I am using Change Sets for the first time. In the past I have deployed with Eclipse.

Right now, I am attempting to deploy an entire object. I would like to deploy everything involved with the object (fields ; page layouts ; validation rules ; field updates ; workflows ; email alerts etc...)

So, I add this Custom Object to my Change Set (we'll call it "ObjectX_c") but then when I press "View/Add Dependencies" I get a huge list of components which includes components whose Parent Object is not ObjectX_c. At first I thought these were possibly components which referred to ObjectX or that it referred to in some way, but that doesn't appear to be the case.

So, is it up to me to fish through that list and pick only the components which have ObjectX as its Parent Object ?

Thank you very much for your help.

2 Answers 2


Yes, the Dependencies displayed sometimes cast a wider net than you'd like but are generally good for catching out stuff you may have missed.

You will need to pick the fields, layouts etc and make sure that you have a deploy-able change set.

If you realise that you've missed out stuff after deploying to the destination, you can clone the change set and add stuff to it and redeploy.

It is a bit manual so works best for discrete things rather than large scale deployments.

  • Thank you @techtrekker ; I thought I was going crazy because so much was showing up which didn't apply to my change set at all. What do you think causes that ?When that happens it essentially takes away the ability to do a Select All for all of the fields in the object which is being deployed which is a VERY big deal when you're deploying an object with about 100 fields !Besides the log which is produced by Change Sets do you know of any benefit in using it instead of Eclipse?If View/Add Depencies isn't consistant then that isn't one of them which is too bad because Eclipse doesn't have that.
    – Zoom_v
    Commented Oct 25, 2013 at 16:48

Change sets are great, primarily because they're way easier to deal with than Eclipse or ANT. That said, building them can be like pulling teeth, and the dependency analyzer doesn't do much to narrow things down and may not really answer the true question of "what's related to this object".

When you're doing larger deployments, i.e. a whole object and all of it's related stuff, Eclipse and ANT will start to become preferred. As an example, the following package.xml will get you all the object details, fields, record types, field sets, buttons, list views for a particular object. For an object with a non-trivial amount of these building a change set (which requires choosing each individual field, record type, button, and list view) makes watching paint dry seem like a reasonable occupational choice.

<?xml version="1.0" encoding="UTF-8"?>
<Package xmlns="http://soap.sforce.com/2006/04/metadata">
  • yha,if the dependency analyzer worked 100% then I would never use Eclipse to deploy again.But as it is now deploying an object in Eclipse can almost be completed with a single click on the checkbox next to the Custom Object's name.Like you said,it includes all of those component types (along with workflow rules ; email alerts ;field updates).I can deploy a good amount of my objects by simply doing that, adding the page layouts, adding a tab, and then adding any email template I need. Doing that same thing in Change Sets seems to require me to jump around to a bunch of screens and clicks.
    – Zoom_v
    Commented Oct 25, 2013 at 18:29

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .