Bug #3710
Facter not working on Mac 10.5.6
| Status: | Rejected | Start date: | 05/03/2010 | |
|---|---|---|---|---|
| Priority: | High | Due date: | ||
| Assignee: | % Done: | 0% |
||
| Category: | - | |||
| Target version: | - | |||
| Keywords: | mac | Affected Facter version: | ||
| Branch: | ||||
| Votes: | 0 |
Description
It appears puppetd is not working properly on Mac OS X 10.5.6 (latest 10.5.x DVD shipped from Apple). It appears puppetd is not able to create /etc/puppet/ssl or /var/puppet/*
We tried with 0.25.1 and 0.25.5rc2
bash-3.2# puppetd --debug
debug: Failed to load library 'selinux' for feature 'selinux'
debug: Failed to load library 'shadow' for feature 'libshadow'
debug: Puppet::Type::User::ProviderPw: file pw does not exist
debug: Puppet::Type::User::ProviderUseradd: file useradd does not exist
debug: Puppet::Type::User::ProviderUser_role_add: file rolemod does not exist
debug: Failed to load library 'ldap' for feature 'ldap'
debug: Puppet::Type::User::ProviderLdap: feature ldap is missing
2010-05-03 09:19:05.095 system_profiler[1034:10b] Exception while calling [SPPlatformReporter updateDictionary:]
*** -[NSCFArray objectAtIndex:]: index (3) beyond bounds (2)
err: Could not create resources for managing Puppet's files and directories in sections [:main, :puppetd, :ssl]: undefined method `each' for nil:NilClass
/Library/Ruby/Site/1.8/puppet/util/pidlock.rb:33:in `initialize': No such file or directory - /var/lib/puppet/run/puppetd.pid (Errno::ENOENT)
from /Library/Ruby/Site/1.8/puppet/util/pidlock.rb:33:in `open'
from /Library/Ruby/Site/1.8/puppet/util/pidlock.rb:33:in `lock'
from /Library/Ruby/Site/1.8/puppet/daemon.rb:46:in `create_pidfile'
from /System/Library/Frameworks/Ruby.framework/Versions/1.8/usr/lib/ruby/1.8/sync.rb:229:in `synchronize'
from /Library/Ruby/Site/1.8/puppet/daemon.rb:45:in `create_pidfile'
from /Library/Ruby/Site/1.8/puppet/daemon.rb:21:in `daemonize'
from /Library/Ruby/Site/1.8/puppet/application/puppetd.rb:233:in `run_setup'
from /Library/Ruby/Site/1.8/puppet/application.rb:216:in `run'
from /Library/Ruby/Site/1.8/puppet/application.rb:306:in `exit_on_fail'
from /Library/Ruby/Site/1.8/puppet/application.rb:216:in `run'
from /usr/sbin/puppetd:160
History
Updated by James Turnbull about 2 years ago
- Category set to OSX
- Status changed from Unreviewed to Accepted
- Assignee set to Markus Roberts
- Priority changed from Normal to High
- Target version set to 0.25.5
Updated by Nigel Kersten about 2 years ago
We’re not seeing these issues with 0.25.4 at all.
This smells familiar though. Are you getting the same errors when simply running ‘facter’ ?
Apple made some changes to the XML output of system_profiler that produce errors like this.
Updated by Roy Nielsen about 2 years ago
I’m getting the following when trying to run facter on my 10.5.6 machine:
$ facter
2010-05-04 13:26:15.567 system_profiler[5019:10b] Exception while calling [SPPlatformReporter updateDictionary:]
*** -[NSCFArray objectAtIndex:]: index (3) beyond bounds (2)
/Library/Ruby/Site/1.8/facter/macosx.rb:28: undefined method `each' for nil:NilClass (NoMethodError)
from /Library/Ruby/Site/1.8/facter/util/loader.rb:73:in `load'
from /Library/Ruby/Site/1.8/facter/util/loader.rb:73:in `load_file'
from /Library/Ruby/Site/1.8/facter/util/loader.rb:38:in `load_all'
from /Library/Ruby/Site/1.8/facter/util/loader.rb:33:in `each'
from /Library/Ruby/Site/1.8/facter/util/loader.rb:33:in `load_all'
from /Library/Ruby/Site/1.8/facter/util/loader.rb:30:in `each'
from /Library/Ruby/Site/1.8/facter/util/loader.rb:30:in `load_all'
from /Library/Ruby/Site/1.8/facter/util/collection.rb:90:in `load_all'
from /Library/Ruby/Site/1.8/facter.rb:91:in `to_hash'
from /usr/bin/facter:138
So – Do you think it would help to update the system to 10.5.8 before trying to run puppet on the machine?
Regards, -Roy
Updated by Nigel Kersten about 2 years ago
I believe a later 10.5.x release changed the XML output, 10.5.7 perhaps?
You can see the differences with system_profiler -xml SPHardwareDataType
Have you tried different versions of Facter?
Once you’ve worked out which point release changes the output, and how, we can add some error checking to Facter to support both differences in formats.
Updated by James Turnbull about 2 years ago
- Assignee changed from Markus Roberts to Nigel Kersten
So this is Factor issue rather than a Puppet one?
Updated by Nigel Kersten about 2 years ago
I believe so, yes.
Updated by James Turnbull about 2 years ago
- Project changed from Puppet to Facter
- Category deleted (
OSX) - Target version deleted (
0.25.5) - Support Urls deleted (
)
Updated by James Turnbull about 2 years ago
- Subject changed from puppet not working on Mac 10.5.x to Facter not working on Mac 10.5.x
Updated by Roy Nielsen about 2 years ago
Updating to 10.5.8 seemed to do the trick for me for both running puppet & for running facter.
Regards, -Roy
Updated by Nigel Kersten about 2 years ago
Roy, is there any chance you can paste into this ticket the differences between 10.5.6 and 10.5.8 for this command?
system_profiler -xml SPHardwareDataType
If there is no difference, then I’ve got the type wrong, and please attach to this ticket two different files for the full:
system_profiler -xml
Thanks.
Updated by Roy Nielsen about 2 years ago
- File SPHardwareDataType10.5.6.txt added
- File SPHardwareDataType10.5.8.txt added
- File full_system_profiler10.5.6.txt added
- File full_system_profiler10.5.8.txt added
I’ve attached the four files.
Regards, -Roy
Updated by Markus Roberts about 2 years ago
Roy —
Thanks.
— Markus
Updated by Nigel Kersten about 2 years ago
Roy, is there any chance we could get this info from you for actual Mac hardware rather than a VMware install?
There are some weird things that get reported via system profiler for virtual instances, and I’d like to check how the 99% case looks for this situation.
You can see how the info for some of this info isn’t really correct under 10.5.6 for the virtual instance, and I have a feeling from memory that the info is actually malformed in that situation and we may simply have to fail gracefully in Facter.
Updated by Roy Nielsen about 2 years ago
Hello,
I’ll try to get that information to you this morning.. I might have to re-install 10.5 on my older mini to get 10.5.6.
Regards, -Roy
Updated by Roy Nielsen about 2 years ago
- File SPHardwareDataType10.5.8.txt added
- File full_system_profiler10.5.8.txt added
Here’s the 10.5.8 stuff attached, more to come in a couple of hours…
Updated by Roy Nielsen about 2 years ago
- File SPHardwareDataType10.5.1.txt added
- File full_system_profiler10.5.1.txt added
My apologies –
The install I thought was 10.5.6 was 10.5.1. Attached are the files.
Updated by Nigel Kersten almost 2 years ago
- Status changed from Accepted to Needs Decision
Huh. I’m even more confused now. The 10.5.1 looks to be the same format as the 10.5.8 output, but the 10.5.6 output does look different to both. Perhaps Apple messed it up and then fixed it later.
It looks like Apple completely forgot to include the _items entry in 10.5.6.
We could fix this, but I’m not sure if we should, given that it’s a single point release that seems to exhibit this problem, and it’s going to complicate the facter code significantly.
Can I get a design decision from one of the core Facter maintainers about this?
Updated by Nigel Kersten almost 2 years ago
- Subject changed from Facter not working on Mac 10.5.x to Facter not working on Mac 10.5.6
Updated by Luke Kanies almost 2 years ago
- Status changed from Needs Decision to Rejected
Sorry for the long delay here – I didn’t realize it needed a decision from me.
I’m comfortable rejecting this, given the constraints you’ve provided.