Hi all,
I''m jumping into learning puppet with the modules on Puppetlabs
websites, but I''m running into a few issues and I could use some help.
I''m currently working on the second exercise here:
http://docs.puppetlabs.com/learning/variables.html
I can write a manifest that installs NTP for CentOS:
package { ''ntp'':
ensure => installed,
}
file { ''/etc/ntp.conf'':
owner => ''root'',
group => ''root'',
mode => ''640'',
content => "server 0.rhel.pool.ntp.org",
notify => Service[''ntpd''],
require => Package[''ntp''],
}
service { ''ntpd'':
ensure => running,
enable => true,
hasstatus => true,
hasrestart => true,
}
But when I try to use the $case variable to select the correct OS, I
get an error. Here''s my manifest:
package { ''ntp'':
ensure => installed,
}
file { ''/etc/ntp.conf'':
content => "server 0.rhel.pool.ntp.org",
owner => ''root'',
mode => ''640'',
require => Package[''ntp''],
}
case $operatingsystem {
centos, redhat: { $ntp = "ntp" }
debain, ubuntu: { $ntp = "ntpd" }
default: { fail("Unrecognized operating system for webserver") }
}
if $ntp == ''ntp'' {
service { ''ntp'':
name => $ntp,
ensure => running,
enable => true,
hasrestart => true,
hasstatus => true,
}
}
else {
service { ''ntpd'':
name => $ntp,
ensure => running,
enable => true,
hasstatus => true,
hasrestart => true,
}
}
I''ve tried to reorder things a few different ways but I have a feeling
this is related to how I''m using the case variable and the if/else
statement. My error is this:
err: /Stage[main]//Service[ntp]/ensure: change from stopped to running
failed: Could not start Service[ntp]: Execution of ''/sbin/service ntp
start'' returned 1: at /root/learning-manifests/ntp.pp:24
Which confuses me, because the after the manifest is finished the /etc/
ntp.conf file is written and configured correctly, NTP is installed,
but it''s calling ''/sbin/service ntp start'' as opposed
to ''/etc/init.d/
ntp start.''
When I run the vanilla script, everything configures correctly and
ntp(d) is started. Why is my second script using /sbin when it seems
that it should be using /etc/init.d/?
Thanks!
Jesse
--
You received this message because you are subscribed to the Google Groups
"Puppet Users" group.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscribe@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.
Wil Cooley
2012-Apr-17 05:34 UTC
[Puppet Users] Re: Question regarding multi-OS NTP manifest
On Apr 15, 4:44 pm, Jesse <anonym...@gmail.com> wrote:> case $operatingsystem { > centos, redhat: { $ntp = "ntp" } > debain, ubuntu: { $ntp = "ntpd" }This is backwards; centos/redhat should be ''ntpd'' and debian/ubuntu should be ''ntp''.> if $ntp == ''ntp'' { > service { ''ntp'': > name => $ntp, > ensure => running, > enable => true, > hasrestart => true, > hasstatus => true, > }} > > else { > service { ''ntpd'': > name => $ntp, > ensure => running, > enable => true, > hasstatus => true, > hasrestart => true, > } > }This conditional is redundant. I think it''s better to use one resource name and change the "name" attribute with the variable as you have done (although I would a more explicit variable name like ''$ntp_service''): service { ''ntp'': name => $ntp_service, ... }> err: /Stage[main]//Service[ntp]/ensure: change from stopped to running > failed: Could not start Service[ntp]: Execution of ''/sbin/service ntp > start'' returned 1: at /root/learning-manifests/ntp.pp:24This is because of the reversal I mentioned above; it should be ''ntpd'' on CentOS, not ''ntp''.> When I run the vanilla script, everything configures correctly and > ntp(d) is started. Why is my second script using /sbin when it seems > that it should be using /etc/init.d/?Have you tried ''man service''? ''/sbin/service'' is a way of running init scripts with a sanitized environment and is the preferred way of running init scripts rather than doing so directly. http://linux.die.net/man/8/service -- You received this message because you are subscribed to the Google Groups "Puppet Users" group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscribe@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
