The work around that we found quite helpful is to update /etc/yum.conf,
to point to a different repository directory path. That way if another
repository is added through CentOS or through any other rpm, we do not
have to retrofit or update the kickstart.
We do this with the following
Item #1 - Taken directly from one of my spacewalks config
/etc/rhn/rhn.conf
#https://www.redhat.com/archives/spacewalk-list/2016-September/msg00016.html
# The number of days to wait to hear from a system before warning that it
is not checking in.
web.system_checkin_threshold = 10
Item #2, I
You can, just pay attention to the NOTE from item 2. (Updated) Sorry
about the duplication, I missed a couple of steps for item 2. They have
been updated and are items e and f.
1. Create a distribution.
You may have this already done. If not, there is plenty of
documentation
You can, just pay attention to the NOTE from item 2.
1. Create a distribution.
You may have this already done. If not, there is plenty of
documentation out there on how to do this.
2. Create a kickstart profile for each parent (base) channel, and select
all of the child channels
Was able to change it with the following... Need to put the schedule in '
'.
# rpm -q spacecmd
spacecmd-2.6
#spacecmd -q -s softwarechannel_listsyncschedule |grep epel6-x86_64
655 epel6-x86_640 0 21 ? * 1
#spacecmd -q -s softwarechannel_setsyncschedule
Attempting to completely remove a configuration file in a global
configuration channel (exposed to 9+ devices), the following error is
being generated.I was able to remove 2 other configuration files, and
all revisions, in the same configuration channel earlier today.
The file is
The image replacement will be interesting, would love to see instructions
on how to do that. However the header and welcome changes should be able
to be done through the following:
Update /etc/rhn/rhn.conf
# The following overrides the defaults in
/usr/share/rhn/config-defaults/*.conf
#
There is another option mentioned in the archive, about updating
/etc/yum.conf with a new reposdir=. This avoids updating/creating the
repositories when the CentOS release rpm is updated with the next kernel
release. Oracle and Redhat do not include the repository information in
the primary
If /etc/sysconfig/rhn/rhnsd is not there, then need to investigate the
package rhnsd. I would try and reinstall the rpm, and check to see if the
file is there.
If the file is not there:
What version of rhnsd is installed on the clients?
What files are being provided by the rpm?
You can also cron the check-in. Also, if the hardware/profile sync is not
put in cron, the hostname or IP changes are never updated.
#!/bin/bash
PATH=/usr/sbin:/sbin:/bin:/usr/bin
# Implemented on the RedHat Linux based systems supported by the five
spacewalk environments.
###
Our solution to this was to add the following to /etc/yum.conf. This
solution also prevents other vendor repositories from dropping in a config
file that would cause the same time out issue, since the servers can not
reach them anyway.
reposdir=/somepath/
- Thanks and good luck.
From:
Typically for my configurations, it is almost immediate. However, after a
successful yum update, I run "rhn-check" and "rhn-profile-sync". This
will update the DB immediately.
- Thanks and good luck
From: Mario Mikocevic
To: spacewalk-list@redhat.com
Yes.. Someone has in the past manually ran a reposync command and pointed
it to the el7 vs the el6 repository URL. You may want to also validate
the repository URLs to Channels mapping.
Channels -> Manage Software Channels -> Manage repositories
To audit all of them: If they all have
se rhn-action commands need to be run on each individual client? Or
do they need to be run on the Spacewalk server?
On Tue, Apr 10, 2018 at 5:50 PM, William H. ten Bensel <whten...@up.com>
wrote:
Spacewalk should be setup to already schedule. IF you want the commands
to run immediately,
We had to go higher in memory allocation -->
# Increase java heap size and logging:
https://access.redhat.com/solutions/43122
taskomatic.java.maxmemory=12288
taskomatic.java.initmemory=512
taskomatic.logfile.maxsize=100m
java.taskomatic_channel_repodata_workers=4
tomcat6 -->
-Xms512m
Spacewalk should be setup to already schedule. IF you want the commands
to run immediately, then install osad on the clients. However there is a
cost with doing that (Will not cover). If you do not install osad, on the
next rhn-check (default every 4 hours) the commands will run.
On the
Just as FYI --> To avoid running the spaceawalk-hostname-rename scripts in
the future and easily migrate from one server to another, you can use a
cname for all of the details.
IE You can have a standby server with spacewalk installed on. Then
all you have to do is flip the cname and
system before warning that it
is not checking in. Does not change the email notifications, just the
GUI.
web.system_checkin_threshold = ###
- Thanks and good luck
From: "William H. ten Bensel" <whten...@up.com>
To: spacewalk-list@redhat.com
Date: 03/21/2018 09:32
Any changes from the default should be placed in /etc/rhn/rhn.conf. The
start process reads the defaults, then processes the rhn.conf for any
additional settings.
Here is an example of increasing the rhn-search memory.
[root]# grep search /etc/rhn/rhn.conf
# Increase rhn-search
a Road, Melbourne, Victoria 3004 Australia
___
Spacewalk-list mailing list
Spacewalk-list@redhat.com
https://www.redhat.com/mailman/listinfo/spacewalk-list
[attachment "smime.p7s" deleted by William H. ten Bensel/UPC]
For the command "satellite-sync", is there a way to indicate not to use a
proxy even when rhn.conf has one specified? Or is there a configuration
on the the child spacewalk to allow the command to bypass the proxy? I
see --http-proxy as an option, but if left blank or any data exists, the
ely.
From: spacewalk-list-boun...@redhat.com [
mailto:spacewalk-list-boun...@redhat.com] On Behalf Of William H. ten
Bensel
Sent: Thursday, June 08, 2017 3:34 PM
To: spacewalk-list@redhat.com
Subject: Re: [Spacewalk-list] Issues with kickstarted systems. [External]
Be aware: [This message came from
error, please
notify the sender by replying to this message and delete this e-mail
immediately.
From: spacewalk-list-boun...@redhat.com [
mailto:spacewalk-list-boun...@redhat.com] On Behalf Of William H. ten
Bensel
Sent: Thursday, June 08, 2017 3:34 PM
To: spacewalk-list@redhat.com
Subj
The registration issues, during kickstart and manual, are related... The
registration failure is caused by the kickstart script, "Registration and
server actions" as it does not do dependency checks when doing the
following command:
rpm -Uvh --replacepkgs --replacefiles
I also clean up the cache.
/var/cache/yum/*
- Thanks and good luck
From: Michael Mraka
To: spacewalk-list@redhat.com
Date: 06/02/2017 03:50 AM
Subject:Re: [Spacewalk-list] How to "unregister" a Spacewalk
client from the server
Sent by:
I believe this can be done:
Linux 6: update /etc/hosts, /etc/sysconfig/network and hostname with what
you want it to be.
Linux 7: update /etc/hosts and /etc/hostname
Then regenerate the SSL.
- Thanks and good luck
From: "Fouts, Christopher"
To:
The issue with disabling is, when you upgrade the settings for the repos
configurations are re-enabled by the package centos-release-x-y. So, you
have to do the entire disable step again. RedHat and Oracle installs do
not do this, because they expect you to be using Satellite/Spacewalk/other
Instead of re-registering each server remotely. Could you just switch
their base channels or add/remove child channels from the spacewalk
itself?
BASE:
spacecmd -q -s myspacewalk.domain.com softwarechannel_listbasechannels
spacecmd -q -s myspacewalk.domain.com system_listbasechannel $myserver
links.
On Tue, Feb 28, 2017 at 11:14 AM, William H. ten Bensel <whten...@up.com>
wrote:
> On the Spacewalk server:
> validate that the installer created:
> /var/www/html/pub/RHN-ORG-TRUSTED-SSL-CERT. If not it should have also
been
> created in /root/ssl-build/. Copy RHN-
by the Spacewalk application itself.
