This is no longer a toy project, but part of a bigger logging/monitoring setup you might find useful. So please refer to the 'surrunding' project https://github.com/chrisbetz/logmon.

I'm currently updating the READMEs, so until I'm done I'll keep the old readme here for reference.

So far, this is a toy project to try out to run Riemann in an OSv unikernel. You might also want to take a look into https://github.com/tzach/riemann/ and the associated blog post http://osv.io/blog/blog/2014/04/22/riemann-on-osv/, but stuff changed inside Riemann, making it a little harder.

On OS X, I had some trouble first building new OSv images. I finally figured out it had to do with my firewall / security setting for qemu. So, if you also encounter problems, try this first:

Make sure you have cleaned old artefacts if you encountered problems.

capstan delete riemann.osv

and delete the related files/folders under ~/.capstan/repository/riemann.osv.

Install qemu using homebrew:

brew install qemu

Find out where your qemu is located:

which qemu-system-x86_64

For me it was /usr/local/bin/qemu-system-x86_64.

Run ``/usr/local/bin/qemu-system-x86_64` and make sure to open your firewall by accepting incoming network requests to qemu.

And make sure to export CAPSTAN_QEMU_PATH (track this issue):

export CAPSTAN_QEMU_PATH=/usr/local/bin/qemu-system-x86_64

Now you should be good to go.

I chose to copy config to ./riemann.config and alter that file to bind riemann to 0.0.0.0 instead of 127.0.0.1 to be able to access Riemann without port forwarding.

Run the image on VirtualBox using this command (which will also build it):

capstan run -p vbox -f 5555:5555 -f 5556:5556 -f 8001:8000

It's essential to forward the ports, as otherwise I cannot access my guest machine. Bridged network is somewhat broken for me here :( You can find the Dashboard/REST API on http://localhost:8001/

You might need to clear the artefacts if you change things.

Use

capstan delete riemann.osv

to delete the instance.

In order to get Capstan to rebuild correctly, you might need to clear the image. I think there's no capstan command for this right now, although you might check your images using capstan I. You just need to move ~/.capstan/respository/riemann.osv to the trash.

Once installed correctly you can run riemann-health to send data to your Riemann setup.

I added a prn to the config in order to push events down to the console for debugging reasons.

Currently, OSv suffers from using the same UUID for its disk images. Thus, you cannot run two (different) instances of OSv unikernels in parallel, because vbox will prevent you from registering a different disk image under the same UUDI.

You can circumvent this by changing the UUID using an internal command on the repository diskimage (before running the instance, obviously)

VBoxManage internalcommands sethduuid ~/.capstan/repository/riemann.osv/riemann.osv.vbox