If you are a software developer or architect then the word security should be of prime importance in everything you do. Forrestor just published 2023 predictions for software development. At the top of the list is a prediction that the recent wide spread adoption of low-code solutions will cause headline security breaches in 2023.
Luckily, Salesforce provides a robust and layered security approach, but if not used properly, it could open up several security vulnerabilities. If you are not familiar with how Salesforce security works, this is the time to beef up your knowledge.
If you are not familiar with how Salesforce security works, this is the time to beef up your knowledge.
Last year, I published a series for Pluralsight that I reference here on my blog. If you are a Pluralsight subscriber, I strongly urge you to check it out.
If you are not a Pluralsight subscriber, you can check out my series about it here on my blog. You can also signup for a free 10-day trial here. And, for a limited time, Pluralsight is running a Black Friday, Cyber Monday sale where they are offering 50% off their annual subscriptions.
UPDATE on 11/16/2022 – This post was written several years ago and is still very popular on my blog, so I decided to do an update. If you are working for a non-profit and using the Nonprofit Success Pack (NPSP) and CumulusCI, then you should strongly consider looking into open-source Snowfakery to generate fake data that adheres to relationships.
I am currently designing a course for lynda.com about improving Visualforce performance. I decided that it was necessary to load my Developer org with a bunch of fake Account data in order to demonstrate some of the performance improvement techniques I was suggesting. I am NOT talking about unit test data, but actual data (just randomly generated). I have tried to generate things like this in the past and it always turned into a big pain. So I went ahead and did some general queries to see if anyone had already slayed this dragon.
At first I was excited about an article I found on DeveloperForce about Generating and Loading Representative test Data for Salesforce and Force.com Orgs. Perfect, I thought. That is just what I need. Unfortunately, I needed it fast and this article details a painfully long way of generating a Ruby on Rails app and installing all sorts of things that I really did not have the time for. I have worked with RoR a bit and knew this method would take way too long.
I did not want to create my own custom Apex factory or install some expensive app off the AppExchange. So back to searching and then I came upon the website GenerateData.com. I was a bit skeptical about it working, but in less than 15 minutes, I had generated my first data set for 100 Accounts and loaded it into my Dev org. I had hit the fake data jackpot!!!!
Now in all fairness, it did take me a few more tries before I got the data exactly the way I wanted, but this method was definitely fast and effective. I will also tell you that you are limited to only generating datasets of 100 records at a time using the Free version, but if you are willing to pay a very small fee (which I was, since I think it is a GREAT deal for all the time it saved me), then you can save data sets and generate as many records as you need.
And to save you a little time, I am including a screenshot of what my final Account Data Set looked like.
With my paid account, I generated 1000 Account records and imported them using the Data Import Wizard in Setup | Data Management. It took me less than 2 hours to get all the data loaded exactly as I needed it. I consider that HUGE!!!
I hope you find this tool as useful as I did. I sure do appreciate the developer, Ben Keen who wrote it! GREAT job!!!!
Today my course about Getting Started with Salesforce Lightning Web Components was published on Pluralsight. This is my 14th course for Pluralsight. This not a tutorial. Instead, I tried to to explain why things were done the way they were and provide tips for how you can do the same when building your own solutions.
10 minute preview video
If you do not have a subscription to Pluralsight, you can always sign up for a free 10-day trial here. I have also created a GitHub repo that contains all the code for the course. Here you will find detailed instructions for how to install this in your own scratch org.
UPDATE: As of 2024, LWC.Studio is shutting down and the tool I used in this post is only available in read mode. Salesforce now recommends you use a tool called StackBlitz to Fork the LWC GitHub repo and create your prototypes, such as what you see here.
This post is in preparation for my soon to be released Pluralsight course, Getting Started with Salesforce Lightning Web Components. While creating that course, I realized how super valuable a tool called webcomponents.dev can be for prototyping an app using Lightning web components (LWC’s).
I created a sample app using their free online IDE. This prototype uses the scaffolding provided for a basic LWC app, with nested components. I am using JSON data to render a list of product tiles. My favorite reasons to use this tool:
See preview of app immediately – In VS Code, you have to configure your meta config file to expose the component, push/deploy to an org and then open that org to see the results. Best of all, you can see it on different screen sizes like tablet and phone.
Integration with GitHub – Create a new repo and then create a branch to make changes.
Share results with others – You can share with a URL (as I have done in this post) or by posting right to Twitter. Share as either a simple view or in play mode where the user is not required to fork for tinkering with the code.
This tool was selected by Salesforce to replace the old playground feature, that has been deprecated. In all fairness, this tool is so much better than the playground. You can use it to create more than just LWC apps. It supports JavaScript, TypeScript, React, Vue 3, etc.