On Tue, Feb 28, 2017 at 7:08 AM, William H. ten Bensel <whten...@up.com>
wrote:
> Hopefully this will assist, a shot in the dark... Most of the following
is
> from a kickstart. The !! comments and actions are additional steps that
may
> or may not be
Hopefully this will assist, a shot in the dark... Most of the following
is from a kickstart. The !! comments and actions are additional steps
that may or may not be needed.
Try reinstalling the following on the client side.
# below pkgs needed for rhn registration and configuration
I believe the User and password are stored in the rhn.conf
awk -F= '/db_[user|password]/ {print $2}' /etc/rhn/rhn.conf
- Thanks and good luck
From: Daryl Rose
To: "spacewalk-list@redhat.com"
Date: 02/24/2017 12:44 PM
Subject:
To address some additional questions...
1. How to make the init.d start/stop script dynamic?
rhnconf = "/etc/rhn.conf"
if [[ "$srvr_product" == "spacewalk" ]]; then
spacecname=$(grep "cobbler.host" $rhnconf |awk -F"=" '{print
$2}'|sed 's/^[[:space:]]//g'|tr -d '\n')
elif
Not crazy at ALL, that is what I do+. Hopefully this brain dump will
help.
GOTCHA's . If you are using ISS to sync other Spacewalks... You have
to put BOTH failoverhost.network.com and primaryhost.network.com in the
list! Spacewalk does not associate ALIAS to connecting server. It
Try and utilize the command lines.
spacewalk-clone-by-date
!!!RESTART TASKOMATIC before running this:
/usr/bin/spacewalk-manage-channel-lifecycle -c $fromchnl --no-children
--promote
Base -> Always updating from the public repo's
date{x}Base -> Frozen at Date {x}
*Note, some 3rd
Hopefully this helps. The times are in quartz standard.
spacecmd -q softwarechannel_listsyncschedule |awk '$3 ~ /[a-zA-Z0-9]/
{print $0}'
- Thanks and good luck
From: Daryl Rose
To: "spacewalk-list@redhat.com"
Date: 01/10/2017
Anybody have any suggestions on the best way of uploading files to config
channels? spacecmd almost meets all of my requirements but one.
Last Requirement is: If it is a text file, the file needs to be
identified as such and viewable in the GUI.
Here is what I have tried and found
Franky nailed it.
Just some additional insight (do not know how many devices/servers that
are supported):
If there ever is a chance of mass rebooting 1k+ servers/devices at once,
the rhn_check's could degrade the spacewalk. Adding something like the
perl command below, will assist in
- Schedule a kernel update from Web GUI, let spacewalk agent do it's job
and get kernel update (/boot/initramfs missing)
Just saw something similar today.
Issue: Timed out (+2 hours) in the middle of the yum upgrade process.
Solution: Reboot to previous kernel, run
I believe this option added to /etc/rhn/rhn.conf should perform the
requested notification time frame that you are looking for:
# The following overrides the defaults in
/usr/share/rhn/config-defaults/*.conf
# The number of days to wait to hear from a system before warning that it
is not
I would like to use the spacewalk api system.provisionVirtualGuest,
however I need the ability to pass extra args to the virt-install line
that spacewalk would create. Is there any way to do this? Am I missing
something obvious?
Reasons:
Do not use DHCP, PXE, etc. static IP only.
one out there authenticating their Spacewalk users against AD? Or
even LDAP. If so…. Any pointers, or links to good information on how to
set it up ?
