On Thu, Dec 12, 2013 at 7:21 AM, Dan Bode <bod...@gmail.com> wrote:
>
>
> On Thu, Dec 12, 2013 at 1:46 AM, Nan Liu <nan....@gmail.com> wrote:
>
>> On Wed, Dec 11, 2013 at 8:51 AM, Dan Bode <bod...@gmail.com> wrote:
>>>
>>> I had a bit of time to research the existing device code to see if I can
>>> use it for an integration with two specific use cases:
>>>
>>
>> I'm not sure what issues are still actively worked on, and I'm keeping an
>> eye on the redmine migration to see what gets ported over. I've had onsite
>> discussion with PL developers, and I would love to get more feedback and
>> roadmap for devices v.s. transport. For now, I'm staying with transport
>> resources. Comments below.
>>
>>
>>> 1. discovery/inventory - access hardware inventory and store it
>>> somewhere where it can be retrieved.
>>>
>>> So far, device supports this use case.
>>> - specify a list of device endpoints in device.conf
>>> - run puppet device to get their facts to serve as inventory (although
>>> puppet device looks like it gets facts and requests catalogs, I will
>>> probably call the facts method directly to just get the facts)
>>> - have the front end query these facts from PuppetDB
>>>
>>
>> puppet device facts are not really invoked via facter, and have some
>> gotchas (such as symbol keys). They are tucked away in
>> lib/puppet/util/network_devices/<device_name>/facts.rb. However since
>> facter is not available for puppet device, the only win for device is
>> inventory in puppetdb. The missing functionality can be implemented as a
>> resource for transport solution which exports facts via puppet face.
>>
>
> I can't think of a case where I need facts from a device to make a
> configuration decisions. Perhaps I'm just not far enough into it :)
>
In theory you should be able to detect the device version and use the
appropriate provider. In practice, the one place I could use this
functionality required different versions of rubygem to even connect to the
device.
2. management - manage the process of bringing up a cluster from scratch
>>>
>>> This is the use case where puppet device is problematic.
>>>
>>> In this use case, an external system needs to specify how a collection
>>> of resources should be configured. The types of these resources are
>>> heterogeneous, for example:
>>>
>>> - Server
>>> - Storage
>>> - Network
>>> - add Port
>>> - create server
>>>
>>> These hardware configuration rules (and their dependencies) map pretty
>>> cleanly to the Puppet DSL and the Resource/Graph model. Where a manifests
>>> represents multiple devices and multiple endpoints.
>>>
>>
>> This is one of the main reason I'm using transport since it expresses
>> cross node dependency using the existing DSL.
>>
>>
>>> I had the following issues with puppet device for this use case:
>>>
>>> 1. It iterates through the endpoints and configures them one at a time
>>>
>>> This is probably the biggest barrier. I need to keep track of a
>>> collection of resources that target multiple endpoints and apply them in a
>>> certain order. Looking at the device code it seems to just iterate through
>>> the endpoints in device.conf and configure them one at a time.
>>>
>>> I spent some time thinking about the current device command and how I
>>> might use it to configure workflows across multiple endpoints.
>>> - on the puppet master, keep a queue (or list) for each endpoint that
>>> needs to be configured
>>> - have an external process (the dispatcher) that keeps track of the
>>> configuration that needs to be applied (along with their endpoints) and
>>> stores the resources that represent that configuration into the correct
>>> queue for it's endpoint.
>>> - have an ENC that checks the certname of a device when it checks in,
>>> maps it to a queue, and clears all entries for a queue (for it to apply)
>>> - If the dispatcher keeps track of all of the resources that it put onto
>>> which queue, it can track the report for those devices to know when it's
>>> entire job is completed.
>>>
>>> The above explanation is the best way I could think of to use the
>>> existing device, but it is cumbersome enough that it warrants not using the
>>> device model.
>>>
>>> 2. it does not allow for the specification of dependencies between
>>> multiple device endpoints. It only allows for certain endpoints to be
>>> processed in a certain order.
>>>
>>> This is pretty much the same as #1, but worth mentioning separately.
>>>
>>> 3. It invents its own command line for doing things (it does not cleanly
>>> operate with puppet resource, puppet apply, puppet agent with represents a
>>> major loss of functionality)
>>>
>>> 4. Management of device.conf
>>>
>>> The existence of device.conf creates its own management issues. You need
>>> to assign a single node to a single device, you have to manage the process
>>> for getting the credentials to that device, you have to figure out how many
>>> devices/which devices go to which nodes as you scale out to a large number
>>> of device endpoints.
>>>
>>> *Solution:*
>>>
>>> The transport model (as created by Nan Liu) seems to get around the
>>> issues mentioned above and would allow a pretty clean integration path.
