I have access to a SQL Server DB system that only allowed "Windows Authentication" - that is, use your windows login credential to access the data. (Not using a username-password). It also controls database permissions.
I want to build a REST API for users not familiar with SQL to get data from that server. The problem is, I need to take user's credential to execute the command, not the machine that hosts the REST API. How could I design the system that allow user to pass their login and let the API send query to the server using that specific login?
I found a reference from Microsoft, not sure if it's relevant.
https://learn.microsoft.com/en-us/sql/relational-databases/security/choose-an-authentication-mode?view=sql-server-2017
What you are after is Kerberos delegation, sometimes also known as "double hop". It is not something you have to explicitly implement in your code, as long as you stick with the Microsoft stack; rather, it is a matter of configuration.
This post describes the setup required for the feature to work in sufficient details. You can also find some basic requirements in this help article. For anything beyond that, just google it.
Related
"Issue description copied..."
I'm building a partner connector, which relies on a user name and password to connect to database (very similar to the existing Postgres / MySQL connectors provided by Google). In order to verify the credentials, I also need the database host information to be present in addition to username and password and this is the base of my problem.
The Google build connectors conveniently are allowed to collect user credentials and the database related information at the same time. Unfortunately, that doesn't seem to be the case for partner connectors as stated in the requirements
Point 5 "Use appropriate authentication method in getAuthType(). Do not request credentials via getConfig()."
The authentication itself happens before any other configuration details are known (there is just a dialog for username and password) and there doesn't seem to be a way to request additional information on the authentication screen itself. Once the credentials have been entered, the verification also happens immediately, before the configuration is being shown in the next step.
Once credentials are validated successfully, Datastudio then assumes the schema and data can be requested.This excludes the option of a dummy confirmation, because there doesn't seem to be a way to tell credentials are invalid and need to be changed after checking the other configuration details on the next screen.
That makes me unsure, how to determine valid credentials in my use case as I need to know the variable endpoint to authenticate against. I definitely want to avoid storing any user credentials myself in an external database, because this opens up another can of worms.
Has anyone successfully solved a similar issue before and can provide guidance here?
This is a known limitation of the authentication methods for Community Connectors.
A workaround would be to use authtype NONE and then request the credentials and database information in the config. This is, however, not a recommended approach.
I am new to Salesforce, but am an experienced developer. I am provided a link to a Salesforce report, which mostly has the right filters (query). I would like to use an REST API to pull that information as CSV or JSON so that I can do further processing on it.
Here are my questions:
Do I need special permissions to make API calls? What are they?
Do I need to create an "app" with client-key & secret? Does my admin need to grant me permission for this too?
There are a lot of REST APIs from Salesforce, which one do I need to get the info from the report? Analytics?
How do I authenticate in code?
You'd have to work with the System Administrator on the security pieces. Anybody who knows how the company works, can all users see everything, is there Single Sign-On in place, how likely is the report to change...
You will need an user account to pull the data. You need to decide if it'll be some "system account" (you know username and password and have them stored in your app) or can it run for any user in this org. It might not matter much but reports are "fun". If there will be data visibility issues 6 months from now, you'll be asked to make sure the report shows only French data to French users etc... you can make it in report filters or have multiple reports - or you can just use current users access and then it's the sysadmin that has to set the sharing rules right. (would you ever think about packaging what you did and reusing in another SF instance? Making a mobile app out of it? Things like that, they may sound stupid now but will help you decide on best path)
The user (whether it'll be system account or human) needs Profile permissions like "API Enabled" + whatever else you'd need normally ("Run Reports" etc). If you're leaning towards doing it with system user - you might want to look at Password Policies and maybe set password to Never Expires. Now this is bit dangerous so there would be other things you might want to read up about: "API only user" (can't login to website), maybe even locking down the account so it can login only from certain IP ranges or at certain times when the job's supposed to be scheduled...
Connected App and OAUth2 stuff - it's a good idea to create one, yes. Technically you don't have to, you could use SOAP API to call login, get session id... But it's bit weak, OAuth2 would give you more control over security. If you have sandboxes - there's little-known trick. You can make connected app in production (or even totally unrelated Developer Edition) and use client id & secret from it to login to sandboxes. If you create app in sandbox and you refresh it - keys stop working.
(back to security piece - in connected app you can let any user allow/deny access or sysadmin would allow only say these 3 users to connect, "pre-authorize". Could be handy)
Login - there are few REST API ways to login. Depends on your decision. if you have 1 dedicated user you'll probably go with "web server flow". I've added example https://stackoverflow.com/a/56034159/313628 if you don't have a ready SF connection library in your programming language.
If you'll let users login with their own credentials there will be typical OAuth "dance" of going to the target page (Google login, LinkedIn, Twitter...) and back to your app on success. This even works if client has Single Sign-On enabled. Or you could let people type in their username and pass into your app but that's not a great solution.
Pull the actual report already
Once you have session id. Official way would be to use Reporting API, for example https://developer.salesforce.com/docs/atlas.en-us.api_analytics.meta/api_analytics/sforce_analytics_rest_api_get_reportdata.htm
A quick & dirty and officially not supported thing is to mimic what happens when user clicks the report export in UI. Craft a GET request with right cookie and you're golden. See https://stackoverflow.com/a/57745683/313628. No idea if this will work if you went with dedicated account and "API access only" permission.
If a user logs in to their computer using a Single Sign-On system such as Active Directory, LDAP, or Kerberos, is it possible for applications they run to know who they are and what system they authenticated with? Can I get enough information out of these systems to verify their identity without requiring any additional user input?
Specifically, I would like to be able to check these things:
Did they log in via a single sign-on system at all, or are they just using a regular user account on this machine?
What system did they use?
Does the system have some URI that would distinguish it from any other directory?
What is the current user's distinguished name in that directory?
Can I get some information which I can pass to another host to prove to that host that the user is who they said they are? For example, a token that can be used to query the SSO system.
I'd assume all of these things should be possible, and in fact encouraged, but I am not positive. I'm sure the method of getting at this information is
SSO (at least with Kerberos which is used by ActiveDirectoy) is based on a token. As soon as the user requests access to a kerberized system the system queries for the token and checks its validity for accessing the system. It's as good as querying for username and password. when the user did not log in with an Kerberos-account there is no tiket so no automated access.
using the token you can get the users login- name and from that you can then use that to query the SSO-backend (typically LDAP) for more information on that user.
LDAP is not an SSO-system as it is simply a storage query protocol but it is often used as backend for SSO-systems.
The problem often is kerberizing an application. for Webapps that means you have to kerberize the webserver so that that one then can handle the authentication process with the SSO-service and then pass that information on to the unferlying webapp.
Hope that answers you questions.
for more information have a look around the web for kerberos
You are really asking about two things:
Authentication: Who are you?
Authorization: What are you allowed to do?
Kerberos really only answers the first question, you need a secondary system like LDAP or Active Directory ( which is both kerberos and ldap in a single server) to answer the second.
If your system is using kerberos correctly, any user login should have an associated kerberos ticket. With this ticket, you can request "service" tickets
to prove your identity to remote servers that support kerberos. The ticket
contains your principal identity in the realm ( user#DOMAIN.NET ) that can be
used to query authorization systems.
However, the details required to get all the moving parts in that sentence working together "on the same page" so to speak can be very complex. The remote service has to support accepting kerberos credentials, it has to be either in the same realm or have a cross realm trust relationship configured.... The list
gets pretty long. Depending on your exact application environment, using all these things can be fairly trivial, or it can be next to impossible.
I am using Entity Framework and have the same model on many databases and servers. I want to set my connection string at runtime based on the logged in user. I am currently overriding CreateObjectContext() in my DomainService and setting the connection string there. This works great, but I'm concerned with the lack of security. For example, in my DomainService, I have an invoke method called SetConnectionString() where I pass in the connections string each time I need a new Entity. This is being done from the DomainContext (client side). Since I am using SQL Integrated Security, I am concerned that someone could manipulate the connection string on the clinet and get access to a database that they are not authorized to use.
