I just wanted to let this mailing list know that for the Cloudtrust project (https://github.com/cloudtrust), we have developed a certain number modules for Keycloak. These are currently compatible with the version 3.4.3.Final of Keycloak, but we will make them compatible with Keycloak 4.X (where X will be the latest sub-version of Keycloak when we start working on this) as soon as we can. These modules are:
* keycloak-wsfed (https://github.com/cloudtrust/keycloak-wsfed): an implementation of the WS-Federation protocol for keycloak. This allows to select the WS-Federation protocol for Keycloak clients and for identity brokers.
* keycloak-authorization (https://github.com/cloudtrust/keycloak-authorization): this module allows the use of the client authorization system to prevent a user which is authenticated in a Keycloak realm to access a given client. It works no matter which protocol is used, and without the client having to support any extra protocol. Note: this solution is a bit hacky, but necessary for one of our use-cases.
* keycloak-export (https://github.com/cloudtrust/keycloak-export): a module adding an endpoint to fully export a realm while Keycloak is still running (no need for restarts!).
PS: I'm mailing this both dev and user mailing lists as I believe it may interest members of both mailing lists
Hello Keycloak Users!
We are planning on using the Client Registration flow for setting up
clients on login.
This is mainly to more clearly identify each individual device a user
has logged in with.
Are there anyone using this feature in production with a large number
With our current stats, we would probably end up with a few million
clients by the end of the year.
1. Will this scale well with the way Keycloak works?
2. If a user loses their device, how should a full revoke & logout be performed?
3. Is there an alternative approach to give each user more control
over their device and session?
I am building a multi-tenant mobile application that uses keycloak as a SSO server. We will pre-load users in keycloak using their email address as their username with a separate realm for each tenant. When a user logs into the mobile app I need to detect the realm from a user's email domain and redirect to the appropriate authorisation end point for the realm. Has anyone faced a similar problem?
My thoughts at the moment is to build a proxy api that the mobile application redirects to that prompts the user for their email address, look up the configured tenant form the email domain and redirects to the appropriate realm's login page passing the mobile app credentials it passes to the proxy api and the entered user email as a login_hint.
Can anyone see any issues with this approach? Or a suggest a better approach?
I have a multi-tenant architecture deployed with keycloak.
At first, to investigate multi-tenant architecture, I have followed what
is available within keycloak:
The same application is deployed in both tenants with
* http://localhost:8080/multitenant/tenant1 and login as
user-tenant1, password user-tenant1
* http://localhost:8080/multitenant/tenant2 and login as user-tenant2,
When you specify http://localhost:8080/multitenant/tenant1, you are
redirected to tenant1, and you need to authenticate.
*2) description of the problem*
The issue I am facing, is that I have a customer client application,
which can redirected to several diffrent realms.
The realm selction is based on the email address.
* user1(a)foo.com ---> should redirect to realm foo
* user2(a)bar.com ---> shou0dl redirect to realm bar
In fact, the email analsys shoudl redirect to the correct realm (foo or
bar , or more).
Once I have the login screen of the corresponding realm1, it is the as
in /introduction/, where user authenticates normally in his specific
*3) Authentication workflow requirement*
In fact the authentication workflow process should be as follows:
* General welcome panel
* the user enter his email address
* based on the analysis of his welcome address, the users is
redirected to a specific authentication realm (foo or bar or more)
* The user enter is login/password in realm login authentication screen
After analysis, it sounds like that the keycloak authentication process
needs to be updated/modified with
1. adding an extra additional step (which is a general form asking
2. based on teh email analysis, the corresponding tenant login
screen is presented to the tenant
3. the user authenticates to the tenant with his login/password.
*4) How to move forward*
For information, Azure and atlassian already implements such a
redirection mechanism in SAAS multi tenant architecture.
Keycloak documentation does not seem to mention about such a possibility
to tailor "out of the box" the authentication workflow to our needs.
Could the mechanism described above being achieved by customizing the
authentication workflow by developing a specific authentication SPI
plugin which could handles the both steps mentioned above ?
Does this approach sounds correct to you, or is it something to rule out ?
Or woudl you advise another approach ?
Tkx for your help.
Gsm: +33(0)682 801 609
Tél: +33(0)489 829 238
Fax: +33(0)955 260 370
I'm working on a multi tenant project where usernames are actually their email addresses and the domain of the email serves as a tenant identifier.
Now in Keycloak I'll have different realms per tenant, but I want to have a single login page for all tenants and the actual realm that will do the authentication to be somehow resolved by the username (email address).
How do I go about doing that?
Best regards, Pedro.
Hello all,I’m working on an organization-based service and want to have resource-specific permissions that are restricted by (from a user perspective) organization-specific roles. Since I’m not familiar with the specific terminology, I’m thinking of something similar to how GitHub manages their permissions:- A single user can be a member of multiple organizations- A user can have a different roles with different organizations that grant them access to all of an organization's resources- A user can have access to a specific resource- That organization-specific role determines access to different organization resourcesAre there any best practices or patterns for this model? Thanks!Justin
I'm trying to sync the LDAP groups into Keycloak but it doesn't update the
membership if I add or remove it from a group in LDAP.
I was able to sync the groups and its users into Keycloak correctly if
those wasn't provisioned before. For example, if the user already exists in
Keycloak DB (provisioned from LDAP) and I remove it from a LDAP group (also
provisioned from LDAP), the user in Keycloak continues to being a member of
the group in the Groups tab of user's details screen and in client's group
mappers. However, if I open the Members tab of group's details screen the
user was removed from the group.
Is there any way to solve this problem? Because of my company policy I
can't use Keycloak to manage the groups.
I'm using Keycloak 2.5.1.
Thanks for the help
The AWS ALB will allow you to authenticate to cognito or OIDC nowadays.
I thought "Great, I can connect it up to my KeyCloak".
Sadly not. Well, I can connect it to KeyCloak and see sensible looking
headers and JWTs flowing back and forth.
And then the ALB says "500 Internal Server Error" :-(
I can see a request to keycloak (from the client) :
And it 302 redirects back to the ALB :
On the KeyCloak server I can see the POST requests from the browser coming
in and hitting the authenticate URL, KC hands back a 302 (the URL above)
Then the ALB does a POST to the token endpoint and gets a 200 response with
a nice chunk of access token. I can decode it and see my details quite
happily. I even validated the signature. (Using jwt.io 's debugger.)
Although the ALB doesn't ask for the certificate at any stage, so I don't
think it even bothers validating it.
But it doesn't seem to like it. And gives me a 500 error.
(I can authenticate with Google OIDC without any trouble...)
(NB Any secrets in any of those strings won't get you very far, there is no
content yet :-) )