forked from CSCfi/pouta-virtualcluster
-
Notifications
You must be signed in to change notification settings - Fork 0
/
TODO
28 lines (27 loc) · 1.07 KB
/
TODO
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
# TODO: resize - add or remove nodes, check first that all are alive
# TODO: boot from volume (as default?) and static IPs
# TODO: parallel provisioning
# TODO: dynamic public IP allocation
# TODO: (ansible) automatic Spark startup
# TODO: (ansible) disable IPv6 (?)
# DONE:
# add proper info command
# handle multiple networks in tenant
# print urls for web interfaces after provisioning
# convert update and reboot to a play
# run ansible provisioning automatically
# (ansible) automatic Hadoop startup
# configurable admin account instead of 'centos'
# REJECT: add users from cluster.yml(?)
# rate-limit reset_nodes
# (ansible) add gridengine
# (ansible) separate common cluster setup from hadoop
# (ansible) add ganglia
# add reset command
# REJECT: fix playbooks to support generic frontend / node notation for hosts
# support groups/roles in config and generate the ansible inventory dynamically
# (ansible) format shared_data and local_data in playbook if necessary
# write log about provisioning actions
# delete a single volume at a time
# clean shutdown
# write ansible inventory