On Tuesday, April 17, 2012 1:34:02 AM UTC-4, Wil Cooley wrote:> > On Apr 15, 4:44 pm, Jesse <anonym...@gmail.com> wrote: > > > case $operatingsystem { > > centos, redhat: { $ntp = "ntp" } > > debain, ubuntu: { $ntp = "ntpd" } > > This is backwards; centos/redhat should be ''ntpd'' and debian/ubuntu > should be ''ntp''. > >After spending over an hour trying to figure out my mistake, of course I notice the errant "d" after I post to the group. Thanks for confirming!> > if $ntp == ''ntp'' { > > service { ''ntp'': > > name => $ntp, > > ensure => running, > > enable => true, > > hasrestart => true, > > hasstatus => true, > > }} > > > > else { > > service { ''ntpd'': > > name => $ntp, > > ensure => running, > > enable => true, > > hasstatus => true, > > hasrestart => true, > > } > > } > > This conditional is redundant. I think it''s better to use one resource > name and change the "name" attribute with the variable as you have > done (although I would a more explicit variable name like > ''$ntp_service''): > > service { ''ntp'': > name => $ntp_service, > ... > } > >Thank you, I"ll use that idea in my next manifest.> > err: /Stage[main]//Service[ntp]/ensure: change from stopped to running > > failed: Could not start Service[ntp]: Execution of ''/sbin/service ntp > > start'' returned 1: at /root/learning-manifests/ntp.pp:24 > > This is because of the reversal I mentioned above; it should be ''ntpd'' > on CentOS, not ''ntp''. > > > When I run the vanilla script, everything configures correctly and > > ntp(d) is started. Why is my second script using /sbin when it seems > > that it should be using /etc/init.d/? > > Have you tried ''man service''? ''/sbin/service'' is a way of running init > scripts with a sanitized environment and is the preferred way of > running init scripts rather than doing so directly. > > http://linux.die.net/man/8/serviceI''ll read up on this as well. I''ve created another manifest based on the lesson plan, this time for an Apache install that is OS sensitive, with a custom 404 and a generic landing page. I think this part is wrong: file { ''$webservice_name.conf'': path => "/etc/${webservice_name}/conf/${webservice_name}.conf", Since in Debian/Ubuntu it would be file { ''$webservice_name.conf'': path => "/etc/${webservice_name}/${webservice_name}.conf", but I''m not sure how to parse that out. Here''s it in entirety (Debian side commented out as per the Puppetlabs instructions to test only for EL/CentOS or fail): #init.pp class httpd { case $operatingsystem { centos, redhat: { $webservice_name = ''httpd'' $conf_file = ''/etc/httpd/conf/httpd.conf'' $docu_root = ''/var/www/html'' $404page = ''404page.redhat.centos.html'' } # debian, ubuntu { # $webservice_name = ''apache2'' # $conf_file = ''/etc/apache2/apache2.conf'' # $docu_root = /var/www/'' # $404page = ''404page.debian.ubuntu.html'' #} default: { fail("Uncrecognized operatingsystem for webserver")} } package { ''$webservice_name'' ensure => installed, } file { ''$webservice_name.conf'': path => "/etc/${webservice_name}/conf/${webservice_name}.conf", source => "puppet:///modules/apache2/${conf_file}", ensure => file, require => Package[''$service_name''], } service { ''ntp'': name => $service_name, ensure => running, enable => true, subscribe => File[''${webservice_name.conf}''], } file { ''404page.html'': path => ''$docu_root/404page.html source => "puppet:///modules/apache2/${404page}", ensure => file, require => Package["$webservice_name"], } } I figured I could start another topic, but I didn''t want to clutter the board. Cheers! Jesse On Tuesday, April 17, 2012 1:34:02 AM UTC-4, Wil Cooley wrote:> > On Apr 15, 4:44 pm, Jesse <anonym...