All that being said, this is NOT a replacement for the Salesforce CLI or the Salesforce Extensions for VS Code.
All that being said, this is NOT a replacement for the Salesforce CLI or the Salesforce Extensions for VS Code. I would just use this for prototyping an app quickly or testing out a LWC base component.
🤯 This weekend, I discovered an issue trying to debug a Lightning Web Component in the Chrome debugger tools.
UPDATE on 11/13/2022: Salesforce includes a way to automatically disable LWS for your scratch orgs through the security settings in your config/project-scratch-def.json file. You will simply need to add a sessionSetting for lockerServiceNext set to false. Instructions below show you how to do this through Setup. Thanks to Grzegorz Skaruz for pointing this out.
If you try to debug JavaScript in a scratch org using the Chrome Debugger, you will no longer see the components listed in the page hierarchy under the Sources tab (like you may be used to doing).
To still debug your LWC’s, you will need to first disable this feature in your scratch org. You can do this in:
Session Settings -> Uncheck “Use Lightning Web Security for Lightning Web Components”. Click Save.
Unfortunately, it is still hard to see where to go in Chrome Debugger. But, you expand the c subfolder under components (see image below).
Chrome Debugger After LWC Disabled in Scratch Org
Good luck and if you find a better way to do this, please leave a comment below to help others.
Top five best practices for Lightning Web Components (LWC)
👏🏽 The following best practices are in order of importance.
# 1 – Always Cache Data
Lightning web components offer two ways to cache data. The first is to use Lightning Data Service, which automatically handles security for you. The best part is that you do not have to write any Apex classes – especially platform required test classes.
If you must use Apex, then you can simply mark your methods as cacheable using the AuraEnabled annotation. The Winter 23 release introduced the global scope option for this annotation.
@AuraEnabled(scope=global)
public static myCacheableMethod() {}
# 2 – Use Conditional Rendering
Conditional rendering means that portions of a component will not be rendered until a condition is met.
For example, let’s assume you had a component that displayed a list of widget data. You would not want the list to be built unless there was data available. If there was no data, then the user would see a message telling them there are no widgets.
<div if:true{widgets}>
<template for:each={widgets} for:item="widget">
<li key={widget.Id}>{widget.Name}</li>
</template>
</div>
<div if:false{widgets.length}>
There are no widgets available
</div>
#3 – Use Pagination with Lists
While we are on the topic of lists, rendering a list of data has the potential for causing lots of performance problems. Many Salesforce orgs have custom objects that contain hundreds, thousands, if not millions of records.
The best way to prevent these lists from getting out of control is to introduce a pagination component. You can see an example of this in the eBikes GitHub repository.
There are now 94 base Lightning web components to choose from. They cover everything from a simple input box to a complex record form.
These components not only offer the CSS from the Salesforce Lightning Design System (SLDS), but they offer a performance advantage. These components are already rendered at the client-side, so they do not require additional download processing.
# 5 – Use SLDS Icons and Styling
And since I mentioned the SLDS, let me remind you that the Lightning Design System website offers hundred of optimized icons. Using your own customized icons can result is low render quality and resolution, so be sure to take advantage to these readily available goodies.
Want to learn more about Lightning Web Components?
In this series, I will be sharing highlights from courses that are part of the Salesforce Certified Sharing and Visibility Designer Skill Path on Pluralsight. The skill path is designed to help anyone trying to pursue the Salesforce Sharing and Visibility Designer certification.
Setting Org-wide Defaults
Org-wide defaults (OWD) provide a default level of access for records owned by users. More importantly, they can be used to limit data access for each standard or custom object. You can set different levels for internal versus external users.
Default levels:
Private is the most restrictive and means that only the record owner and users above them in the role hierarchy can view or edit records.
Public read only allows all users to view records, but not edit them.
Public read/write is the least restrictive level and means full access for all users.
Public read/write/transfer is used only for case and lead records, since these types of records can be transferred to another owner.
For each object there is a “Controlled by parent” checkbox. As you might guess, this affects objects that are children. They will inherit the access level of the parent.
Every time that user attempts to access a record from a particular object, Salesforce will first check the OWD. If it is set as private or read/only, the system will look at the object’s sharing table and join the group membership table based on the ID of the user trying to access the record.
Depending on what is found, the least restrictive access will be granted. The access grant, or sharing row cause will be stored as a record in the sharing tables that I told you about in the last post. This is the record access calculation process.
Designing a Role Hierarchy
Role hierarchies provide data access for a group of users. It allows managers access to records owned by their employees. Essentially, record access is opened up vertically to users higher up in the hierarchy. By default, peers or other members assigned to the role will not have access to these records.
