Offline installation support

[dpkkumar01]

We have tested the ST work flow Engine in our local lab which is having internet connectivity. Now we would like to install ST in our other server which doesn't have connectivity with internet and my other server.

Kindly provide your valuable suggestion to complete my offline installation.

[arm4b]

You can rely on Docker installation for the StackStorm community. All you need is to download and then copy Docker images to your offline system: https://docs.stackstorm.com/install/docker.html

Apart of that @LindsayHill has in plans to write a more detailed technical doc how to install StackStorm in offline systems.

And finally, we're actively working on the solution to bundle everything in one virtual appliance. That will be available for StackStorm Enterprise (BWC) customers.

[LindsayHill]

@dpkkumar01 does your other server have proxy access, or does it have access to an internal yum/apt mirror, that in turn has Internet access? Or is it completely air-gapped?

If you have proxy access, that is straightforward (see the notes here: https://docs.stackstorm.com/install/)

If you don't have proxy access, but do have an internal APT mirror, then you have options like this https://stackstorm.com/2017/02/10/installing-stackstorm-offline-systems/

If you are completely airgapped, then you will need to manually download the RPMs/Debs, and transfer them across using your normal procedures for handling this sort of thing. You will not be able to use the scripted install, but will have to follow a version of the manual install instructions https://docs.stackstorm.com/install/deb.html

[dpkkumar01]

Thanks for the response.

@LindsayHill we don't have the proxy access on that system. We have to move the complete package to my new machine and start the installation.

[LindsayHill]

@dpkkumar01 as you know, ST2 packages pull in other dependencies, e.g. python, PostgreSQL, etc. So while it's possible download the ST2 rpms/debs, then transfer them and install using dpkg/rpm commands. But then you'll run into missing dependencies.

How does your site handle this sort of thing normally? Do you run internally accessible mirrors of RHEL/Ubuntu base repos? Or if you need to install something, do you download every possible dependency (and their dependencies), and then transfer every possible package that might be required on top of a base RHEL/Ubuntu install?

[dpkkumar01]

@Lindsay Hill, I understood your point. usually we will download the required package to our local machine and transfer to my other server which is not having internet access.

There won't be any direct connection between these machines. we will transfer the package via some VPN channel.

On Sat, Nov 11, 2017 at 1:11 AM, Lindsay Hill notifications@github.com wrote:

@dpkkumar01 https://github.com/dpkkumar01 as you know, ST2 packages pull in other dependencies, e.g. python, PostgreSQL, etc. So while it's possible download the ST2 rpms/debs, then transfer them and install using dpkg/rpm commands. But then you'll run into missing dependencies.

How does your site handle this sort of thing normally? Do you run internally accessible mirrors of RHEL/Ubuntu base repos? Or if you need to install something, do you download every possible dependency (and their dependencies), and then transfer every possible package that might be required on top of a base RHEL/Ubuntu install?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/StackStorm/st2/issues/3841#issuecomment-343567083, or mute the thread https://github.com/notifications/unsubscribe-auth/AYV7xGzYQRDk63Xt5VoLPoLqLcDf7Padks5s1KbMgaJpZM4QX-cd .

[LindsayHill]

So what is installed on your system that has no Internet access? Is it just a minimal RHEL/Ubuntu install, and you need to transfer every single dependency?

[LindsayHill]

The reason I ask is because the instructions are somewhat different if you need to download & transfer every possible lower-level dependency.