@gmail.com> wrote: > > > case $operatingsystem { > > centos, redhat: { $ntp = "ntp" } > > debain, ubuntu: { $ntp = "ntpd" } > > This is backwards; centos/redhat should be ''ntpd'' and debian/ubuntu > should be ''ntp''. > > > if $ntp == ''ntp'' { > > service { ''ntp'': > > name => $ntp, > > ensure => running, > > enable => true, > > hasrestart => true, > > hasstatus => true, > > }} > > > > else { > > service { ''ntpd'': > > name => $ntp, > > ensure => running, > > enable => true, > > hasstatus => true, > > hasrestart => true, > > } > > } > > This conditional is redundant. I think it''s better to use one resource > name and change the "name" attribute with the variable as you have > done (although I would a more explicit variable name like > ''$ntp_service''): > > service { ''ntp'': > name => $ntp_service, > ... > } > > > > err: /Stage[main]//Service[ntp]/ensure: change from stopped to running > > failed: Could not start Service[ntp]: Execution of ''/sbin/service ntp > > start'' returned 1: at /root/learning-manifests/ntp.pp:24 > > This is because of the reversal I mentioned above; it should be ''ntpd'' > on CentOS, not ''ntp''. > > > When I run the vanilla script, everything configures correctly and > > ntp(d) is started. Why is my second script using /sbin when it seems > > that it should be using /etc/init.d/? > > Have you tried ''man service''? ''/sbin/service'' is a way of running init > scripts with a sanitized environment and is the preferred way of > running init scripts rather than doing so directly. > > http://linux.die.net/man/8/serviceOn Tuesday, April 17, 2012 1:34:02 AM UTC-4, Wil Cooley wrote:> > On Apr 15, 4:44 pm, Jesse <anonym...@gmail.com> wrote: > > > case $operatingsystem { > > centos, redhat: { $ntp = "ntp" } > > debain, ubuntu: { $ntp = "ntpd" } > > This is backwards; centos/redhat should be ''ntpd'' and debian/ubuntu > should be ''ntp''. > > > if $ntp == ''ntp'' { > > service { ''ntp'': > > name => $ntp, > > ensure => running, > > enable => true, > > hasrestart => true, > > hasstatus => true, > > }} > > > > else { > > service { ''ntpd'': > > name => $ntp, > > ensure => running, > > enable => true, > > hasstatus => true, > > hasrestart => true, > > } > > } > > This conditional is redundant. I think it''s better to use one resource > name and change the "name" attribute with the variable as you have > done (although I would a more explicit variable name like > ''$ntp_service''): > > service { ''ntp'': > name => $ntp_service, > ... > } > > > > err: /Stage[main]//Service[ntp]/ensure: change from stopped to running > > failed: Could not start Service[ntp]: Execution of ''/sbin/service ntp > > start'' returned 1: at /root/learning-manifests/ntp.pp:24 > > This is because of the reversal I mentioned above; it should be ''ntpd'' > on CentOS, not ''ntp''. > > > When I run the vanilla script, everything configures correctly and > > ntp(d) is started. Why is my second script using /sbin when it seems > > that it should be using /etc/init.d/? > > Have you tried ''man service''? ''/sbin/service'' is a way of running init > scripts with a sanitized environment and is the preferred way of > running init scripts rather than doing so directly. > > http://linux.die.net/man/8/serviceOn Tuesday, April 17, 2012 1:34:02 AM UTC-4, Wil Cooley wrote:> > On Apr 15, 4:44 pm, Jesse <anonym...@gmail.com> wrote: > > > case $operatingsystem { > > centos, redhat: { $ntp = "ntp" } > > debain, ubuntu: { $ntp = "ntpd" } > > This is backwards; centos/redhat should be ''ntpd'' and debian/ubuntu > should be ''ntp''. > > > if $ntp == ''ntp'' { > > service { ''ntp'': > > name => $ntp, > > ensure => running, > > enable => true, > > hasrestart => true, > > hasstatus => true, > > }} > > > > else { > > service { ''ntpd'': > > name => $ntp, > > ensure => running, > > enable => true, > > hasstatus => true, > > hasrestart => true, > > } > > } > > This conditional is redundant. I think it''s better to use one resource > name and change the "name" attribute with the variable as you have > done (although I would a more explicit variable name like > ''$ntp_service''): > > service { ''ntp'': > name => $ntp_service, > ... > } > > > > err: /Stage[main]//Service[ntp]/ensure: change from stopped to running > > failed: Could not start Service[ntp]: Execution of ''/sbin/service ntp > > start'' returned 1: at /root/learning-manifests/ntp.pp:24 > > This is because of the reversal I mentioned above; it should be ''ntpd'' > on CentOS, not ''ntp''. > > > When I run the vanilla script, everything configures correctly and > > ntp(d) is started. Why is my second script using /sbin when it seems > > that it should be using /etc/init.d/? > > Have you tried ''man service''? ''/sbin/service'' is a way of running init > scripts with a sanitized environment and is the preferred way of > running init scripts rather than doing so directly. > > http://linux.die.net/man/8/serviceOn Tuesday, April 17, 2012 1:34:02 AM UTC-4, Wil Cooley wrote:> > On Apr 15, 4:44 pm, Jesse <anonym...@gmail.com> wrote: > > > case $operatingsystem { > > centos, redhat: { $ntp = "ntp" } > > debain, ubuntu: { $ntp = "ntpd" } > > This is backwards; centos/redhat should be ''ntpd'' and debian/ubuntu > should be ''ntp''. > > > if $ntp == ''ntp'' { > > service { ''ntp'': > > name => $ntp, > > ensure => running, > > enable => true, > > hasrestart => true, > > hasstatus => true, > > }} > > > > else { > > service { ''ntpd'': > > name => $ntp, > > ensure => running, > > enable => true, > > hasstatus => true, > > hasrestart => true, > > } > > } > > This conditional is redundant. I think it''s better to use one resource > name and change the "name" attribute with the variable as you have > done (although I would a more explicit variable name like > ''$ntp_service''): > > service { ''ntp'': > name => $ntp_service, > ... > } > > > > err: /Stage[main]//Service[ntp]/ensure: change from stopped to running > > failed: Could not start Service[ntp]: Execution of ''/sbin/service ntp > > start'' returned 1: at /root/learning-manifests/ntp.pp:24 > > This is because of the reversal I mentioned above; it should be ''ntpd'' > on CentOS, not ''ntp''. > > > When I run the vanilla script, everything configures correctly and > > ntp(d) is started. Why is my second script using /sbin when it seems > > that it should be using /etc/init.d/? > > Have you tried ''man service''? ''/sbin/service'' is a way of running init > scripts with a sanitized environment and is the preferred way of > running init scripts rather than doing so directly. > > http://linux.die.net/man/8/serviceOn Tuesday, April 17, 2012 1:34:02 AM UTC-4, Wil Cooley wrote:> > On Apr 15, 4:44 pm, Jesse <anonym...@gmail.com> wrote: > > > case $operatingsystem { > > centos, redhat: { $ntp = "ntp" } > > debain, ubuntu: { $ntp = "ntpd" } > > This is backwards; centos/redhat should be ''ntpd'' and debian/ubuntu > should be ''ntp''. > > > if $ntp == ''ntp'' { > > service { ''ntp'': > > name => $ntp, > > ensure => running, > > enable => true, > > hasrestart => true, > > hasstatus => true, > > }} > > > > else { > > service { ''ntpd'': > > name => $ntp, > > ensure => running, > > enable => true, > > hasstatus => true, > > hasrestart => true, > > } > > } > > This conditional is redundant. I think it''s better to use one resource > name and change the "name" attribute with the variable as you have > done (although I would a more explicit variable name like > ''$ntp_service''): > > service { ''ntp'': > name => $ntp_service, > ... > } > > > > err: /Stage[main]//Service[ntp]/ensure: change from stopped to running > > failed: Could not start Service[ntp]: Execution of ''/sbin/service ntp > > start'' returned 1: at /root/learning-manifests/ntp.pp:24 > > This is because of the reversal I mentioned above; it should be ''ntpd'' > on CentOS, not ''ntp''. > > > When I run the vanilla script, everything configures correctly and > > ntp(d) is started. Why is my second script using /sbin when it seems > > that it should be using /etc/init.d/? > > Have you tried ''man service''? ''/sbin/service'' is a way of running init > scripts with a sanitized environment and is the preferred way of > running init scripts rather than doing so directly. > > http://linux.die.net/man/8/serviceOn Tuesday, April 17, 2012 1:34:02 AM UTC-4, Wil Cooley wrote:> > On Apr 15, 4:44 pm, Jesse <anonym...