Underscore characters are not valid in DNS entries unless you want to be
pedantic and read through the RFCs involved in DNS. Use dash characters
instead and youll be fine.
https://stackoverflow.com/questions/2180465/can-domain-name-subdomains-have-
an-underscore-in-it
From: Robinson,
First, I think confignics has been superseded by confignetwork; Regardless of
that, you should only call one of them.
Second, are you using the ‘nics’ table? If not, please look into it. If yes,
please show the settings for the node.
From: aaronhc...@gmail.com
Sent: Wednesday, May
The postscript is 'hardeths'
> On August 20, 2019 at 3:54 PM Kevin Keane wrote:
>
> Remember that the DHCP in xCAT is quite substantially different from
> "normal" enterprise DHCP, since it needs to only serve a very specific,
> well-defined, set of computers (all of which normally will
I have a hierarchical cluster setup running xCAT v2.12.2.
I do not wish to run the DHCP server on the service node (SN) - the master node
(MN) will serve the DHCP leases.
When I run 'nodeset node01 osimage' on the SN I get the following message:
[root@c2-service]# nodeset node01 osimage
node01:
Please ensure that "awk" is installed and located at the location called for by
the first line of those scripts.
Check the package list that you're using in your OS image and explicitly add
"gawk.x86_64".
Sometimes dependencies change and things that used to get implicitly installed
are no
Did you properly set up the deps repo for your OS? You have to run the setup script from within the OS-specific directory.On Jul 6, 2018 10:35 AM, Javier Ron wrote:
Hello,
I am trying to install xcat 2.12 for a dev environment & after a yum install xCAT, I get conflicts
-->
You're exactly right - the "otherpkgs" repo is for packages that are not part
of the official OS distribution.
Note that if you use "osimage" then you can specify a comma-separated list of
package directories for both pkgdir and otherpkgsdir. You will need to run
"createrepo" in each of the
Apparently it's normal to have "bootparams" empty now. The values of kcmdline that appear in the tftp file get built from other sources when nodeset is run.My problem with dracut was related to IP addressing and IPoIB.Lots of options need to be set for the addkcmdline value in the osimage to get
There is no need to keep the xCAT master node around once the machines have been imaged.But it sure makes life easier if you do.--
Check out the vibrant tech community on one of the world's most
engaging tech sites,
The otherpkgs script is run as a postbootscript and is part of the "xcatdefaults" entry.I'm not surprised that a free OS like CentOS acts this way.I guess I'm a little surprised that it hasn't been discussed before. I would expect Ubuntu to be similar.On Oct 5, 2017 11:37 AM, Kevin Keane
According to the documentation on hierarchical clusters, one can specify
primary and backup service nodes for a given compute node.
When I do that, and then run nodeset for the compute node, it appears that
both of the service nodes execute the prescripts for the node.
That's undesirable, and the
You have to set the root password in the passwd table. I recommend setting it for "compute" .On May 10, 2017 1:57 PM, Fırat Yılmaz wrote:
Hi Mark,
Yes i set the password and checked with tabdump passwd | grep
ipmi command.
I
I just noticed that there are two world-readable pem files in /xcatpost after a diskfull image of a node. Shouldn't those files be restricted or deleted?--
Check out the vibrant tech community on one of the world's most
Check on the node that the script is executable. I recently saw that when the scripts get copied to the node that all execute bits get removed.I would also recommend that you use the new offer ofed installation script. The one with 2 on the end of the name. Look in the /opt/xcat tree for the
Does this pdu support only apply to IBM hardware?--
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot___
You can use "prescripts" to do stuff on the MGMTNODE.
Create /install/prescripts and put your code in there.
Then set the "prescripts" in the prescripts table (or use osimage).
They can be run either at the beginning or end of the action.
See the man page for prescript table.
xCAT will pass the
ZhongGuanCun Software Park,No.8 Dong Bei Wang West Road,
Haidian District,
> Beijing P.R.China 100193
>
> IBM环宇大厦
> 北京市海淀区东北旺西路8号,中关村软件园28号楼
> 邮编:100193
>
> > > Russell Auld ---12/21/2016 08:15:19 AM---Do people still use software
kits? --
You definitely need to use a RH7 compatible script. Look in /opt/xcat/share for sample filesOn Jan 23, 2017 11:25 AM, Damir Krstic wrote:I am trying to install RH7.1 (stateful) and the install keeps failing at various points of my kickstart file. I think I may be using RH6
Upgrading is pretty painless. Unfortunately, there's no way to really
know if there have been any code changes that will affect you.
