

it sounds like this is not really the point with docker either - to replace existing hypervisors like VMWare/Hyper-V etc.

Then the question:

Or are the companies that run a pure docker setup for their whole machine farm and if so how do they manage resources and keep an overview of all servers?

I just did a fresh install of jenkins master on a private virtual server having 1 GB of RAM. I have not yet installed anything else besides Jenkins on that server. I reduced the number of executors to 1. When I start the jenkins services it consumes approx. 700 MB of RAM just being idle:

What is up with all those child processes??

As a comparison here is after I stopped the Jenkins service:

We would like to move our git repository to an atlassian stash installation. I have created a new and empty repository in stash and from our existing git repository on my local machine I have executed:

git push stash-remote refs/remotes/origin/*:refs/heads/*

Where I have configured/added the stash-remote in the repository I am standing in. It shows up fine in the stash web interface.

Is that all I need to do to get a complete copy of our git repository into stash so our developer can continue working on the new stash repository?

In vSphere 5.1.0 I have created a snapshot of a windows 7 machine. I would to restore the machine to this snapshot each time its restarted. But it only seems to be possible if I manually go to the menu and select Revert to current snapshot.

How do I force the machine to restore to the snapshot each time its rebooted?

I am trying to reinstall python2.6 on debian 6 but I get:

root@ff: apt-get install --reinstall python2.6
Reading package lists... Done
Building dependency tree       
Reading state information... Done
You might want to run 'apt-get -f install' to correct these:
The following packages have unmet dependencies:
 python-pkg-resources : Depends: python (< 2.7) but 2.7.5-5 is to be installed
 python-setuptools : Depends: python (< 2.7) but 2.7.5-5 is to be installed
E: Unmet dependencies. Try 'apt-get -f install' with no packages (or specify a solution).

Notice that starting python2.6 works fine:

:~$ python2.6
Python 2.6.6 (r266:84292, Dec 26 2010, 22:31:48) 
[GCC 4.4.5] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> 

But its the application that I need to use which depend on python that is now broken, e.g pip:

~$ pip
Traceback (most recent call last):
  File "/usr/bin/pip", line 7, in <module>
    from pip.log import logger
ImportError: No module named pip.log

I previously installed python2.7 and have used:

update-alternatives --config python

to switch back to 2.6.But it seems that application installed using python 2.6 has been corrupted by the 2.7 install.

I have tried to uninstall python2.7. But it says that its not installed, even though I can start it from a terminal:

# apt-get purge python2.7
Reading package lists... Done
Building dependency tree       
Reading state information... Done
Package python2.7 is not installed, so not removed
0 upgraded, 0 newly installed, 0 to remove and 460 not upgraded.
# python
Python 2.7.6 (default, Jan 11 2014, 14:34:26) 
[GCC 4.8.2] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> 

It seems that a complete re-install of the system is required, but I would like to avoid that if possible.

And also tried restoring an older status file :

# cp /var/backups/dpkg.status.2  /var/lib/dpkg/status
# apt-get clean
# apt-get update

And my /etc/apt/sources list:

# deb cdrom:[Debian GNU/Linux 6.0.4 _Squeeze_ - Official amd64 NETINST Binary-1 20120129-00:39]/ squeeze main

#deb cdrom:[Debian GNU/Linux 6.0.4 _Squeeze_ - Official amd64 NETINST Binary-1 20120129-00:39]/ squeeze main

deb http://ftp.se.debian.org/debian/ squeeze main
deb-src http://ftp.se.debian.org/debian/ squeeze main

deb http://security.debian.org/ squeeze/updates main
deb-src http://security.debian.org/ squeeze/updates main

# squeeze-updates, previously known as 'volatile'
deb http://ftp.se.debian.org/debian/ squeeze-updates main
deb-src http://ftp.se.debian.org/debian/ squeeze-updates main
deb http://mirror.cse.iitk.ac.in/debian/ testing main contrib

Now of I try to install python-twisted I get:

# apt-get install python-twisted
Reading package lists... Done
Building dependency tree       
Reading state information... Done
The following packages were automatically installed and are no longer required:
  iso-codes libserf-0-0 python-apt-common lsb-release
Use 'apt-get autoremove' to remove them.
The following extra packages will be installed:
  cpp-4.4 g++-4.4 gcc-4.4 gcc-4.4-base gcc-4.8-base libapr1 libapr1-dev libaprutil1 libaprutil1-dbd-sqlite3 libaprutil1-dev libaprutil1-ldap libc-dev-bin libc6
  libc6-dbg libc6-dev libc6-i386 libdb5.1 libexpat1 libexpat1-dev libffi6 libgcc1 libgmp10 libgomp1 liblzma5 libmpfr4 libpython-dev libpython-stdlib
  libpython2.7 libpython2.7-dev libpython2.7-minimal libpython2.7-stdlib libsasl2-2 libsasl2-modules libsasl2-modules-db libsctp-dev libsctp1 libserf-1-1
  libsqlite3-0 libsqlite3-dev libssl1.0.0 libstdc++6 libstdc++6-4.4-dev libsvn1 libtinfo5 libxml2 linux-headers-2.6.32-5-common locales make multiarch-support
  python python-crypto python-dev python-libxml2 python-lxml python-minimal python-openssl python-pkg-resources python-setuptools python-sqlite python-support
  python-svn python-twisted-bin python-twisted-conch python-twisted-core python-twisted-lore python-twisted-mail python-twisted-names python-twisted-news
  python-twisted-runner python-twisted-web python-twisted-words python-yaml python-zope.interface python2.7 python2.7-dev python2.7-minimal subversion

The above list contains a bunch of python2.7 dependencies....where do they come from? Should it not be possible to install python-twisted for python2.6?