Kind Regards,
Jeff
Jeff Baldwin
8335 Classic Drive
Charlotte, NC 28262
800.368.1056 toll free
704.597.9015 main
knowclassic.co
This is a solution (might not be the best way) to prevent re-registering,
hours of importing/exporting channels, and setting up an entire new
spacewalk server with lots of downtime. Can be used in case the Active
Spacewalk needs to be down longer then the Service level agreement. For
this
I am making the assumption that you are referring to the spacewalk profile
id, let say abc123 and the hostname is now xyz123:
On server/device /etc/sysconfig/rhn/systemid will have the abc123:
profile_name
abc123
Now you want to rename the spacewalk profile:
# spacecmd -s
Just a side note, next time the package centos-release is updated, it is
going to run into the same issue again. The redhat-release does not
include /etc/yum.repos.d/* files but the centos-release does.
Refer to the yum.conf modification in this post, as a work around.
For security, it would be best to setup a proxy.
https://fedorahosted.org/spacewalk/wiki/HowToInstallProxy
If unable to setup a proxy, the ports are listed in the documentation.
- Thanks and good luck
From: Daryl Rose
To: "spacewalk-list@redhat.com"
I just responded today on how to pass console information to a kvm
virtual:
https://www.redhat.com/archives/spacewalk-list/2016-April/msg00070.html
- Thanks and good luck
From: Lachlan Musicman
To: spacewalk-list@redhat.com
Date: 04/11/2016 09:53 PM
Subject:
- Thanks and good luck
From: Konstantin Raskoshnyi
To: spacewalk-list@redhat.com
Date: 04/11/2016 08:27 PM
Subject:Re: [Spacewalk-list] PRE Section doesn't work at all KVM
Guest
Sent by:spacewalk-list-boun...@redhat.com
There's no file in tmp
Hopefully this example will help. You should be able to configure your
hostname logic like the following:
We had to do the following with CentOS 7.2 and Oracle Linux 7.2, because
there is a bug with how the static --network line is being generated. (Bug
opened with support and they have
Spacecmd keeps a cache in ~/.spacecmd/{localhost/spacewalk}
Try:
spacecmd -u -p x -s xx clear_caches
or
cd ~/.spacecmd/xx (localhost or spacewalk)
rm the files in there.
Sometimes, just to clear things out .. rm -rf ~/.spacecmd
- Thanks and good luck
From:
It is in extra arguments, ie:
virt-install --name= --ram=8192 --vcpus=4 --mac=
--location=http://spacewalk.domain/distro/centos --disk path=disk
--bridge= --extra-args="ks=http://spacewalk.domain/kickstart noipv6
noselinux serial console=tty0 console=ttyS0,115200 ip=x
- Thanks and good luck
From: Lachlan Musicman
To: spacewalk-list@redhat.com
Date: 04/04/2016 01:36 AM
Subject:Re: [Spacewalk-list] Can't install post kickstart install
Sent by:spacewalk-list-boun...@redhat.com
This email originated from outside of
- Thanks and good luck
From: Lachlan Musicman
To: spacewalk-list@redhat.com
Date: 03/31/2016 07:31 PM
Subject:[Spacewalk-list] Kickstart, Config management and
debugging
Sent by:spacewalk-list-boun...@redhat.com
This email originated from outside
ked, because we can use the "hack"
If any of the files in a configuration channel needs to be pushed,
you can use the UI. or if you need all of the configuration channels
pushed use spacecmd.
From: "William H. ten Bensel" <whten...@up.com>
To: spacewalk-lis
nt
Sent by:spacewalk-list-boun...@redhat.com
This email originated from outside of the company. Please use discretion
if opening attachments or clicking on links.
On 30 March 2016 at 22:48, William H. ten Bensel <whten...@up.com> wrote:
That is a "hack" work around. After the 1
ent
Sent by:spacewalk-list-boun...@redhat.com
This email originated from outside of the company. Please use discretion
if opening attachments or clicking on links.
On 30 March 2016 at 22:48, William H. ten Bensel <whten...@up.com> wrote:
Note: To prevent updating multiple kickstarts whe
is
this a hack to get around the "something isn't working, but we don't know
what it is" problem? I don't mind - I am happy to have it working :)
L.
--
The most dangerous phrase in the language is, "We've always done it this
way."