I got burned by a couple things when I upgraded from 2.9.x to 2.12.x.
1 - When I added some new nodes, and then ran makedhcp, the new DHCP
stanzas were
http://xcat-docs.readthedocs.io/en/latest/guides/admin-guides/reference
s/man5/cfgmgt.5.html
How are the "roles" actually applied to a node by using this table when
the cfgmgr="chef"?
--
Check out the vibrant tech
I have a situation where postbootscripts do not run. I traced the problem back
to "/opt/xcat/xcatinstallpost" script which is called
by the "/etc/init.d/xcatpostinit1" service after the first boot.
The trouble is with this section of code:
MACADDR=`grep MACADDRESS= /xcatpost/mypostscript.post
I have a situation where postbootscripts do not run. I traced the problem back
to "/opt/xcat/xcatinstallpost" script which is called
by the "/etc/init.d/xcatpostinit1" service after the first boot.
The trouble is with this section of code:
MACADDR=`grep MACADDRESS= /xcatpost/mypostscript.post
I can't reproduce this now. There must have been something wrong with the pool or the order of operations.
I recreated the local pool and can use dir:/// now.
It's working as expected.
On Oct 24, 2016 3:42 PM, Russell Auld <russa...@comcast.net> wrote:Yes, I could see the volume in the p
"
>
> You can also run "virsh vol-info " to check the size.
>
> Mark Gurevich
> Poughkeepsie Development Lab
> HPC Software Development - xCAT
>
> "If we knew what it was we were doing, it would not be called research, would it?"
> --Al
eijing P.R.China 100193
- Original message -
From: "Russell Auld" <russa...@comcast.net <mailto:russa...@comcast.net> >
To: "'xCAT Users Mailing list'" <xcat-user@lists.sourceforge.net
<mailto:xcat-user@lists.sourceforge.net> >
Cc:
Subject: Re
China System Technology Laboratory
Tel: 86-10-82453455
Email: w...@cn.ibm.com <mailto:w...@cn.ibm.com>
Address: 28,ZhongGuanCun Software Park,No.8 Dong Bei Wang West Road, Haidian
District Beijing P.R.China 100193
- Original message -
From: "Russell Auld" <r
t;
Address: 28,ZhongGuanCun Software Park,No.8 Dong Bei Wang West Road, Haidian
District Beijing P.R.China 100193
- Original message -
From: "Russell Auld" <russa...@comcast.net <mailto:russa...@comcast.net> >
To: "'xCAT Users Mailing list'" <xcat
e of it. The problem is
that the device specification in the file is also specified as "bootnic".
Of course, there is no such device on the system, which means that after the
first reboot, there is no network connection and the postbootscripts do not run.
-----Original Message-
Fr
or the "doxcat" script reconfigures the
IP address using DHCP a second time, just before it
does the SSL certificate generation and the handshaking.
From: Russell Auld [mailto:russa...@comcast.net]
Sent: Saturday, August 6, 2016 8:16 AM
To: 'xCAT Users Mailing list' <xcat-user@lists.so
: Russell Auld [mailto:russa...@comcast.net]
Sent: Thursday, May 19, 2016 11:09 PM
To: xCAT Users Mailing list <xcat-user@lists.sourceforge.net>
Subject: Re: [xcat-user] Prescripts running twice
No, the script only exists in the prescripts directory.
On May 19, 2016 10:42 PM, Xiao Peng W
You could remove/unset "managedaddressmode" and use a postscript to change the
nic config from dhcp to static. We used to use a combination of "confignics"
and "hardeths".
You would need to be running dhcpd on the master node for that to work.
-Original Message-
From: Dr. Markus
he nodeset command?
On May 19, 2016 10:30 AM, Russell Auld <russa...@comcast.net> wrote:
I noticed that my prescripts are being executed twice when I do a “nodeset osimage”
I have the scripts set as “prescripts-end” in the prescripts table.
What’s the meaning of
I noticed that my prescripts are being executed twice when I do a "nodeset
osimage"
I have the scripts set as "prescripts-end" in the prescripts table.
What's the meaning of "begin" and "end" with respect to "osimage"?
Why is my script running twice?
The first execution occurs when I run
Looks like a pre script will run as root on the node's master
https://sourceforge.net/p/xcat/wiki/Postscripts_and_Prescripts/
--
Mobile security can be enabling, not merely restricting. Employees who
bring their own
What's the right way to get the sync files to the service nodes?
