Best practices in Tekla Model Sharing
To keep your shared models in good shape and to share your changes successfully, follow the Tekla Model Sharing best practices below.
The users of the same shared model need to have the same Tekla Structures version, and use the same latest service pack.
For general Tekla Model Sharing troubleshooting instructions, see Troubleshooting Tekla Model Sharing.
Install the Cache service for Tekla Model Sharing
Tekla Model Sharing Cache service downloads model data from the Tekla Model Sharing service and caches the data in the file system inside a LAN. Using the cache service reduces the use of the Internet, resulting in faster download times when the same data is requested more than once. Note that the cache is not used for packets that are written out.
The cache service needs an Internet connection to download model data from the sharing service. The first time a user fetches a packet it will be loaded to the cache service. Any following requests to the same packet will then be served fast from the cache service inside the LAN.
The cache service is useful even if there is only one Tekla Model Sharing user in the same office. For example, rejoining a model is faster as the model data is available in the cache service. In addition, the model data is always loaded as small data blocks. This is useful if the downloading is interrupted for some reason, because the cache service can just download the missing blocks later. The Cache service content can also be cleared at any point by simply deleting the unnecessary files from the service. This way, you can easily remove old model data.
If there are several Tekla Model Sharing users in the same office, we highly recommend you to install the Tekla Model Sharing Cache service. The Cache service is recommended especially in regions where the download speed may be limited.
The image below shows how the model data is stored to the sharing service and used with the Tekla Model Sharing Cache service.
If the Cache service cannot be reached for any reason, Tekla Structures will use the cloud storage for model changes directly. This can also happen when a download operation is on-going, if the connection to Cache service times out.
You can see whether the Cache service is in use by checking the ClientLog_cat.txtClientLog_dog.txt log files in the \Users\<user>\AppData\Local\Tekla DataSharing folder.
Software and system requirements for a cache installation:
- Windows Server 2008 R2, or later
- .NET Framework 4.5.1
Install a cache service in the following way:
Troubleshoot the cache service installation
If you cannot connect to the service from Tekla Structures:
-
Ensure that the Tekla Model Sharing Windows Service is running.
-
Ensure sure that the firewalls do not block TCP/IP ports configured to Tekla Structures, for example
9001
or9998
when you use the default ports.
If the service does not start:
-
Check the Windows Event Viewer's Application Log for possible errors.
Use GUIDs correctly in shared models
Tekla Structures objects have an identifier that is shown as an object GUID, Globally Unique Identifier, that is also used in Tekla Model Sharing.
This means that features that do not use GUIDs need to be changed to use GUIDs:
-
Interoperability import/export actions:
-
FabTrol XML
-
ASCII
-
-
All other applications, macros and report processes that rely on static IDs.
Save local versions of shared models on your computer
We recommend that you save the local versions of shared models on your computer instead of a network drive for two main reasons:
-
The performance of shared models is better when the local models are saved on your own computer. This means that the models open faster, for example.
-
Saving shared models on your own computer prevents other users from accessing them and accidentally blocking important files.
If you still want to save local models on a network drive, use a private network drive that other users cannot access.
Create baselines regularly
The Owner of a model should create baselines of the model regularly. For example, you can create a baseline once a week.
We recommend creating a new baseline each time a new user has been invited to the model. This way, joining the shared model is faster.
Back up shared models
We recommend you to back up the models used in Tekla Model Sharing. In case there are problems with a shared model, it is possible to select any user's local version of the model, or a model that has been backed up, and continue working using that model. Make sure that you have the complete backed up model in use and that the model folder includes, for example, drawings and different databases. This ensures that the model functions properly and you do not lose any data. If the backed up version of the model is old, reading in all the changes may take some time.
Back up your models according to your company conventions, for example, by using Windows Backup. You can also use the
command to create a backup copy of the model. The backup copy will have the same GUIDs as the original model.Note that the Save as command cannot be used for backing up the model. If you use Save as, the model gets new IDs and it has no relation to the original model. If you use the Save as command, the model history is not copied with the saved model.
Number model objects in Tekla Model Sharing
Numbering a shared model has three main phases: reading in the changes made by other users, numbering a series of objects, and writing out the numbering changes. Always use the Number series of selected objects command when you number parts in a shared model. To avoid unnecessary work and conflicts, do not use the Number modified objects command.
Before you start numbering objects in a shared model, you should plan the numbering carefully. We recommend that you divide the model into phases, and each user only numbers objects within the phase they are working on. This way, you can avoid numbering conflicts in your shared models.
To number a phase, do the following:
Use XS_FIRM and XS_PROJECT folders with shared models
Using project folders and firm folders in shared models is the easiest when you use a sub-folder in a Trimble Connect project as the project or firm folder.
The project and firm folder information is only updated from the Trimble Connect project to local versions of the shared models, not vice versa. In practice, this means that Tekla Structures downloads new files from the project or film folder to the local model and updates any changed files. If a local file is not in the Trimble Connect project sub-folder, it is removed from the local model.
When Tekla Structures detects a change between the local model data and the data in the Trimble Connect project or firm folder, you get the following notification:
If you click Yes, the model is saved and closed. Then, the local model data is updated and re-opened.
To set a folder in a Trimble Connect project as the XS_FIRM
or the XS_PROJECT
folder: