Contact   •   Articles   •   Products   •   Search

Rick Strahl's Web Log

Wind, waves, code and everything in between...
ASP.NET • C# • HTML5 • JavaScript • AngularJs

LINQ to SQL and Serialization

One of the important that things that you will likely have to do at one point or another with entity objects is serialize them. This could be for Web Services most commonly but could be for any sort of storage or transport scenario.

There are a number of pitfalls with this scenario however, primarily because LINQ to SQL will very likely generate circular references into your entity model from your data and XML Serialization will fail outright in that scenario. For example, say you have a customer and projects table and if you let LINQ generate the one to many relationship it will create Customer entity with a Projects property and a Project entity with a Customer property.

For code scenarios this is probably a good thing - you want to be able to see all Projects and filter that list in your code.

Unfortunately in a serialization scenario this doesn't work because you essentially have a circular reference - Customer ->Projects -> Customer. So we have a 1 -> Many and a 1 - 1 relationship going back to the original object here. To demonstrate you can actually do whacky stuff like this:

CustomerEntity entity = Customer.Context.CustomerEntities.Single(c => c.Pk == 1);
string address = entity.Projects.ToArray()[0].Customer.Address;

And while that's a whacky example, this sort of setup makes sense. When you look at an Project you will liklely want to know about the Customer that is associated with the Project. And when you're looking at a Customer you'll want to know about all the Projects associated with the customer. A classic circular reference scenario.

At an abstract level that's fine but when you want to serialize this arrangement there are problems because the serializer doesn't know what do to with the circular references. So if you run code like this:

CustomerEntity entity = Customer.Context.CustomerEntities.Single(c => c.Pk == 1);
MemoryStream ms = new MemoryStream();
XmlTextWriter writer = new XmlTextWriter(ms, new UTF8Encoding());
XmlSerializer serializer =
        new XmlSerializer( typeof( CustomerEntity ) );
writer.Formatting = Formatting.Indented;
writer.IndentChar = ' ';
writer.Indentation = 3;
serializer.Serialize(writer, entity);
byte[] Result = new byte[ms.Length];
ms.Position = 0;
ms.Read(Result, 0, (int)ms.Length);
string XmlResultString = Encoding.UTF8.GetString(Result, 0, (int)ms.Length);

You'll get:

A circular reference was detected while serializing an object of type TimeTrakker.CustomerEntity.

There are a couple of ways to get around this but they're not exactly pretty as they involving changing the model.

So when you think about Customers and Projects relationship for example, is it really appropriate to have a Projects property in your CustomerEntity? By itself this expression doesn't really represent anything useful, especially if the system grows and there are lots of projects. So for serialization purposes you really wouldn't actually like to see this relationship.

So you can actually hide this relationship for serialization by marking the child property as internal. On the other the ProjectEntity probably should have a CustomerEntity and that always should be visible because it's a 1-1 relationship and you'll likely need that data and it would likely be OK to serialize so that relationship stays public:

You can also set the ChildProperty  option to false altogether to completely cut off the relationship through the entity. Using Internal still makes the relationship available in the middle tier (if you use one that is), but hides it for XmlSerialization (which works off public properties only).

As I said this is not a great workaround because by removing this relationship you're also removing the ability to run entity relation ship queries through LINQ. So by doing the above this sort of thing will no longer work:

CustomerEntity entity = Customer.Context.CustomerEntities.Single(c => c.Pk == 1);            
var query2 = from c in Customer.Context.CustomerEntities  
             from p in c.Projects
             where p.CustomerPk == 1
             select new { c.Company, p.ProjectName};
this.gdList.DataSource = query2;

Remove the relationship or set it to internal (outside of the bus object) you loose the ability to the implicit relationship that is implied and you'd have to explicitly define it like this:

var query2 = from c in Customer.Context.CustomerEntities  
             join p in Customer.Context.ProjectEntities on c.Pk equals p.CustomerPk
             select new { c.Company, p.ProjectName};
this.gdList.DataSource = query2;

So the above workaround of turning off the relationship mapping might get you by in SOME scenarios. The internal flagging allows you to hack around the problem by making the property essentially invisible for XmlSerialization. However, that's not going to work for Binary Serialization and is also going to be a problem for WCF based Web Services that don't use the basicHttpBinding - all other formats use the lower level variations of binary serialization that work of field state rather than property values.

There's a solution for WCF however -  the Entity designer has a Serialization option for the generating a WCF [DataContract] and [DataMembers] on the entities generated:

This option controls only whether the entities are generated with a WCF [DataContract] attribute in the classes:

public partial class CustomerEntity : INotifyPropertyChanging, INotifyPropertyChanged

and for each property:

[Column(Storage="_LastName", DbType="NVarChar(50)", UpdateCheck=UpdateCheck.Never)]
public string LastName

If you specify a non-public flag (Internal, Protected or Private) for the relationship (or any field for that matter), the DataMember property is not generated. This gives you pretty good control over WCF serialization as well as XML Serialization and ASMX Web Services.

For Binary Serialization that's not going to do anything however - but then again there shouldn't be a lot of need for binary serialization anymore with WCF filling that niche in most places.

.Serialization - a Special Scenario

All of this also got me to thinking about these object relationships. In code having an Projects property is useful because you can apply LINQ filter expressions against it. But when you persist via serialization you don't get that flexibility. So if you were to serialize you would get ALL of the child entities serialized which is probably not at all what you want. For example, in my simple time tracking app I'm using as my playground I have a customer entity with individual entries. I really don't want to serialize ALL entries. I MIGHT want to serialize a handful of new entries, but certainly not ALL of them once the application has been running for a few months.

The problem now is that we have a logical AND a physical relationship in this scenario. As non parameterized view the Projects property doesn't make any sense in a Customer object. However,  a Entries property on an Invoice object might make perfect sense. In some cases it makes perfect sense to expose the relationship directly.But if you decide not to express the relationship explicitly between say Customer and Projects you loose some of the cool functionality that LINQ provides by automatically understanding relationships.

I'm not sure there's a solution to this. I think this same problem also exists with the ADO.NET Entity Framework (which uses a very similar approach although it's even more implicit about how relationships are defined by explicitly removing key fields).


Maybe some of the folks who've been doing OR/M development for some time with other tools like nHibernate can chime in. This is a thorny problem.

It seems to me that LINQ to SQL could actually solve these issues quite easily by providing a few more properties to set on entities and relationships. Such as a Serialized property on relations, and explicit serialization attributes on the entities. But even as it is you can bend LINQ to SQL to do as you need it seems.

Make Donation
Posted in LINQ  Visual Studio   ADO.NET  

The Voices of Reason


September 03, 2007

# re: LINQ to SQL and Serialization

Although this is a oversimplified, the relationship definitions need to have something to indicate composition/constituency - e.g., that B is part of A. When serializing, I often want to serialize A and everything it it references that constitutes it. I may optionally want serialize things it references that are non-constituent, but I want constrol based on how strongly or loosely related the two are.

Rick Strahl
September 03, 2007

# re: LINQ to SQL and Serialization

I hear 'ya but the question becomes how do you map that? You can set up relationships manually so if an on-again, off-again scenario using manual relations will work. I guess the issue is that we really want two different behaviors based on whether we're navigating through the data in code or whether we are serializing.

It's almost like a second set of configurations for the relations would be needed.

I haven't thought this all the way through - I need to play with live data to see - but I think for now I'd rather stick with 'loose' relations that aren't visible as relations on the object (ie. Customers -> Projects) because in reality you'd rarely need to see Customers.Projects in an object graph. It's just about as easy to set up the relation or directly query projects for whatever operations you'd perfom on projects:

var q = context.CustomerEntities.All();

foreach(CustomerEntity cust in q)
     foreach(Project p in context.ProjectEntities.Where( p.CustomerPk == cust.Pk)
          // ... do whatever with projects

This code isn't really any less workable than if there was a relation - it's basically just an extra .Where() to enforce the relationship here.

It matters only for persistence; if you pass the object somewhere or if you serialize. Again I'm not sure if that's a realistic - or even desirable - object structure that you'd want to pass around because once there are a few projects, that relationship without a filter is pretty much meaningless for persistence or object passing.

September 03, 2007

# re: LINQ to SQL and Serialization

I've been serializing object graphs with NH objects - but pretty much in the way you work around the problem.

