[jboss-as7-dev] add operation in CLI GUI

ssilvert at redhat.com ssilvert at redhat.com
Thu Jan 12 14:39:46 EST 2012 

On 1/12/2012 1:09 PM, Kabir Khan wrote:
>> On 1/12/2012 11:39 AM, Alexey Loubyansky wrote:
>>> I could suggest two ways to workaround this:
>>>
>>> - node type as a 'folder', currently the unit is node-type=node-name, 
>>> instead node-type could be a folder that contained node-names and you 
>>> could click on the node-type to add new nodes.
>> Early on I was going to do it this way, but it became clear that it
>> would be overly complicated and would break with the directory notion in
>> the CLI where a directory is /subsystem=jca/, not /subsystem/jca/.  Once
>> you replace = with / you lose sight of the underlying address scheme.
> Could the structure perhaps be 
> subsystem/
>   subsystem=jca
>   subsystem=jmx
> Or is that too weird?
I just tried that and it did indeed get weird.  You end up with nodes
that don't have any correspondence to the address path.  So you have to
code around it all over the place.  It also made any sort of refresh
impossible to implement without significant hacks.  I do need to improve
refresh at some point and I don't want to make that any harder than it
already is.

Try my three latest commits found here and let me know what you think:
https://github.com/ssilvert/jboss-as/commits/cli-gui-fixadd

Alexey, once you see it, let me know if you still think that "?" makes
more sense than "*".  That's an easy change.

