Halil, I am not sure I understand the question. There is also too few information in your post to answer. You can create Baselines and Snapshots on Streams or on repository workspaces.

If you create a snapshot or baseline at a repository workspace, all change sets that are outgoing for each component will be completed and included in the baseline (and snapshot, because that creates a baseline if necessary). It is basically creating a baseline of the state of our repository workspace on the server as it is right now.

If you have unresolved changes you get a warning. If you continue to create the baseline the unresolved changes are not part of it.

I don't know, can this scenario happen or if so frequency?

I will write more to be clear and simple..

I've a  Stream "StrA" which includes "CompA" and "CompB". As a developer I've also created repository workspace following StrA and loaded both components to my local. I developed some code and checked in. Then In my repo workspace, I've outgoing change sets for CompA waiting to go..

At that time I click to create snapshot for strA. Let's say It takes long creating snapshot process (I mean clicking to create to finishing all baselines takes 3 min.). When creating baseline process continues I've decided to deliver the outgoing change set to the stream's CompA.

So the question is:

* What happens at that time? For example,  Deliver process waits the streams snapshot creation process than deliver starts. or snapshot process stops, deliver occurs than snapshot creation restarts..

Our customer ask this because they want to automate the baseline creation process and want to mitigate the risk for conflicting deliver and baseline creation process and wants to do a synchronization.

Hi Halil,

I would suggest to use a repository workspace to create the baselines and snapshots. In this case you are in total control and there is no racing condition that can happen. This is as far as I understand done for example by the build engine. While the baseline/snapshot is created, the state of your repository workspace represents what will be included in the baseline. It can not change during the baseline operation. The isolation the repository workspace provides makes sure nothing can happen to the files in the baseline. This can also be automated using the SCM tools.

For streams, I would assume that changes that occur during the snapshot/baseline process can't get into the baseline. However I think you can't be sure what the stream contains when you create the baseline.