openstack
/
diskimage-builder
Code Issues Proposed changes
diskimage-builder/elements/hwdiscovery
History
Ben Nemec 16be6d7ce0 set -u and -o pipefail everywhere 
As with the previous similar changes, this is intended to catch
problems as they happen instead of ignoring them and continuing on
to potentially fail later.  Setting this on all existing scripts
will allow us to enforce use via Jenkins.

Change-Id: Iad2d490c86dceab148ea9ab08f457c49a5d5352e
 		2014-05-06 15:51:07 -05:00
..
binary-deps.d Build ramdisks in an image chroot. 2013-07-12 11:09:35 +01:00
init.d Make pxe_mac accurate in two common cases. 2013-11-17 21:43:59 -06:00
install.d set -u and -o pipefail everywhere 2014-05-06 15:51:07 -05:00
README.md Further fleshing out of hwdiscovery element 2012-12-06 21:55:30 +00:00

README.md

A ramdisk to report the hardware of a machine to an inventory service.

This will collect up some basic information about the hardware it boots on:

  • CPU cores
  • RAM
  • Disk
  • NIC mac address

This information will then be collated into a JSON document, base64 encoded and passed, via HTTP POST, to a URL that you must specify on the kernel commandline, thus:

HW_DISCOVERY_URL=http://1.2.3.4:56/hw_script.asp

This is currently fairly fragile as there can be a huge variability in the number of disks/NICs in servers and how they are configured.

If other elements wish to inject data into the hardware discovery data, they can - they should be specified before hwdiscovery to the image building script, and they should contain something like this in their init fragment:

_vendor_hwdiscovery_data="$_vendor_hwdiscovery_data "some vendor key" : "some data you care about", "some other vendor key" : "some other data you care about","

Note that you are essentially feeding JSON into the main hwdiscovery JSON.

This will allow any number of vendor specific hwdiscovery elements to chain together their JSON fragments and maintain consistency.