- Grace Hopper
On 30 March 2016 at 13:1
*Deployed*... Does that mean activate/re-activate OR the installation of
the rpms did not get the files OR an rhncfg-client get failed?
Activate/re-activate should have pushed the files.
Installation of the rpms by themselves, will not push the files.
rhncfg-client get or a schedule through
gt; Also, /etc/tomcat6/tomcat6.conf has some settings for JAVA_OPTS that are
>>>>>> interesting for tuning purposes:
>>>>>>
>>>>>> JAVA_OPTS="-XX:NewRatio=4 -XX:PermSize=1024m -XX:MaxPermSize=2048m
>>>>>>
.
The spacewalk version is 2.4, Oct 7th, 2015
Thanks!
On Thu, Mar 24, 2016 at 6:21 PM, William H. ten Bensel <whten...@up.com>
wrote:
What version of spacewalk are you running?
Bill
On Mar 24, 2016, at 20:15, Konstantin Raskoshnyi <konra...@gmail.com>
wrote:
This email originated
What version of spacewalk are you running?
Bill
> On Mar 24, 2016, at 20:15, Konstantin Raskoshnyi wrote:
>
> This email originated from outside of the company. Please use discretion if
> opening attachments or clicking on links.
>
> Hi Community!
> We have a spacewalk
activation keys for a group
of repositories including upstream Cetnos and a repository for our
internal code.
On Thu, Mar 10, 2016 at 6:41 AM, William H. ten Bensel <whten...@up.com>
wrote:
We do this, however it is by design. Hopefully this will help.
It is accomplished by:
A Kickst
Python would be the best way, but if you are interested in bash/spacecmd
examples. Just some quick examples.
for server in `spacecmd -s xx -u y -p zz -q group_listsystems
group`; do
for errata in `spacecmd -s xx -u y -p zz -q
system_listerrata $server |egrep
We do this, however it is by design. Hopefully this will help.
It is accomplished by:
A Kickstart profile is pointed to a "clone-by-date" base channel, updates,
extras, debuginfo, and spacewalk-client. This gives us 6.1, 6.2, 6.5,
6.5r1, 6.5r2, 6.6, kickstarts.
You can do the following with an external DB and a cname. This is not an
active/active solution, but active/standby. This method has it's own
pros and cons.
1. Install spacewalk/satellite (same version) on one server at location x
and another server at location y (do not update schema on
Uncertain about the getVar commands. However the ks= can be found during
the kickstart in /proc/cmdline as it has the entire command line that was
sent via the anaconda boot options. This would include the ks= and other
variable that can be obtained as well. ie. ip, gateway, netmask,
Some additional information:
A cname will also work. It does not have to be FQDN.
hostname cname
update /etc/hosts with the cname information.
This allows an easier migration from one spacewalk to another. ie. To a
hot standby (all relevant spacewalk data replicated with external DB).
- Thanks
We ran into issues where "yum update" updated the centos-release RPM,
which in turn would push updated .repos to /etc/yum.repos.d/. This would
cause a similar issue to the errors described.
The solution we came up with, was to put the following in the post section
of the default kickstart:
Have you tried the spacewalk-manage-channel-lifecycle utility?
RPM which provides the utility: spacewalk-utils
/usr/bin/spacewalk-manage-channel-lifecycle --help
Examples:
List the base channels and child channels
/usr/bin/spacewalk-manage-channel-lifecycle -u $conf{org1user} -p
Is there any way to reduce the CPU usage via a configuration change (ie
only schedule on % servers during these specified times), batch run time
change, or increase the time between the scheduled tasks for each server?
Any other ideas/suggestions on how to reduce the CPU usage for this task?
-
Hello Spacewalk community,
Issue:
Servers will not register
Root Cause:
We updated an older kickstart centos6.6 with a root password
change (not a major change), and now the kickstart will no longer register
the servers. We found that when the kickstart was updated, the
71 matches
Mail list logo