What is the most secure way to dynamically set the connection string for the domain entity I want to use at runtime?
I do have the database name and server settings in a database that I could look up from the DomainService if I passed in my project id or some other piece of information, but I am then right back to the same issue - someone could easily spoof this info.
Ideas??
-Scott
Hopefully my answer is relevant as the same security issues are faced by flash developers...
The only way to make this really secure ("this" being database connections from your client side code) is to make each user only have read-only access to their data in the database.
As you are realizing, there is no way to stop the user from monitoring/modifying data that you are sending back to the server. Even if you encrypt the communication, the user still has access to the code and in the case of flash, source code due to trivial decompilation.
I am not a DB developer, but I do security audits of stuff like you are asking often, so the following advice could be wrong, but I just googled, and it seems like maybe you could use views to restrict users (sets of credentials) to specific information.
If not, this article seems relevant to what you want:
http://technet.microsoft.com/en-us/library/cc966395.aspx
What about using the ASP.NET membership framework? Tie the user login with a Login call in the framework to make sure the user an authorized user. Add a "RequiresAuthentication" attribute to each of the functions you want to protect in your domain service.
Here are more details on using ASP.NET membership with Silverlight: WCF RIA Services - Authentication, Roles, and Profiles
i'm developing a .net winforms desktop application intended to be run at several bank's branches as a backup application whenever the main one (a web application) is unavailable due to connection issues with the bank's central node. The branchs themselves don't count with any enterprise services besides a SQL-Server database. For that reason, the application should be able to connect directly to the SQL-Server. My problem arises when I have to provide the application with a password to connect to the database:
1) Storing the password in clear text in a app.config file or similar is not an option (the customer requires the password to be encrypted)
2) Storing the password encrypted in a configuration file leads to the need of having an encryption key locally available. The encryption key could be just hardcoded in the application's code, but it would be easily readable by using a .net-decompiler or similar.
3) Using a custom algorithm to encrypt/decrypt wouldn't work either due to the same reasons as 2).
4) Integrated security is not supported by the bank
Additionally, the customers requires that they should be able to change the password in one location (within a branch) without the need to go from one computer to another updating config files (that rules out the possibility of using the machine's key to encrypt the password in individual machine's config files like asp.net does)
Would you provide any other approach or suggestion to deal with this problem?
I would appreciate any help.
Thanks in advance,
Bernabé
I don't think that encyrpting the password by any means is going to solve your problem. If the user has to send the password to server, and the password is located on the box, then by definition the user running the application must have access to the password and be able to decrypt it. Otherwise, you wouldn't be able to authenticate them. This means that there will always be a way for the user to get at the password, regardless of where you store it.
I can think of 2 ways that could possibly work, but I'm afraid they're not exactly what you're looking for.
Remove the requirement of having the
user send the password to the server
by using some sort of local proxy
(for example using a WCF windows
service) to take your winform
requests and then send them on your
behalf to the DB server. If you
install the service using an account
different from the user's account,
then you can secure the password by
any of the means mentioned in the other
answers. They key here is to make
sure the application user does not
have access to the resources that
the service account needs to decrypt
the password.
Don't store the password in the web config. Assign each user a different user account and password at the database level and have them type it in when they log in.
You could use the protected configuration built into .Net. See Encrypting Configuration Information Using Protected Configuration in the MSDN docs. One of it's raison d'etres was to encrypt data such as connection strings in config files.
You could
To use DPAPI to store a encryption/decryption key securely: How To: Use DPAPI to Encrypt and Decrypt Data
To install a SQL Server Compact Edition (or another small database) into your workstations and to synchronize data when your web application comes online again.
To ask for help inside that institution, as other people could have solved that problem and could to help you.
Definitely agree with the above regarding DPAPI. Microsoft's Enterprise Library makes this an absolute breeze too, so I would consider looking there first.