I'm running this cron job every night on the MN, and the CN target in the
command is one that is serviced by the SN.
/opt/xcat/bin/updatenode -f
Is there a better way to do this?
The problematic NICs were on the nodes, not on the xCAT master in my case. I just got replacement NICs.
--
___
xCAT-user mailing list
xCAT-user@lists.sourceforge.net
Could this approach be used to get vendor-supplied network drivers into the PXE
sequence?
Or the RH Anaconda installation kernel?
From: Russell Jones [mailto:russell-l...@jonesmail.me]
Sent: Monday, January 11, 2016 6:33 PM
To: xCAT Users Mailing list
e original eyes (mine) don't know exactly what
> to look for.
>
> Best regards,
>
> Steve
>
> On Wed, Nov 4, 2015 at 8:34 AM, Russell Auld <russa...@comcast.net> wrote:
>>
>> You will definitely need to change "master" to the IP address that the nodes
ion about what MAC
addresses are mapped to the IP addresses? I think that ultimately the
information comes from the mac and node and hosts tables but I mean what files
are the DHCP using to pick up that information. I don't see it in the usual
place.
Thanks a lot for your help Russell.
Steve
On T
Check the site table.
On Nov 3, 2015 1:14 PM, Stephen Cousins wrote:
>
> Hi,
>
> I had improperly set my network to 172.0.0.0/16 thinking that that was in
> private space. I later found that 172.16.0.0/12 is the real private space so
> I changed entries in all of the
ot;,,
"SNsyncfiledir","/var/xcat/syncfiles",,
"nodesyncfiledir","/var/xcat/node/syncfiles",,
"tftpdir","/tftpboot",,
"xcatdport","3001",,
"xcatiport","3002",,
"xcatconfdir","/etc/xcat"
Looks like xcatdsklspost does get called even for stateful installs. Look at
the script header.
Usually in cases like this, the issue is that the node being imaged can't
resolve the name of the master node. Make sure your dns is working properly.
On Oct 23, 2015 12:52 PM, "Heald, Nathan T."
Did you try “dhcpinterfaces” in the site table?
See my thread from this past July 27 entitled:
“Is there a way to add outside subnets to dhcpd.conf?”
From: John Keenleyside [mailto:keen...@ca.ibm.com]
Sent: Sunday, October 4, 2015 9:48 PM
To: xcat-user@lists.sourceforge.net
Subject:
Define "work with".
Do you want to do node discovery?
On Sep 1, 2015 1:08 PM, "Waldron, Michael H" wrote:
>
> Does anyone know if xCAT will work with Dell PowerEdge M1000e blade chassis?
> I see references to various Dell servers but none relating to blades.
>
> Mike
if this is required or not, and I am not able to login to a
system right now to verify, however try adding !remote! to
site.dhcpinterfaces and see if your remote VLAN shows up when you do a
makedhcp -n
On 7/25/2015 2:11 PM, Russell Auld wrote:
I added definitions to the networks table
I added definitions to the networks table that identify separate VLANs in our
environment.
When I run makedhcp -n' it appears that only the subnets that the MN's
interface are attached to get created.
The MN is a on routable VLAN 10.10.10.0/24. The MN is also connected to another
local,
I think you need to make some symbolic links in
/opt/xcat/share/xcat/netboot/centos pointing to the dracut directories in
/opt/xcat/share/xcat/netboot/rh
From: yang feng电话 [mailto:hoking.y...@gmail.com]
Sent: Monday, July 13, 2015 4:03 AM
To: xcat-user@lists.sourceforge.net
Subject:
,ZhongGuanCun Software Park,No.8 Dong Bei Wang West Road, Haidian
District Beijing P.R.China 100193
Russell Auld ---2015/04/11 05:15:29---At the end of the script I see this
block: if [ $return_value -eq 0 ]; then
From: Russell Auld russa...@comcast.net mailto:russa...@comcast.net
To: 'xCAT Users Mailing
At the end of the script I see this block:
if [ $return_value -eq 0 ]; then
updateflag.awk $MASTER 3002 installstatus booted
else
updateflag.awk $MASTER 3002 installstatus failed
fi
I always wondered why the status would be status=failed when I ran an 'lsdef'
on a node.
I think the status
to my surprise.
From: Russell Auld [mailto:russa...@comcast.net]
Sent: Friday, March 27, 2015 6:48 AM
To: 'xCAT Users Mailing list'
Subject: [xcat-user] Function keys under rcons and wcons
Is there a way to send a function key combo to the system while using rcons?
