Have you check this link:
https://xcat-docs.readthedocs.io/en/stable/guides/admin-guides/manage_clusters/common/deployment/prepostscripts/post_script.html
after "rinstall" command , you can run command "xcatprobe osdeploy -n
<nodename>" to probe operating system provision process.
The output will show from begin of provision process until end of node
installation completion. set attribute "xcatdebugmode=1" on the site
table will show more of debug information.
On the CN, check the /var/log/xcat/xcat.log .
when you deploy the VM template, How did u change the network setting? I
think /opt/xcat/xcatinfo just capture some information for the CN. Don't
think will use them during the provision.
Thanks,
Casandra Qiu
...................................................................
Casandra Hong Qiu
Phone: (845) 433-9291, t/l 293-9291
Office: Building 8, 3-B-04
cxh...@us.ibm.com
From: Thomas HUMMEL <thomas.hum...@pasteur.fr>
To: <xcat-user@lists.sourceforge.net>
Date: 11/02/2020 02:56 PM
Subject: [EXTERNAL] [xcat-user] Postscripts : generation/download
workflow
Hello,
I'm using xCAT 2.16 on CentOS 8.2 x86_64.
I mostly uses stateless nodes but also have a few stateful ones. I
installed the first batch of the latter using the standard method (which
basically comes to an xCAT wrapped kickstart install) and everything
works fine.
Then, for some reason, I had to provide users a VMWare template of a
stateful node so they could just right click "New VM from this template"
to create a new node (some ansible runs afterwards).
In order to to this, I just xCAT installed a new stateful node on a new
VM, then after minor cleanup (ssh keys, history, ...) convert it into a
template.
For this second case I noticed that for VM's created after this template
the /opt/xcat/xcatinfo file would come with info from the initial VM
(the one which was used to create the template). Obviously I did not
cleaned enough for that matter.
I know about postscripts and I've read the doc (I'm using them via
xcatpostinit1.service in the stateless case and through updatenode -P in
the stateful case) but it is still not quite clear to me what is
generated and what is downloaded or already present on the node.
My understanding is that :
it is ultmiately the /xcatpost/mypostscript which wraps and runs the
configured postscripts.
It is dynamically generated from a (standard or customised )
mypostscript.tmpl template located on the xCAT server
Note: I'm not using precreatemypostscripts
Client side, my understanding is that
/opt/xcat/xcatinfo is generated and that, depending on if this is the
first REBOOT after install or not:
- either /opt/xcat/xcatdsklpost
- or /opt/xcat/xcatinstallpost
gets run. The latter itself patches and runs
- /xcatpost/mypostscript.post
Finally, it seems that the mypostscript is could be downloaded from the
server.
Could you quicly explain/clarify the relation between all those objets ?
- what is generated and is not (and thus can/must be remove in a general
purpose template) ?
-> server side
-> node side
- what may exist or not on node's side depending on contexts
- what is downloaded ?
- in what general circumstances ? For instance I noticed
/xcatpost/mypostscript.post presence seems to depend on the REBOOT context.
My goal, as you understand is to know what to remove in the template so
the VM's instanciated after it would have the correct info (NODE=, ...)
into the proper files.
Thanks for your help
--
Thomas HUMMEL
_______________________________________________
xCAT-user mailing list
xCAT-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/xcat-user
_______________________________________________
xCAT-user mailing list
xCAT-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/xcat-user