We use Castle ActiveRecord attributes to declare only the relationships we're interested in having in our model code. To be honest, I don't see this as a drawback, since in our case it has always been clearly one side that owns the relationship (if not, we revisited our model), and so we put our declarations on the side that owns it.

I'd love a counter-example for which this does not work well though, if you have one though.

I also prefer this approach since it forces one to think hard about how you will be using your entity objects and when they will be loaded. Having everything automagic is nice and RAD until you realize you're accidentally eager loading from a table with 10 million rows :)

(Culprits shall remain unnamed).

Ryan Haney
September 03, 2007

# re: LINQ to SQL and Serialization

If you make a call to a web method called GetCustomerWithProjects(), you should get a CustomerWithProjects object. I haven't worked with it much, but I think this type of scenario is where ADO.NET Entities comes to play, possibly with inheritance. Imagine a SerializableCustomer class that inherits from Customer but provides potentially overriding meta-data specifically for XML or binary serialization. This way we can create different instances of entities for different purposes. Something to chew on at least.

Rick Strahl
September 03, 2007

# re: LINQ to SQL and Serialization

>> I'd love a counter-example for which this does not work well though, if you have one though.

Well I'd think that would be somewhat common. Invoice -> LineItems -> Invoice. A lineitem might need information about the invoice. OTOH, the logical business object probably manages the invoice which logically also deals with the lineitems.

Not sure offhand but I suppose I am just brainstorming and trying to think ahead a bit for scenarios that might affect later dead ends.

@Ryan - yeah I suppose you can map additional objects that represent just the right relationship structure, but I think you'd want to avoid that as much as possible in order to minmize pollution your entity model with redundant types. But it may be the only way to do this. I think I'm with nexus on this though - being more conservative and using a minimalist approach with relations for the base model seems a prudent approach. With LINQ to SQL at least you do have the Internal option which precludes the relation from showing up during serialization but can still be used in the middle tier which potentially gives you the best of both worlds.

September 20, 2007

# re: LINQ to SQL and Serialization

I had similar serialization problem with related entities(vs2008 beta2). I was exposing a webservice based on a linq query and i ve come accross this problem. I found a rough way to solve my problem. Perhaps it helps/gives an idea to somebody else:
My table hierarchy is Photos->Tags

and when i serialized my photo object , i took:

System.InvalidOperationException: There was an error generating the XML document. ---> System.InvalidOperationException: The type System.Collections.Generic.List`1[[Photo, App_Code.irdi89y2, Version=, Culture=neutral, PublicKeyToken=null]] may not be used in this context.

I ve overcome this problem by using an partial class implementation.

public partial class Photo
    public Photo EmptifyTags()
        this.Tags= null;
        return this;

 PhotoDBDataContext photoDB = new PhotoDBDataContext();
            var templates = (from photo in photoDB.Photos
                         where photo.IsTemplate.Equals(true)
                         select photo.EmptifyTags()).Skip(start-1).Take(numVisible*2);

October 26, 2007

# re: LINQ to SQL and Serialization

I use the CodeSmith tool and change the LinqToSql templeate to add a [System.Xml.Serialization.XmlIgnore] attribute on every public EntitySet<T> properties. It works.

Mark Lee
December 21, 2007

# re: LINQ to SQL and Serialization

There is an very good article on to address the serialization issue for Linq to Sql.

Isaac Eckert
February 26, 2008

# re: LINQ to SQL and Serialization

I stumbled across a possible solution while trying to get your exception in my own code where I am developing a web service. My first attempt tried to return an entity object that I initialized myself instead of querying the database. The result was not exception and, after some further testing, I found out why. When I initialized the entity myself (
CustomerEntity entity = new CustomerEntity{property1="test", ...}</code&gt) the properties within the entity representing the relational information like Projects in your example are not initialized and are null.

By setting the relational properties in the entity to null before returning the entity object in a webservice (or using XmlSerialer) the entity serializes just fine.  In your example if Projects is the only relationship to customers defined in your Linq to SQL classes you can make customers serialize like this:

<code lang="c#">CustomerEntity entity = Customer.Context.CustomerEntities.Single(c => c.Pk == 1);

     entity.Projects = null;

Now, the only thing is that the Projects property is still serialized as an empty tag. Something kind of like this:

<Projects />

If this is not a problem then setting Projects to null is a fast solution. For my entity classes I added a member function called MakeSerializable to my partial class that sets all the properties to null and I just call that before returning the object.

Ian Cooper [MVP]
April 06, 2008

# Architecting LINQ To SQL Applications, part 8

Previously: Architecting LINQ To SQL Applications, part 7 Tiers A layer, such as we discussed Part 2

June 17, 2008

# re: LINQ to SQL and Serialization

Hi Rick,

Thanks for all the articles on these topics they've been very helpful.

I tried this using a WCF service with BasicHTTPBinding. I managed to get the Parent Property to appear on the client side but it is always null. I have checked while debugging that the property is populated prior to passing up from the service so the contents is getting lost on the way.
Did you manage to get your example actually working with WCF.


Rick Strahl
June 18, 2008

# re: LINQ to SQL and Serialization

@Justin - what are you passing back exactly? You need to make sure that you are actually passing back a non-null value. You should also pass back not an IQueryable, but a concrete instance - ie. a ToList() or ToArray() result.

September 15, 2008

# re: LINQ to SQL and Serialization

Thanks a lot man,

I have applied your article in my project and it worked superb.

Bill Konrad
October 26, 2008

# re: LINQ to SQL and Serialization

Just a thought for everyone. In the case where you are trying to serialize one of these circularly linked entities for the purposes of a webmethod, try this:

select new
ID = p.ID,
Title = p.Title,
Subcategory = p.Subcategory,
Summary = p.Summary

The point is, how often in a webmethod are you really just interested in maybe a field or two from one of the object's circular references? I eliminated my issue by simply grabbing what I needed, creating a new generic type and serializing that instead. You can keep the ID if you need to reference it later.

Maybe I'm off base here, thoughts?

February 19, 2009

# re: LINQ to SQL and Serialization

Thanks for the info

The real problem, I think, is actually a combination of the basic problems with XML [it's a tree, not a graph] and the way the XmlSerializer works [which is to serialise objects on the presumption that they're a tree]. It would actually be possible to do this OK by writing out entity ref's and simply referring to these later on in the XML (which is why entity ref's were added to XML), and by doing this it's possible to serialise any arbitrary graph. Unfortunately, XmlSerialiser fails to do so. Just to demonstrate that it's perfectly possible, try this:

class A
private A _sibling;
public A Sibling
get { return _sibling; }
set { _sibling = value; }

static void Main(string[] args)
A a = new A();

a.Sibling = a; // now that really *is* circular!

SoapFormatter sf = new SoapFormatter();
using (MemoryStream ms = new MemoryStream())
sf.Serialize(ms, a);
Console.WriteLine(Encoding.Default.GetString(ms.GetBuffer(), 0, (int)ms.Length));
Console.WriteLine("Hit the any key...");

Note that the SoapFormatter is essentially useless now, though, because it doesn't support generics!

March 13, 2009

# re: LINQ to SQL and Serialization

Signum Framewor has an ORM with full Linq provider and it's designed with entities serialization in mind, enabling N-layer architectures.

You can see some videos explaning how to make a client - server application here:

Finally, in order to make the resrialization we use NetDataContractSerializer, that enables old remoting-style serialization over WCF services. This way it uses SharedTypes between client and server, enabling validation and change tracking in both sides. It also enables sending full objects graphs (not just trees).

Hope you find the framework as usefull as we do.

Disclaimer: I'm the lead developer of Signum Framework, but nothing I've said is an exageration :)

April 02, 2009

# re: LINQ to SQL and Serialization

FWIW i shonked my way around this problem with the XmlSerializer using the XmlAttributeOverrides to tell the serializer to ignore foreign key properties:

public static void AddIgnoreForLinqForeignKeys(this XmlAttributeOverrides overrides, System.Type type)
    var ignoreAttributes = new XmlAttributes();
    ignoreAttributes.XmlIgnore = true;

    foreach (var propInfo in type.GetProperties())
        foreach (var attr in propInfo.GetCustomAttributes(false))
            var associationAttr = attr as System.Data.Linq.Mapping.AssociationAttribute;
            if (associationAttr != null && associationAttr.IsForeignKey)
                overrides.Add(type, propInfo.Name, ignoreAttributes);

Then i need only add my linq types and i'm done:

XmlAttributeOverrides overrides = new XmlAttributeOverrides();

var serializer = new XmlSerializer(typeof(DiscussionDocument), overrides);

Rick Strahl
April 03, 2009

# re: LINQ to SQL and Serialization

@Aaron - good idea. But how do you deal with this if your serialization behavior will vary? In some situations you might want child collections to serialize, in others you might not.