…role hierarchy should NOT be a duplicate of your company org chart.
The role hierarchy is just one tool included in the Salesforce Sharing Model. The role hierarchy sits right in the middle. This means that baseline access, implicit sharing or org-wide defaults will override access provided by the role hierarchy.
Salesforce Internal Sharing Model
When designing a role hierarchy, the following things should always be kept in mind:
Your role hierarchy should NOT be a duplicate of your company org chart.
Users should be grouped into access levels. Only users assigned to roles above them will have the same access as the record owner.
Roles should only be created for permanent groups of users and not a group that is considered temporary because changes cause expensive sharing recalculations to take place.
Orgs created prior to the Spring 21 release are limited to 500 roles.
Sharing Rules and Manual Sharing
It is important to realize that the entire Salesforce sharing model is model is built around record ownership. When a user creates a new record, they automatically become the record owner. Records can also be assigned to queues.
…the entire Salesforce sharing model is model is built around record ownership
Sharing rules are created for objects, but you only do this if the OWD for that object is set to private or read-only. Otherwise, there is no need to create a sharing rule since everyone has access to that object. They open up object access for a select group of users.
When thinking about selecting a group of users, they can be assigned to a public group, which is not the same things as a queue. A public group consists of one or more users. These users can be individual users, or they can belong to a role or territory. Users that are part of a group cannot own a record, but they can be part of a sharing rule.
Queues on the other hand are generally used to manage ownership of objects such as leads, cases, and even custom objects. The record is owned by the queue and not any of the users assigned to the queue. However, queues are not used in sharing rules. Sharing rules are configured with public groups or roles.
Manual sharing happens when one user wants to share a record with another user. They can do this by clicking a button in the user interface. But keep in mind that manual sharing is only available for Accounts, Contacts, Cases, opportunities, leads and custom objects.
Only certain users can grant this kind of access. This includes, the:
Record owner
Users above the record owner in the role hierarchy
Any user that has been granted full record access
Administrators
Manual sharing is primarily used for special exceptions. That is why it sits above sharing rules in that upside down triangle you saw earlier.
Welcome back to the second post for this series. In this post you will be introduced to the Salesforce Record Sharing Model for internal users. This will involve sharing Salesforce data across users, groups and roles.
In this series, I will be sharing highlights from courses that are part of the Salesforce Certified Sharing and Visibility Designer Skill Path on Pluralsight. The skill path is designed to help anyone trying to pursue the Salesforce Sharing and Visibility Designer certification.
Diagnosing User Access Issues
Most Salesforce Administrators will eventually be asked why some user cannot access certain data. In situations such as these, it is helpful to refer to a diagram such as the one below.
Salesforce Sharing Model for Internal Users
The Salesforce sharing model can be imagined as an upside down triangle. Tools at the top of the triangle provide the widest level of access to the greatest number of users. Tools located at the bottom can be used more precisely to grant certain kinds of access to the least number of users.
When trying to figure out user access problems, you would start at the bottom. In other words, baseline access should be the first thing you look at, followed by implicit sharing and so on. Keep going up the triangle until you determine what the problem is.
Sharing Tool
Description
Baseline Access
Involves a combination of a profile permissions along with permission sets.
Implicit Sharing
Salesforce’s built-in sharing behavior between account and child records (contacts, cases and opportunities)
Org-wide Defaults (OWD’s)
Determines an objects’ default access level and is the only way to limit record level access.
Role Hierarchy
Expands data access vertically. Allows managers to access records owned by the users they manage.
Sharing Rules
Define criteria for sharing access with specific users or users in public groups/roles
Manual Sharing
Typically used for special circumstances, users can intentionally grant record access to a user that would not normally have access.
Team Access
Used to grant access to teams, which are groups of users that work together on objects like accounts, opportunities or cases.
Territory Hierarchy Access
Used to manage and grant account access to users assigned to sales territories.
Salesforce Record Sharing Tools
I am not going to lie to you, the Salesforce record sharing model is complex. But don’t be overwhelmed because throughout this series, I will be introducing you to all of these tools.
Working with Access Grants
Access grants are what Salesforce uses to determine who sees what data. The process of determining this all starts with an object sharing table. Object Sharing tables are completely separate from the object table itself where all the Salesforce data lives, such as the information about an account lives.
Sharing tables will store information about the grant (or sharing type) such as whether it is explicit or implicit. Implicit grants happen when there are children records associated with a parent.
For instance, accounts and contacts are designed with this kind of relationship. Contacts are considered children of a parent account and understandably users that can access a contact can also access the account.