On ubuntu 13.10 I can install gcovr just fine: sudo apt-get install gcovr But on debian 7 I get:

# apt-get install gcovr
Reading package lists... Done
Building dependency tree       
Reading state information... Done
E: Unable to locate package gcovr

How do I install gcovr on debian 7 which I need for this:

http://meekrosoft.wordpress.com/2010/06/02/continuous-code-coverage-with-gcc-googletest-and-hudson/

I have a filezilla server runnning on a remote windows 7 machine. I have a running jenkins master with a slave installed on debian 7.

If I ssh to the debian7 slave and do a ftp/put to the filezilla server I can upload a test.txt file just fine:

$ ftp myfileZillaFtp 21
Connected to myfileZillaFtp.
220-FileZilla Server version 0.9.41 beta
220-written by Tim Kosse ([email protected])
220 Please visit http://sourceforge.net/projects/filezilla/
Name (myfileZillaFtp:jenkins): admin
331 Password required for admin
Password:
230 Logged on
Remote system type is UNIX.
ftp> cd test
250 CWD successful. "/test" is current directory.
ftp> put test.txt
local: test.txt remote: test.txt
200 Port command successful
150 Opening data channel for file transfer.
226 Transfer OK
6364 bytes sent in 0.00 secs (24959.2 kB/s)
ftp> 

But if I run the same step in a Jenkins freestyle job where I have added a FTP publish step based on the instruction here:

https://wiki.jenkins-ci.org/display/JENKINS/Publish+Over+FTP+Plugin

I get:

FTP: Connecting from host [debian7-slave]
FTP: Connecting with configuration [myfileZillaFtp] ...
220-FileZilla Server version 0.9.41 beta
220-written by Tim Kosse ([email protected])
220 Please visit http://sourceforge.net/projects/filezilla/
FTP: Logging in, command printing disabled
FTP: Logged in, command printing enabled
CWD test
250 CWD successful. "/test" is current directory.
FTP: Remote root is not absolute, getting absolute directory from PWD
PWD
257 "/test" is current directory.
TYPE I
200 Type set to I
CWD /test
250 CWD successful. "/test" is current directory.
PASV
227 Entering Passive Mode (172,22,80,8,192,46)
FTP: Disconnecting configuration [myfileZillaFtp] ...
ERROR: Exception when publishing, exception message [Connection timed out]
Build step 'Send files over FTP' changed build result to UNSTABLE
Finished: UNSTABLE

Why does the ftp upload fail when running it through jenkins?

EDIT:

Most likely a bug in: https://wiki.jenkins-ci.org/display/JENKINS/Publish+Over+FTP+Plugin

It works with this plugin: http://wiki.hudson-ci.org/display/HUDSON/FTP-Publisher+Plugin

Any suggestions are still welcome though.

I am trying to configure my debian 6 machine to use DHCP. I have modified the /etc/network/interfaces to only contain:

auto eth0
iface eth0 inet dhcp

But when I reboot it hang at NFS Common utilities: statd . After around 5 minutes comes to the login screen, any ideas?

I have installed Ubuntu 12.04.1 LTS (GNU/Linux 3.2.0-23-generic x86_64). Some info about the machine:

$ grep --color "model name" /proc/cpuinfo
model name  : Intel(R) Xeon(R) CPU           E5430  @ 2.66GHz
model name  : Intel(R) Xeon(R) CPU           E5430  @ 2.66GHz
model name  : Intel(R) Xeon(R) CPU           E5430  @ 2.66GHz
model name  : Intel(R) Xeon(R) CPU           E5430  @ 2.66GHz

I need to install jdk5 to support an old application.

I have tried:

~$ sudo apt-get install openjdk-5-jdk
Reading package lists... Done
Building dependency tree       
Reading state information... Done
E: Unable to locate package openjdk-5-jdk

I have also tried:

~$ sudo apt-get install sun-java5-jdk
Reading package lists... Done
Building dependency tree       
Reading state information... Done
E: Unable to locate package sun-java5-jdk

So its not available in the repos. I have tried to follow this guide (adding the jaunty repos):

http://leonardo-pinho.blogspot.dk/2010/11/java-15-no-ubuntu-1010.html

but same result.

Then I have tried to download jdk-1_5_0_22-linux-i586.bin from here:

http://www.oracle.com/technetwork/java/javasebusiness/downloads/java-archive-downloads-javase5-419410.html#jdk-1.5.0_22-oth-JPR

and do:

~$ chmod a+x jdk-1_5_0_22-linux-i586.bin 
~$ sudo ./jdk-1_5_0_22-linux-i586.bin 
Sun Microsystems, Inc.  Binary Code License Agreement

yes
Unpacking...
Checksumming...
0
0
Extracting...
./jdk-1_5_0_22-linux-i586.bin: 424: ./jdk-1_5_0_22-linux-i586.bin: ./install.sfx.19556: not found
./jdk-1_5_0_22-linux-i586.bin: 1: cd: can't cd to jdk1.5.0_22

Any suggestions?