Brooklyn Data Directory setting not working as expected

[Dave Finnegan]

I've set my brooklyn.datadir in my brooklyn.properties file in order to persist state to my home directory rather than /tmp/brooklyn-${user}.

I was expecting this to change the data directory for persisted state, but it has changed the data directory for new instance creation on, say, EC2.  I then get an error that it can not create my home directory on the new node.  Which, of course, is correct.

So, how do I persist state from my brooklyn server to a stable directory, but not effect all VMs that I create?

It seems like there should be a 'brooklyn.datadir.server=<some dir>' and let the 'brooklyn.datadir=<some dir>' be directed at the remote instances that get created.

[Aled Sage]

Hi Dave,

This is a very timely question - we merged a fix for this just a few days ago [1]!

With the latest brooklyn code, you can set:

• brooklyn.base.dir for "Directory for reading and writing all brooklyn server data"
  This now defaults to ~/.brooklyn/
• onbox.base.dir for "Default base directory on target machines where Brooklyn config data is stored; default depends on the location, either ~/brooklyn-managed-processes or /tmp/brooklyn-${username} on localhost"
  

Setting just one of these will not affect the other, so for your use-case setting "brooklyn.base.dir" will do the trick.

Aled

[1] https://github.com/brooklyncentral/brooklyn/pull/1271

--
You received this message because you are subscribed to the Google Groups "brooklyn-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to brooklyn-dev...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Alex Heneveld]

And if you turn persistence on ("--persist auto" on the CLI) it will store data for deployments in ~/.brooklyn/brooklyn-persisted-state.  Once we iron out a couple of issues (see issue #315 [1]) this will be the default.

Best
Alex

[1] https://github.com/brooklyncentral/brooklyn/issues/315