>>> - add node-type=? and use it for add and non-instance specific 
>>> operations. 'add' will still be listed if you right-click on a specific 
>>> instance, though (the same true about the command line actually). We 
>>> could filter those, although that would be covering the weirdness of the 
>>> management model.
>> I like that idea.  I'll give it a try.  But I'm thinking the folder will
>> be "node-type=*" instead of "node-type=?".
>>
>> WDYT?
> That sounds fine although that should only really be done for things with the add on a wildcard. Also, we haven't solved the problem of things that have expected children, e.g. the 
> /subsystem=jmx/connector=jmx where jmx is the ONLY name allowed for the connector. There are a few other places in the model that have this same setup as well.  
>
>>> Alexey
>>>
>>> On 01/12/2012 04:33 PM, Kabir Khan wrote:
>>>> Alexey,
>>>>
>>>> Can you provide a bit more input on the discussion on https://github.com/jbossas/jboss-as/pull/1081 but on this thread?
>>>> On 11 Jan 2012, at 18:08, ssilvert at redhat.com wrote:
>>>>
>>>>> I believe that add operations are now done correctly as far as the tool
>>>>> is concerned.
>>>>> https://github.com/jbossas/jboss-as/pull/1081
>>>>>
>>>>> We needed to prompt for a resource name that gets filled into the
>>>>> address path.   So for:
>>>>> /subsystem=logging/logger=foo/:add
>>>>>
>>>>> The gui asks the user to fill in the resource name "foo".  The add
>>>>> operation should not have "name" as one of its declared parameters
>>>>> because that would be redundant.  (The logging subsystem is doing this
>>>>> correctly, btw)
>>>>>
>>>>> Stan
>>>>>
>>>>> On 1/10/2012 3:56 PM, Kabir Khan wrote:
>>>>>> On 10 Jan 2012, at 20:51, Kabir Khan wrote:
>>>>>>
>>>>>>> On 10 Jan 2012, at 18:29, ssilvert at redhat.com wrote:
>>>>>>>
>>>>>>>> On 1/10/2012 11:30 AM, Kabir Khan wrote:
>>>>>>>>> That worked for me so I pushed that change. It looks great!
>>>>>>>>>
>>>>>>>>> The one thing which is a bit strange is the ability to add a child resource. e.g. the add operation should really by on the parent of the resource being targeted. For example, the ee extension has the add operation in its list, which gives the op: "/extension=org.jboss.as.ee/:add(module=sasasa)". Which is a bit pointless since it already exists, and there is no way to add a new extension. I had similar issues when writing the management jmx facade, see ChildAddOperationFinder and its callers.
>>>>>>>> Yea, I agree that it's a bit strange.  But I'm not sure if this is a
>>>>>>>> problem that can be solved by the tool.  What I have seen is that often
>>>>>>>> you need to give an address that doesn't exist yet, and append :add to
>>>>>>>> it.  For instance, to add a new logger you say
>>>>>>>>
>>>>>>>> /subsystem=logging/logger=foo/:add
>>>>>>>>
>>>>>>>> But if you look at subsystem=logging you see that there are many kinds
>>>>>>>> of entities that might need to be created (loggers, handlers, etc.).
>>>>>>>> That makes it hard for the tool to guess what you are trying to add.
>>>>>>> I got around this in the jmx facade. Connect to the running process with jconsole, and go to the jboss.as:management=server mbean (appears as simply 'server' in the list. This corresponds to the root resource. In the operations it has an addXXX operation for each child type. e.g.
>>>>>>>
>>>>>>> addExtension ->  /extension=*:add
>>>>>>> addSubsystemEjb3 ->  /subsystem=ejb3:add
>>>>>>> addSystemProperty ->  /system-property=*:add
>>>>>>> addSubsystemJaxrs ->  /subsystem=jaxrs:add
>>>>>> Or to use your logging subsystem example, go to jboss.as:subsystem=logging (appears as just 'logging' in jconsole), and you will see addSizeRotatingFileHandler, addPeriodicRotatingFieldHandler, addLogger etc.
>>>>>>
>>>>>>>> Beyond that, there is no way for the tool to know what params are
>>>>>>>> required for a new entity.  :read-operation-description(name=add)
>>>>>>>> doesn't usually help in this case.
>>>>>>>>
>>>>>>>> IMO, each management handler should fully implement the add operation,
>>>>>>>> specify the expected params, and never rely on default behavior.  But
>>>>>>>> that would be a lot of work to go back and fix them all.
>>>>>>> I'm not 100% sure what you mean here so I might be misunderstanding what you're saying. The add handlers are different for different places e.g.:
>>>>>>>
>>>>>>> /system-property=*:read-operation-description(name=add)
>>>>>>> /interface=*:read-operation-description(name=add)
>>>>>>> /subsystem=jmx:read-operation-description(name=add)
>>>>>>> /subsystem=ee:read-operation-description(name=add)
>>>>>>>
>>>>>>> all return different descriptions. (Some have an add operation for '*' meaning all children take the same parameters, others one for each child like the subsystems)
>>>>>>>
>>>>>>>> Anyone else have comments/suggestions?
>>>>>>>>
>>>>>>>>> On 10 Jan 2012, at 16:16, Kabir Khan wrote:
>>>>>>>>>
>>>>>>>>>> I've been out so I didn't see this until my previous reply, I'll try it now
>>>>>>>>>>
>>>>>>>>>> On 10 Jan 2012, at 14:11, ssilvert at redhat.com wrote:
>>>>>>>>>>
>>>>>>>>>>> Kabir (or someone with OS X),
>>>>>>>>>>>
>>>>>>>>>>> Can you give this a try?
>>>>>>>>>>>
>>>>>>>>>>> https://github.com/ssilvert/jboss-as/commit/9f90081c86c54278ec3be444f9fb1906051653e8
