~andreserl/maas:chrony

Last commit made on 2018-02-26
Get this branch:
git clone -b chrony https://git.launchpad.net/~andreserl/maas
Only Andres Rodriguez can upload to this branch. If you are Andres Rodriguez please log in for upload directions.

Branch merges

Branch information

Name:
chrony
Repository:
lp:~andreserl/maas

Recent commits

97c7783... by Andres Rodriguez on 2018-02-26

Core changes for chrony

c38a589... by Newell Jensen on 2018-02-25

LP: #1706438 -- Add zone to Pods and only allocate machines from Pod if zones match.

* Adds ForeignKey to Zone on the BMC model.
* Updates machines allocate API endpoint to only use machines from Pod's that have matching zone if zone parameter is supplied.

28059ad... by Anthony Dillon on 2018-02-24

LP: #1750020 - Create the script reveal functionalty

f8b36e0... by Lee Trager on 2018-02-24

Modify rescue mode to run commissioning or test scripts if available.

The script runner is now executed when transistioning into rescue
mode. Any commissioning or test script available will be run. Because
rescue mode allows the user to SSH into the ephemeral environment MAAS
does not check for loss of heart beat or remote checking if the script
has run past its runtime.

This does not implement the necessary API or UI changes to allow
users to select scripts to run in rescue mode.

a1a52f4... by Anthony Dillon on 2018-02-24

General UI improvements for node details

198016f... by Lee Trager on 2018-02-24

Allow scripts to be selected using the for_hardware field using system or mainboard fields.

cloud-init is now sending a FAIL when a script reboots. This is ignored if
MAAS knows a script is running which is known to reboot.

0000d66... by Anthony Dillon on 2018-02-24

LP: #1749871 - Fix the layout of the first user journey

15a4c14... by Lee Trager on 2018-02-24

Only cache processed ScriptResults for the current nodes being dehydrated.

Previously MAAS would cache the current ScriptResults for all nodes
to minimize database lookups. This caused all ScriptResults to be
loaded into RAM regardless of how many nodes are actually being
loaded. Now only the latest ScriptResults from the current set of
nodes being dehydrates are loaded. As nodes are dehydrated their
ScriptResults are cleared from cache to reduce memory usage on the
region.

1223c03... by Andres Rodriguez on 2018-02-24

LP: #1711203 - Use grub instead of shim to workaround issues with Secure Boot deployment

23142d8... by Anthony Dillon on 2018-02-24

LP: #1750015 - Reinstate the logic around node logs