For purely dynamic situations assigning null to references also works to prevent circular references and excessive serialization data, but it gets to be a pain to do this every time and this is where your approach might work better.

I suppose DataMember attributes though are the ones that you really need to worry about too.

April 03, 2009

# re: LINQ to SQL and Serialization

Hi Rick,

Your blog is quite useful. I pieced together information on LINQ to SQL Serialization and WF that you and your readers may find useful.

I spent a lot of time looking for a good solution, I hope this helps others.

Thank you for your very useful blog.


April 29, 2009

# re: LINQ to SQL and Serialization

Actually you can add the [XmlIgnore] tag to all the Association properties of de linq data context.

but I know this is not useful when we are working with many tables.

Nice blog, see yeah!

Marcos Boyington
April 30, 2009

# Per-property overriding of XML attributes

A better alternative to Aaron's post, that allows you to specify per-property what values to ignore, and even use more advanced settings, such as changing attribute names.

using System.Xml.Serialization;
using System.Reflection;
using System.ComponentModel;
using System.ComponentModel.DataAnnotations;

public static class XsLinq
    public static void SetupXmlAttributes(this XmlAttributeOverrides overrides, System.Type type)
            var metadataAttrib = type.GetCustomAttributes(typeof(MetadataTypeAttribute), true).OfType<MetadataTypeAttribute>().FirstOrDefault();
            var buddyClassOrModelClass = metadataAttrib != null ? metadataAttrib.MetadataClassType : instance.GetType();
            var buddyClassProperties = buddyClassOrModelClass.GetProperties();

            var properties = from buddyProp in buddyClassProperties
                                      join modelProp in type.GetProperties() on buddyProp.Name equals modelProp.Name;

            foreach (PropertyInfo prop in properties)
                overrides.Add(type, propInfo.Name, new XmlAttributes(prop));

We setup the runtime Linq class as follows:

    public partial class User
        public class User_Meta
            public object FirstName { get; set; }

            public object Friend { get; set; }


And finally:

XmlAttributeOverrides overrides = new XmlAttributeOverrides();

var serializer = new XmlSerializer(typeof(User), overrides);

Marcos Boyington
April 30, 2009

# Correction to my previous post

Slight correction to my previous post (unnecessary to do a join against the original class, as the XmlSerialiser will already have added these attributes):

public static class XsLinq
        public static void SetupXmlAttributes(this XmlAttributeOverrides overrides, System.Type type)
            var metadataAttrib = type.GetCustomAttributes(typeof(MetadataTypeAttribute), true).OfType<MetadataTypeAttribute>().FirstOrDefault();
            if ( metadataAttrib != null )
                var buddyClassProperties = metadataAttrib.MetadataClassType.GetProperties();

                foreach (PropertyInfo prop in buddyClassProperties)
                    overrides.Add(type, propInfo.Name, new XmlAttributes(prop));

David Lewis
May 12, 2009

# re: LINQ to SQL and Serialization

I'm relatively new to the issue of serializing LinqToSQL Objects (and none of this is set in code) but it seems to be the case that adding an attribute like [IgnoreForSerializationGroup(string groupName)] to a field and having a Serialiser that took a group argument and recognized the tag and simply did not attempt to serialize fields where the group name and attribute groupName corresponded. the best part is this could be done on a MetadataType associated with the entity type so no monkeying with the generated code.

With this setup you could define serialization slices across your entity model without having to change it's structure.

An example:
    public partial class Product

    public class ProductMetadata
        Object Categories;

    public partial class Category

    public class CategoryMetadata
        Object Products;

then a call to GroupableSerializer i imagine to be like this:
var productXML =  GroupableSerializer(Product,"ProductGroup")

var categoryXML =  GroupableSerializer(Category,"CategoryGroup")

May 26, 2009

# re: LINQ to SQL and Serialization

Thanks for this post,it solves my problem for circular dependancy. and i can get the result of my xml web service.

