Recently active topics in CohesiveFT

Hi
Just created a lamp server, based on daisy linux, with minimum stuff : apache2, mysql, php, created as a vmware image
Everything went smootly. Except I am French and use a french keyboard. So I tried installing locales, reconfiguring console-data, but all of that does not seem to be available in daisy linux.
So should I change apt sources to regular etch ? won't that cause issues elsewhere
More generally, how should I go about changing keyboard layout to another language that english ?
Thanks for any pointer. The service is pretty useless if we do not have the correct keyboard layout !!
Cheers

Great service. However after setting up my server and deploying it to EC2 I can't access the machine via SSH.

I've confirmed that the necessary security group exists with the appropriate permissions but all attempts to SSH into the machine are met with a 'Connection Refused' response like the sshd isn't running.

The only change from the standard Rails 2.1 server is that I've associated an elastic IP address to the server before attempting the first connection.

Any pointers would be really welcome!

Thanks.

https://elasticserver.com/server/7611...

Update: I also don't have any of the additional tabs mentioned here -> http://blog.elasticserver.com/2008/03...

I have 2 EC2 servers connected to the manager with an elastic IP but now I'm trying to connect remote clients outside EC2. I have the client pack installed on the remote client, but it just times out.

Here's the message log:

Wed Jun 24 13:51:15 2009 NOTE: --user option is not implemented on Windows
Wed Jun 24 13:51:15 2009 OpenVPN 2.0.9 Win32-MinGW [SSL] [LZO] built on Oct 1 2006
Wed Jun 24 13:51:15 2009 IMPORTANT: OpenVPN's default port number is now 1194, based on an official port number assignment by IANA. OpenVPN 2.0-beta16 and earlier used 5000 as the default port.
Wed Jun 24 13:51:15 2009 LZO compression initialized
Wed Jun 24 13:51:15 2009 UDPv4 link local (bound): [undef]:1194
Wed Jun 24 13:51:15 2009 UDPv4 link remote: xxx.xxx.xx.xxx:1194
Wed Jun 24 13:52:15 2009 [UNDEF] Inactivity timeout (--ping-restart), restarting
Wed Jun 24 13:52:15 2009 SIGUSR1[soft,ping-restart] received, process restarting
Wed Jun 24 13:52:17 2009 IMPORTANT: OpenVPN's default port number is now 1194, based on an official port number assignment by IANA. OpenVPN 2.0-beta16 and earlier used 5000 as the default port.
Wed Jun 24 13:52:17 2009 Re-using SSL/TLS context
Wed Jun 24 13:52:17 2009 LZO compression initialized
Wed Jun 24 13:52:17 2009 UDPv4 link local (bound): [undef]:1194
Wed Jun 24 13:52:17 2009 UDPv4 link remote: xxx.xxx.xx.xxx

Thanks in advance

Vpn cubed is mentioned to run on rackspace/mosso cloud servers. Is the free edition available for that platform?

All the flexibility is great but I just need a quick Ubuntu 9.04, stock server. What's the quickest way to get that?

I’m trying to use the free vpn cubed, using the steps you described in your document.
I payed and I received an activation key.

Here is the complete list of things that I’ve done:

set EC2_URL=https://us-east-1.ec2.amazonaws.com
ec2-add-group vpncubed-mgr -d "vpncubed managers"
ec2-add-group vpncubed-client -d "vpncubed clients"
ec2auth vpncubed-mgr -P udp -p 1194 -o vpncubed-client -u MY_ACCOUNT_ID
ec2auth vpncubed-mgr -P tcp -p 8000 -o vpncubed-mgr -u MY_ACCOUNT_ID
ec2auth vpncubed-mgr -P udp -p 1195-1197 -o vpncubed-mgr -u MY_ACCOUNT_ID
ec2auth vpncubed-mgr –P tcp 8000 –s IP_WHERE_YOU_WILL_MANAGE_CUBE/32 not executed; I didn’t understand what IP I should enter here

set EC2_URL=https://eu-west-1.ec2.amazonaws.com
ec2-add-group vpncubed-mgr -d "vpncubed managers"
ec2-add-group vpncubed-client -d "vpncubed clients"
ec2auth vpncubed-mgr -P udp -p 1194 -o vpncubed-client -u MY_ACCOUNT_ID
ec2auth vpncubed-mgr -P tcp -p 8000 -o vpncubed-mgr -u MY_ACCOUNT_ID
ec2auth vpncubed-mgr -P udp -p 1195-1197 -o vpncubed-mgr -u MY_ACCOUNT_ID
ec2auth vpncubed-mgr –P tcp 8000 –s IP_WHERE_YOU_WILL_MANAGE_CUBE/32 not executed

set EC2_URL=

ec2run -U https://us-east-1.ec2.amazonaws.com ami-2ad53243 -n 1 -g vpncubed-mgr -d "abc1234567890" <->https://eu-west-1.ec2.amazonaws.com ami-4ee9c13a -n 1 -g vpncubed-mgr -d " abc1234567890"

Now I’m having 2 machines running:
ec2-79-125-59-0.eu-west-1.compute.amazonaws.com
ec2-67-202-7-204.compute-1.amazonaws.com

I assumed that their public IP are the following: 79.125.59.0 and 67.202.7.204; is this correct?

ec2auth -U https://us-east-1.ec2.amazonaws.com vpncubed-mgr -P udp -p 1195-1197 -s 79.125.59.0/32
ec2auth -U https://us-east-1.ec2.amazonaws.com vpncubed-mgr -P tcp -p 8000 -s 79.125.59.0/32

ec2auth -U https://eu-west-1.ec2.amazonaws.com vpncubed-mgr -P udp -p 1195-1197 -s 67.202.7.204/32
ec2auth -U https://eu-west-1.ec2.amazonaws.com vpncubed-mgr -P tcp -p 8000 -s 67.202.7.204/32

When trying to access:
https://67.202.7.204:8000 nothing happens, and I’m stucked!

I tried to ping 67.202.7.204. I receieved Request time out.
I tried to ping ec2-67.202.7.204.compute-1.amazonaws.com I receieved Request time out.

Did I do something wrong?
Where is the step where the activationkey (received from you) is required ?

system updates hang when run through elastic server manager web gui

i have vmware servers running on Pentium IIIs and a Pentium M
http://www.flickr.com/photos/osde-inf...

and i just got an error saying :
kernel requires features not on your cpu
unable to boot your vm
http://www.flickr.com/photos/osde-inf...

are your VMs intendeded to work on PIIIs, Ms and older P4s ?

It's the only free option I know of that runs cross-platform

http://www.virtualbox.org/

I'd really like to see full support for opensolaris..be even willing to help out getting it going!

It was stated within the email that "Using the cftuser you will be able to run sudo for root-level access", but I can't do any sudo or even create and save a new file via vi using cftuser account.

Are we allow to sudo get-apt packages, e.g. phpmyadmin? Can please help me out, thanks.

Too much of a security risk!!! How do you expect people to give you full access to their EC2s? There is no technical reason for this (other than you can automatically check the installation - which is a nice feature but the risks far outweigh this)?

I have successfully established a VPN-Cubed environment between a physical datacenter, an EC2 elastic server (Ubuntu 8.10 built from elasticserver.com), and a remote client.

However, I am unable to connect to any of the services on the EC2 elastic server from the other VPN-Cubed clients.

External access to the EC2 server (ports 22, 8080, etc.) works fine.
For testing I turned off the firewall on the EC2 elastic server and verified that the Amazon security group is allowing port 22 for SSH.
On any of the other VPN-Cubed clients I can not access port 22 (or any other port) on the EC2 Elastic server, yet I can create an SSH session FROM the EC2 server to the clients.
The other VPN-Cubed clients have full access to each other.
I have tried numerous changes to the Amazon security groups and nothing seems to work.
I have also noticed that only one machine can be connected to the VPN-Cubed manager per physical ip address. In my case this is the physical address of a firewall with a number of machines behind it in a datacenter. I would like to connect multiple machines behind the firewall as VPN-Cubed clients.

Any suggestions would be greatly appreciated,

Joel

No surprisingly, it seems like the main focus of elastic servers at this point it dedicated, long-running servers that have a single narrow focus. Basically, someone wants to run a certain Django/RonR app or an AMQP server in virtualized environment. In this usage case, you typically want to deploy a few long running production servers.

In scientific computing, I want to deploy a huge number (32, 64, 128, 256, ...) of servers, for a relatively short period of time (maybe an hour, or a day). During this time, I would use the servers all together to solve some problem.

The current pricing scheme/business model for ES doesn't seem setup to allow this type of usage. Have you thought about parallel computing/HPC focused pricing models. I have access to a number of different clusters that have VM support that I could deploy to, but if I have to spend lots of money to start 100's of ES based VMs, it would be a no go.

Sure some folks in scientific computing would want to deploy a cluster 24/7, but the big attraction of cloud computing (such as EC2) is that I can get a massive number of CPUs for just the time I actually need them. It would be nice if CFT had a pricing model that addressed this usage case. I am more than willing to discuss more of the requirements that we have.

Guys, I want to create an OpenLDAP server on EC2, any bundles out there that would support this?