>>>
>>> For folks not familiar with the transport model. It uses regular types
>>> and providers that accept a parameter called transport that can be used to
>>> indicate that it should be applied against some remote endpoint.
>>>
>>> For example:
>>>
>>> Transport { 'ssh':
>>> url => some_url
>>> password => 'some_password'
>>> }
>>>
>>> port {
>>> transport => Transport[ssh]
>>> }
>>>
>>> This will work perfectly for my use case.
>>>
>>> *The problem:*
>>>
>>> This is fundamentally incompatible with the device model. I will not be
>>> able to leverage resources implemented using this model, people using the
>>> device model will not be able to leverage resources that I/we write.
>>>
>>
>> I think it's possible to switch between them. I'll use the F5 module as
>> an example which is currently puppet device:
>>
>> We just need check in the following order
>> 1. If facter connection value exist, use facter value to connect (fixes
>> problem for puppet resource inspection).
>>
>
> Yep, facts would totally work, but AFAIK, puppet resource does not invoke
> facter (are you saying that it should)
>
No, this was abusing facter to pass in connection info (don't care about
facts and they won't work anyhow). So normally you can't run puppet
resource with device type resources, and this allowed inspection of
resources (see more issues below):
FACTER_connection=https://user:pass@url/ puppet resource network_device_type
2. If resource has transport parameter use the catalog value.
>> 3. Use the setting in device.conf
>>
>
> This is pretty much what I was thinking.
>
>
>>
>> Change the following code to (much abbreviated):
>>
>> https://github.com/nanliu/puppetlabs-f5/blob/master/lib/puppet/provider/f5.rb#L29-L39
>>
>> @transport ||=
>> Puppet::Util::NetworkDevice::F5::Device.new(Facter.value(:url)).transport
>> If Facter.value(:url)
>> @transport ||= PuppetX::Puppetlabs::Transport.retrieve(:resource_ref =>
>> resource[:transport], :catalog => resource.catalog, :provider => 'f5') if
>> resource[:transport]
>> @transport ||= Puppet::Util::NetworkDevice.current.transport
>>
>> I would feel much more confident in transport if it was possible to still
>>> leverage the logic encoded in Puppet devices. This is impossible b/c
>>> devices label themselves in such a way that means that they can only be
>>> consumed by the puppet device command (while I would use resource, apply,
>>> and agent).
>>>
>>
>> I'm not sure what advantage of puppet device you would like to see in
>> transport.
>>
>
> my main concern here is with duplicated code. Currently, even if you did
> your above recommendations, you still wind up with connection information
> encoded in both the Puppet device as well as the transport. I would like to
> see a reasonable pattern for how they can share code.
>
Let me boil it down this way. If the module is designed correctly,
transport is just retrieving the catalog connection info and initializing
it the same way as the facter code above.
1. Transport search catalog for connectivity info.
2. Invoke Puppet::Util::NetworkDevice::F5::Device.new(catalog_credentials)
So we are just replacing device.conf with the info in the catalog and
preserving everything else. It's not much overhead and no duplication in
code. The problem I have with things like vShield is I need a simultaneous
connection to vCenter for objects moref id, so I can't make it compatible
with puppet device. I could see a way to add transport functionality to
resources implemented via puppet devices. So if there's a puppet device
resource that manages DNS, firewall rules, loadbalancing, I could see some
value adding transport. The next questions is how do you sync the source of
truth in this case? Network device is often shared between many application
and how would you get a complete picture when you start decentralizing
control?
Is this something we could just fix in the device type and providers? To
>>> have them either get their credentials from device.conf or transport
>>> resources? Could we get rid of the code that allows for puppet devices to
>>> only be applied using the puppet device command?
>>>
>>
>> See above. Certainly there are a few bugs such as apply_to_device that
>> needs to be fixed,
>>
>
> I might just be struggling to understand exactly how this works.
> appy_to_device is required for using the credentials from device.conf, and
> means that the resources can't be used elsewhere?
>
There was suppose to be three types of resources, apply_to_device,
apply_to_host(???), apply_to_both(???). apply_to_device resource only works
with puppet device command, apply_to_host only with puppet agent, and
apply_to_both should work with both. I was hoping to toggle resource such
as notify to apply_to_both, but the feature is broken. apply_to_device also
prohibits device type resource to be used in puppet resource, so I actually
kept two sets of resources so I can use them in puppet resource and puppet
device.
HTH,
Nan
--
You received this message because you are subscribed to the Google Groups
"Puppet Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to puppet-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/puppet-users/CACqVBqDGwMY6bUgXAQ34jhs3tYZyLv2UXa0decgo4hUiTjAqiQ%40mail.gmail.com.
For more options, visit https://groups.google.com/groups/opt_out.
