Revision history [back]

Can i write directly to openstack db without plugins?

Hi,

first of all: I can get some concepts wrong, so feel free top point this out.

What i understood from heat source code (taken from repo on github) is that when i run heat stack update, it sends data to some kind of db. What i assume, is that this is a kind of centralized db, where all stuff related to running openstack are being kept. Most important for me is userdata stuff.

Help me see this trough: this works similarly when i run server create, right? user data is being sent to cloud db and then instance on wakeup pulls it via metadata service, right?

So. Is there a way to start multiple instances using same entry into db without using userdata upload and stacks? I am preparing piece of code in python that scales faster than heat and provides more customizeable algorithms of scaling and i see some extra time to save in operations like uploading stuff to cloud.

Thanks in advance

Can i write directly to openstack db without plugins?

Hi,

first of all: I can get some concepts wrong, so feel free top point this out.

What i understood from heat source code (taken from repo on github) is that when i run heat stack update, it sends data to some kind of db. What i assume, is that this is a kind of centralized db, where all stuff related to running openstack are being kept. Most important for me is userdata stuff.

Help me see this trough: this works similarly when i run server create, right? user data is being sent to cloud db and then instance on wakeup pulls it via metadata service, right?

So. Is there a way to start multiple instances using same entry into db without using userdata upload and stacks? I am preparing piece of code in python that scales faster than heat and provides more customizeable algorithms of scaling and i see some extra time to save in operations like uploading stuff to cloud.

Thanks in advance