Numerous model activation errors, are seen in the debug file, after large imports via the Modeling Gateway.

Document ID : KB000006683
Last Modified Date : 14/02/2018
Show Technical Document Details
Issue:

When importing a large quantity of models, via the Modeling Gateway, we get a large number of failures, due model inactivation, similar to the following:

 

Number of device models created: 292

Identifying device ports...

Importing Schedule section.

Resolving 457 connections...

Number of errors in Topology section: 457

Number of errors in Auto_Discovery section: 3

Creating 1 associations in Schedule section.

Number of errors in Schedule section: 1

Updating 1835 models...

Number of errors in Update section: 1554

Number of models created:         590

Number of associations created:   363

Number of total errors:           2015

Modeling Gateway import finished with error.

 

Environment:
Any large Spectrum environment.
Cause:

The normal process of importing via the Modeling Gateway, involves importing the models, activating them and then importing the child models that depend on the activated models. e.g. interfaces and global collections etc.

However, when importing a large number of models, due to the limited number of threads, it's probable that not all will have activated, by the time the import process attempts model the child models.  This prevents the parent device model from being found, causing the modeling of its child models to fail.

This can be seen in the Modeling Gateway debug as follows:

 

Model creation complete, waiting for activation...  

The import process will continue if no new models activate within 9 minutes.  20170522.12.48.02

The import process will continue if no new models activate within 8 minutes.  20170522.12.49.02

The import process will continue if no new models activate within 7 minutes.  20170522.12.50.02

The import process will continue if no new models activate within 6 minutes.  20170522.12.51.02

The import process will continue if no new models activate within 5 minutes.  20170522.12.52.02

The import process will continue if no new models activate within 4 minutes.  20170522.12.53.02

The import process will continue if no new models activate within 3 minutes.  20170522.12.54.02

The import process will continue if no new models activate within 2 minutes.  20170522.12.55.02

The import process will continue if no new models activate within 1 minutes.  20170522.12.56.02

The import process will continue if no new models activate within 0 minutes.  20170522.12.57.03

Get model info for model x.x.x.x

device x.x.x.x model_handle=0xxxxxxx|model_name=xxxxxx|network_address=x.x.x.x was not active.  

Get model info for model x.x.x.x

device x.x.x.x model_handle=0xxxxxxx|model_name=xxxxxx|network_address=x.x.x.x was not active.  

Get model info for model x.x.x.x

device x.x.x.x model_handle=0xxxxxxx|model_name=xxxxxx|network_address=x.x.x.x was not active.  

 

Resolution:

There is an attribute for model activation timeout, that can be used to extend the model activation time.  

This should be added to the Import tag of the xml import file, at root node as follows, where we increased the timeout to 50:  

<SPECTRUM_Export landscape_name=<xxxxxxx> landscape_handle="0xXXX00000" model_activation_time="50">

 

Please be aware that this is referenced in the DTD, but should not be edited there.

Additional Information:

https://docops.ca.com/ca-spectrum/10-2-1/en/customizing/modeling-gateway-toolkit