For example, when Anaconda
I upgraded from v2.8.4 to v2.8.5 on a RHEL5.5 machine - see [1] and [2] below
for version info.
When I tried to restart the xcatd daemon, I got the following error:
[root@head]# service xcatd restart
Restarting xCATd Unknown warnings category 'Sys::Syslog' at
in version 2.8.5 with
regard to ‘configib’.
I just wanted to close the loop on this issue and have this info out there for
future reference.
From: Russell Auld [mailto:russa...@comcast.net]
Sent: Thursday, November 20, 2014 1:14 PM
To: 'xCAT Users Mailing list'
Subject: Re: [xcat-user] using
Ensure that the node being deployed can fully resolve the name of the
management node and itself. In an earlier email, you showed
the result of resolving both of the hosts from the management node. The compute
node being provisioned needs to be able to resolve
names too.
If name resolution
Einstein
From:Russell Auld russa...@comcast.net mailto:russa...@comcast.net
To:'xCAT Users Mailing list' xcat-user@lists.sourceforge.net
mailto:xcat-user@lists.sourceforge.net
Date:11/18/2014 04:57 PM
Subject:Re: [xcat-user] using gateway=xcatmaster
I know this is a bit late, but is there any xCAT activity at Super Computing
this year?
--
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and
External: mailto:ling...@us.ibm.com ling...@us.ibm.com, 845-433-5692
I never worry about the future. It comes soon enough. --- Albert Einstein
From:Russell Auld mailto:russa...@comcast.net
russa...@comcast.net
To:'xCAT Users Mailing list'
mailto:xcat-user
I have xcatmaster set as the value for gateway
#netname,net,mask,mgtifname,gateway,dhcpserver,tftpserver,nameservers,ntpservers,logservers,dynamicrange,staticrange,staticrangeincr
ement,nodehostname,ddnsdomain,vlanid,domain,comments,disable
Using xCAT 2.8.4 with RHEL 6.5
Trying to get xCAT to setup the Infiniband card and having trouble.
I set up my node definitions as I did in xCAT 2.7.5 (on a different cluster),
namely like this:
lsdef node1:
.
installnic=eth0
: Russell Auld [mailto:russa...@comcast.net]
Sent: Monday, November 10, 2014 12:31 PM
To: 'xCAT Users Mailing list'
Subject: [xcat-user] confignics/configib not configuring the Infiniband card
Using xCAT 2.8.4 with RHEL 6.5
Trying to get xCAT to setup the Infiniband card and having trouble.
I
I would argue that the Infiniband (IB) fabric is unrelated to xCAT.
xCAT handles the provisioning of the nodes. It will install the operating
system and any IB drivers.
However, the driver stack on any given node is completely irrelevant to:
1) The IB fabric topology.
2) The IB
If you continue with the configuration of xCAT, you'll get to the part where
you have to define a template for the provisioning.
The default template should have something like this in it:
#
# SE Linux
#
selinux --disabled
That template ultimately gets turned into an anaconda script
that part, but what I'm after is the lower level stuff like "kcmdline" and the AnacondFrom: Russell Auld russa...@comcast.netTo: xCAT Users Mailing list xcat-user@lists.sourceforge.netDate: 04/01/2013 07:54 PMSubject: Re: [xcat-user] Help with old-style provisioning
I'm good with that par
I've noticed that in version 2.8 of xCAT users can specify more than one value
for pkgdir in osimage definition.
There appears to be a restriction that the first entry points to the image
source (e.g. /install/rhels6.3/x86_64).
The documentation suggests that the second value for pkgdir
It seems like applying updates to a RHEL6.3 has to be a two-step process and it
seems like it could be just one.
For instance, we have a basic RHEL6.3 installation directory that was created
with the 'copycds' command.
Call it /install/rhels6.3/ x 8 6 _ 6 4
Every quarter we pull
ompute' (profile.osver.architecture.tmpl is most specific, architecture or osver or both being omitted is checked for as the more specific matches are not found) in the respective 'platform' directories ('rh/' for redhat for example)Russell Auld ---04/01/2013 09:05:29 AM---Is there any documentati
Greeting,
First time poster here.
I'm working with an xCAT 2.7 system.
When I run 'lsdef node1', it shows me the postscripts and postbootscripts.
The thing I can't understand is that there is an additional postbootscript that
gets run that doesn't appear in the list generated by
66 matches
Mail list logo