Anuj Pandey
June 15, 2009

# re: LINQ to SQL and Serialization


My problem involves Json Serializer of MVC framework, what would you suggest to this ? since the error is still the same.

IVAPM.Models.IVAPMEntities context = new IVAPM.Models.IVAPMEntities();

int pageIndex = Convert.ToInt32(page) - 1;
int pageSize = rows;
int totalRecords = context.MappUser.Count();
int totalPages = (int)Math.Ceiling((float)totalRecords / (float)pageSize);

var mapusers = from ob in context.MappUser.OrderBy("it." + sidx + " " + sord).Skip(pageIndex * pageSize).Take(pageSize)
select ob;

var result = new JsonResult();
result.Data=new {
total = totalPages,
page = page,
records = totalRecords,
rows = mapusers

return result;

Jonathan Ulfeng
September 14, 2009

# re: LINQ to SQL and Serialization

Thank you so much for your example. You saved me hours of grief here trying to figure out why my silverlight client was crashing on my one-many tables with constraints in SQL Server.

Marcel Peereboom
January 21, 2010

# re: LINQ to SQL and Serialization

This post saved my day. Thanks a lot

Rafe Kemmis
March 23, 2010

# re: LINQ to SQL and Serialization

Just thought I would mention my solution to this same problem when I was trying to use the JavaScriptSerializer class to serialize Linq-to-Sql entities to json. It turns out that if you only need to serialize down one level of entities, then you can set the RecursionLimit to 1 on the JavaScriptSerializer instance, and it will inherently not cause a circular reference.

JavaScriptSerializer json = new JavaScriptSerializer();
json.RecursionLimit = 1;

Option #2 is to use a 3rd party json library like JSON.Net (, which gives you options on how to handle circular references:

JsonSerializerSettings jsSettings = new JsonSerializerSettings();
jsSettings.ReferenceLoopHandling = ReferenceLoopHandling.Ignore;
JsonConvert.SerializeObject(myLinqEntityInstance, Formatting.None, jsSettings)

John Leitch
March 25, 2010

# re: LINQ to SQL and Serialization

I created a utility that generates a WCF data access service based on LINQ-to-SQL generated classes.
It can be found at

April 14, 2010

# re: LINQ to SQL and Serialization

I have a problem with LINQ to SQL and C#, I try to change the Association Properties but this are atributes, how can i do? For example, just for one WebMethod I want to set the Table1-Table2 Association Child Property Access as Internal.
This is my code,

              public  List<Perfil> SE_ListarPerfilesXIdSistema (int IdSistema)
            SeguridadDataContext seguridad = new SeguridadDataContext();
         List<Perfil> perfil = (from p in seguridad.Perfil
                                   where p.Sist_Id == IdSistema
                                   select p).ToList();
            return perfil;

May 29, 2010

# re: LINQ to SQL and Serialization

Hi FYI, I am using WCF Service with Linq SQL. at that time above Error shows. System.Data.Linq.Table`1'[MyProject.DAL.Library.Product]' can not be serilized with DataMemberAttribute attribute.
Can any one guide me?

August 20, 2010

# re: LINQ to SQL and Serialization

OMG! ... do not create stupid data model, and you will never faced with stupid problems...

January 26, 2012

# re: LINQ to SQL and Serialization

I have tried the following 2 things to avoid this circular reference error:

- Added [System.Xml.Serialization.XmlIgnore] to all EntitySet<t>
- Set all child associations to Internal

I double-checked that I got ALL of them. In fact I wrote a little parser to go through and parse and update the items, because it's about a thousand lines of xml.

Neither of these approaches worked. I see some suggestions that look interesting but would require a hundred hours of refactoring my code to implement; not an option for me.

It's driving me crazy that I've seen lots of posts saying that 1 or both of the above items worked for them, yet it doesn't work for me. I've triple checked, read lots of articles, recompiled, tried variations on different suggestions. Nothing.

Anything I might have missed?

West Wind  © Rick Strahl, West Wind Technologies, 2005 - 2015