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> On 1/10/2012 8:28 AM, ssilvert at redhat.com wrote:
>>>>>>>>>>>> On 1/10/2012 8:06 AM, Kabir Khan wrote:
>>>>>>>>>>>>> This has been merged. However, I've not been able to play with it. I added https://github.com/jbossas/jboss-as/commit/999beb90268d94e4e6547986b7277f166b1c2f69 to get around this problem
>>>>>>>>>>>> Thanks.  I don't have an OS X box, so I don't have any way to test.
>>>>>>>>>>>>> The next thing is that right-clicking does not seem to work on OS X so I don't see the menu
>>>>>>>>>>>> Looks like this is a well known issue with OS X and the Swing
>>>>>>>>>>>> isPopupTrigger() method.  What I'm reading says that you need to hold
>>>>>>>>>>>> down the control key while pressing button one.
>>>>>>>>>>>> http://stackoverflow.com/questions/5736872/java-popup-trigger-in-linux
>>>>>>>>>>>>
>>>>>>>>>>>> I'll see if there is a workaround.  I'll probably also add a regular
>>>>>>>>>>>> menu at the top so we don't have to rely on right-click.
>>>>>>>>>>>>> On 10 Jan 2012, at 12:20, ssilvert at redhat.com wrote:
>>>>>>>>>>>>>
>>>>>>>>>>>>>> On 1/10/2012 4:47 AM, Darran Lofthouse wrote:
>>>>>>>>>>>>>>> Do we want to verify the various security features already in the CLI
>>>>>>>>>>>>>>> are supported before it is pulled or follow up once it is in?
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>> Regards,
>>>>>>>>>>>>>>> Darran Lofthouse.
>>>>>>>>>>>>>> The GUI is built on top of the CLI.  So it's running the same
>>>>>>>>>>>>>> authentication/authorization.  The only difference is that it brings up
>>>>>>>>>>>>>> a GUI instead of a command line.
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> BTW, I have to give props to Alexey on this.  His CLI code is extremely
>>>>>>>>>>>>>> well organized and easy to extend.
>>>>>>>>>>>>>>> On 01/09/2012 09:42 PM, Jason T. Greene wrote:
>>>>>>>>>>>>>>>> You have to merge his pull to get it
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>> On 1/9/12 3:41 PM, Kabir Khan wrote:
>>>>>>>>>>>>>>>>> Does it need to be enabled somehow?
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>> [~/sourcecontrol/jboss-as7/git/jboss-as]
>>>>>>>>>>>>>>>>> $./build/target/jboss-as-7.1.0.Final-SNAPSHOT/bin/jboss-admin.sh --gui --connect
>>>>>>>>>>>>>>>>> '--gui' is not a valid operation name.
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>> [~/sourcecontrol/jboss-as7/git/jboss-as]
>>>>>>>>>>>>>>>>> $./build/target/jboss-as-7.1.0.Final-SNAPSHOT/bin/jboss-admin.sh --connect --gui
>>>>>>>>>>>>>>>>> '--gui' is not a valid operation name.
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>> On 9 Jan 2012, at 19:33, Jason T. Greene wrote:
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>> Just saw Stan did this:
>>>>>>>>>>>>>>>>>> http://community.jboss.org/docs/DOC-17457
>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>> I haven't played with it yet, but it looks awesome!
>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>> IMO this is exactly what we need for an "advanced" mode in the console
>>>>>>>>>>>>>>>>>> (post 7.1 obviously)
>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>> --
>>>>>>>>>>>>>>>>>> Jason T. Greene
>>>>>>>>>>>>>>>>>> JBoss AS Lead / EAP Platform Architect
>>>>>>>>>>>>>>>>>> JBoss, a division of Red Hat
>>>>>>>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>>>>>>>> jboss-as7-dev mailing list
>>>>>>>>>>>>>>>>>> jboss-as7-dev at lists.jboss.org
>>>>>>>>>>>>>>>>>> https://lists.jboss.org/mailman/listinfo/jboss-as7-dev
>>>>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>>>>> jboss-as7-dev mailing list
>>>>>>>>>>>>>>> jboss-as7-dev at lists.jboss.org
>>>>>>>>>>>>>>> https://lists.jboss.org/mailman/listinfo/jboss-as7-dev
>>>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>>>> jboss-as7-dev mailing list
>>>>>>>>>>>>>> jboss-as7-dev at lists.jboss.org
>>>>>>>>>>>>>> https://lists.jboss.org/mailman/listinfo/jboss-as7-dev
>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>> jboss-as7-dev mailing list
>>>>>>>>>>>> jboss-as7-dev at lists.jboss.org
>>>>>>>>>>>> https://lists.jboss.org/mailman/listinfo/jboss-as7-dev
>>>>>>>>>>> _______________________________________________
>>>>>>>>>>> jboss-as7-dev mailing list
>>>>>>>>>>> jboss-as7-dev at lists.jboss.org
>>>>>>>>>>> https://lists.jboss.org/mailman/listinfo/jboss-as7-dev
>>>>>>>>>> _______________________________________________
>>>>>>>>>> jboss-as7-dev mailing list
>>>>>>>>>> jboss-as7-dev at lists.jboss.org
>>>>>>>>>> https://lists.jboss.org/mailman/listinfo/jboss-as7-dev
>>>>>>>> _______________________________________________
>>>>>>>> jboss-as7-dev mailing list
>>>>>>>> jboss-as7-dev at lists.jboss.org
>>>>>>>> https://lists.jboss.org/mailman/listinfo/jboss-as7-dev
>>>>> _______________________________________________
>>>>> jboss-as7-dev mailing list
>>>>> jboss-as7-dev at lists.jboss.org
>>>>> https://lists.jboss.org/mailman/listinfo/jboss-as7-dev
>>> _______________________________________________
>>> jboss-as7-dev mailing list
>>> jboss-as7-dev at lists.jboss.org
>>> https://lists.jboss.org/mailman/listinfo/jboss-as7-dev
>> _______________________________________________
>> jboss-as7-dev mailing list
>> jboss-as7-dev at lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/jboss-as7-dev

More information about the jboss-as7-dev mailing list