Hi Mark,
Good to see you on-board
All the latest docs are located on the new documentation site
readthedocs [1], but you may want to start on sourceforge [2], but note
that a lot of the ducmentation currently is being migrated from
sourceforge to readthedocs.
To answer your questions, see below on my take on it, others may have
more and different answers
That said I am looking for information on how to do the following
thing. I have extensively Googled and haven't found answers for these
as of yet, even links would be greatly appreciated!
Deploy diskless nodes over IB
- I saw this as a priority three on the 2.10 roadmap but haven't see
much of anything about it. If anyone has this working I don't mind
doing the legwork. I have managed to get somewhere on this but end up
with an error: "dracut warning: No root device "1" found" and haven't
managed to get past that. This is the most important piece to be honest.
I have also seen this in the roadmap as well, but actually seen it in
action, I can't see it in the release notes [3], maybe something I am
missing. I would love to be able to do this myself
Configure non-node entities via DHCP
- A good for instance would be setting up IPMI or switches via DHCP
without knowing the mac. We have quite a bit of volume and automatic
setup of these items would greatly speed up our process.
On this note, the key word is discovery for search criteria in the
readthedocs. Depending on the HW you have, and the switches you have,
you can discover in ~3 methods
1. Automagic through switch discovery, This is detailed in [4], [5] and [6]
2. Sequential Discovery -- resources include [7], [8] and [9]
3. Flex or bladecenter node discovery -- resource include [10]
Hopefully, this will give you a lot of reading material to go through
the paces. I know the documentation is in quite a few places, but that
is due to things getting migrated to readthedocs.
If you have any issues, give us a shout on IRC, some of us hang out on
the #xcat channel
[1] http://xcat-docs.readthedocs.org/en/latest/
[2] http://sourceforge.net/p/xcat/wiki/Main_Page/
[3] https://github.com/xcat2/xcat-core/wiki/XCAT_2.10_Release_Notes
[4]
http://sourceforge.net/p/xcat/wiki/XCAT_NeXtScale_Clusters/#node-definition-and-discovery
[5]
http://sourceforge.net/p/xcat/wiki/XCAT_iDataPlex_Cluster_Quick_Start/#option-2-switch-discovery
[6]
http://xcat-docs.readthedocs.org/en/latest/guides/admin-guides/manage_clusters/ppc64le/discovery/switch_discovery.html
[7] http://sourceforge.net/p/xcat/wiki/Sequential_Discovery/
[8]
http://sourceforge.net/p/xcat/wiki/XCAT_iDataPlex_Cluster_Quick_Start/#option-1-sequential-discovery
[9]
http://xcat-docs.readthedocs.org/en/latest/guides/admin-guides/manage_clusters/ppc64le/discovery/seq_discovery.html
[10]
http://sourceforge.net/p/xcat/wiki/XCAT_system_x_support_for_IBM_Flex/#cmm-discovery-and-configuration
regards,
Arif
------------------------------------------------------------------------------
Presto, an open source distributed SQL query engine for big data, initially
developed by Facebook, enables you to easily query your data on Hadoop in a
more interactive manner. Teradata is also now providing full enterprise
support for Presto. Download a free open source copy now.
http://pubads.g.doubleclick.net/gampad/clk?id=250295911&iu=/4140
_______________________________________________
xCAT-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/xcat-user