@gmail.com> wrote: > > > case $operatingsystem { > > centos, redhat: { $ntp = "ntp" } > > debain, ubuntu: { $ntp = "ntpd" } > > This is backwards; centos/redhat should be ''ntpd'' and debian/ubuntu > should be ''ntp''. > > > if $ntp == ''ntp'' { > > service { ''ntp'': > > name => $ntp, > > ensure => running, > > enable => true, > > hasrestart => true, > > hasstatus => true, > > }} > > > > else { > > service { ''ntpd'': > > name => $ntp, > > ensure => running, > > enable => true, > > hasstatus => true, > > hasrestart => true, > > } > > } > > This conditional is redundant. I think it''s better to use one resource > name and change the "name" attribute with the variable as you have > done (although I would a more explicit variable name like > ''$ntp_service''): > > service { ''ntp'': > name => $ntp_service, > ... > } > > > > err: /Stage[main]//Service[ntp]/ensure: change from stopped to running > > failed: Could not start Service[ntp]: Execution of ''/sbin/service ntp > > start'' returned 1: at /root/learning-manifests/ntp.pp:24 > > This is because of the reversal I mentioned above; it should be ''ntpd'' > on CentOS, not ''ntp''. > > > When I run the vanilla script, everything configures correctly and > > ntp(d) is started. Why is my second script using /sbin when it seems > > that it should be using /etc/init.d/? > > Have you tried ''man service''? ''/sbin/service'' is a way of running init > scripts with a sanitized environment and is the preferred way of > running init scripts rather than doing so directly. > > http://linux.die.net/man/8/serviceOn Tuesday, April 17, 2012 1:34:02 AM UTC-4, Wil Cooley wrote:> > On Apr 15, 4:44 pm, Jesse <anonym...@gmail.com> wrote: > > > case $operatingsystem { > > centos, redhat: { $ntp = "ntp" } > > debain, ubuntu: { $ntp = "ntpd" } > > This is backwards; centos/redhat should be ''ntpd'' and debian/ubuntu > should be ''ntp''. > > > if $ntp == ''ntp'' { > > service { ''ntp'': > > name => $ntp, > > ensure => running, > > enable => true, > > hasrestart => true, > > hasstatus => true, > > }} > > > > else { > > service { ''ntpd'': > > name => $ntp, > > ensure => running, > > enable => true, > > hasstatus => true, > > hasrestart => true, > > } > > } > > This conditional is redundant. I think it''s better to use one resource > name and change the "name" attribute with the variable as you have > done (although I would a more explicit variable name like > ''$ntp_service''): > > service { ''ntp'': > name => $ntp_service, > ... > } > > > > err: /Stage[main]//Service[ntp]/ensure: change from stopped to running > > failed: Could not start Service[ntp]: Execution of ''/sbin/service ntp > > start'' returned 1: at /root/learning-manifests/ntp.pp:24 > > This is because of the reversal I mentioned above; it should be ''ntpd'' > on CentOS, not ''ntp''. > > > When I run the vanilla script, everything configures correctly and > > ntp(d) is started. Why is my second script using /sbin when it seems > > that it should be using /etc/init.d/? > > Have you tried ''man service''? ''/sbin/service'' is a way of running init > scripts with a sanitized environment and is the preferred way of > running init scripts rather than doing so directly. > > http://linux.die.net/man/8/service-- You received this message because you are subscribed to the Google Groups "Puppet Users" group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/LoiCirfBCgkJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscribe@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Ooops, forgot the default page:
file { ''default.html'':
path => "${docu_root}/default.html",
source => "puppet:///modules/apache2/default.html",
ensure => file,
require => Package["$webservice_name"],
--
You received this message because you are subscribed to the Google Groups
"Puppet Users" group.
To view this discussion on the web visit
https://groups.google.com/d/msg/puppet-users/-/dqZp4afWNP4J.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscribe@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.