Implicit grants will override explicit grants, which happen when a record is shared with manual sharing or sharing rules. So it is important to always keep implicit grants, or implicit sharing in the back of your mind. If you ever have a situation where you cannot figure out why a user is accessing a record, consider implicit sharing.
If you ever have a situation where a user is accessing a record you think they should not have access to, consider implicit sharing.
Object sharing tables are created automatically and follow a very specific naming pattern. For example, when the object record table is named Account, the sharing table will be named AccountShare. And the thing that ties these two tables together is the owner of the record. When dealing with a custom object such as one named myCustomObject, the sharing table will have the object name followed by two underscores and the word Share.
Determining what values go into a sharing table occurs when record access is calculated. This is a separate process from when a user attempts to access the actual record in the user interface or with an API. This process only happens during a configuration change, such as creating a new custom object. And you should know that It is a very complicated resource intensive process known as sharing recalculation.
Things are done this way in order to improve record access performance. If all this checking was done in real time, the system would not perform well at all and users would be very unhappy.
Record access calculations can happen when changes are made to:
Group membership
Role hierarchies
Territory hierarchies
Kicked off manually by an Administrator
It is important to realize that record access calculations can act like a ripple effect in a Salesforce org. For this reason, large orgs should be especially careful when kicking this off or making any changes that might trigger it. The process could negatively impact the orgs performance.
Since this is such a HUGE topic, I will not be covering everything in this one post. Stay tuned for the next post where I will be covering more.
In this series, I will be sharing highlights from courses that are part of the Salesforce Certified Sharing and Visibility Designer Skill Path on Pluralsight. The skill path is designed to help anyone trying to pursue the Salesforce Sharing and Visibility Designer certification.
The first course in the series, Salesforce Security: Getting Started, will use complex customer scenarios for a global robotics provider named Globomantics. You will learn how to evaluate the use of object, field-level, role and security settings to secure the Salesforce org.
Salesforce Security Levels
From a high level, Salesforce enforces security through multiple layers or levels. At the outermost layer there is Organization access. This is controlled by login tools that allow you to control when and how a user logs in.
High-level overview of the different Salesforce security layers
Once logged in, access to objects and fields is controlled through a combination of profiles and permission sets. Profiles must be assigned to each user and permission sets are assigned to specific users.
Record level sharing sits between the object and field level. Once a user is granted access to an object, then specific data records can be shared with them. But access can be restricted on a field level basis, providing even more granular access.
At the record level, access is controlled through data sharing. For each object, this access can be restricted through Org-wide defaults (OWD’s), which apply to all users. Roles and sharing can then be used to open this access back up to certain users.
Controlling Access to a Salesforce Organization
Controlling access to a Salesforce organization (or org) is the first line of defense. Authentication is the process of verifying that a user is who they say they are.
Salesforce authentication covers a broad spectrum of available tools where tools on the left are the least complex, such as passwords associated with a username, and tools on the right are the most complex and offer tighter security.
Salesforce login tools on the left-side are the least complex
Multi-factor authentication (MFA), which you may know as 2-factor authentication involves verifying a user’s identity with two or more pieces of info. Starting in 2022, Salesforce will require customers to use MFA for internal users.
Network-based security deals with limiting where and when a user can login to Salesforce. Device activation (or identity confirmation) involves tracking information about the device used to verify a user’s identity.
Every computer or device that is connected to a public network is assigned a unique IP (Internet Protocol) address, such as 172.16.254.1. When a user logs in to Salesforce for the first time they are sent an activation email that is associated with that address.
When the Salesforce user clicks a link in that email using the same device they logged in with they will be directed back to Salesforce and the device used will be activated and considered secure. Salesforce Admins can allow users logging in from an internal network access without having to activate a device by setting a trusted IP range. This is done at the org-level and applies to all internal users.
Besides setting a trusted IP range, Administrators can restrict a user’s login IP range, along with specific login hours through permissions assigned to user’s at the profile level. By setting a login IP range, all user’s assigned to that profile, will not be able to login from an IP address outside of that range.
So remember, adding a Trusted IP ranges at the org level will only remove computer activation requirements for users logging in from a device within that range. Login IP ranges, which are applied at the profile level will instead prevent a user from logging in from a device outside the range.
Controlling Access to Object and Fields
Salesforce launched their CRM product back in 2000 and at that time the only way to grant user permissions to objects and fields was through the profile, which had a one to one relationship with the user object. As the platform grew, along with the number of permissions, things got much harder to manage.
Salesforce eventually introduced permission sets as a way to alleviate the pain points that were associated with profiles. Permission sets can be assigned to more than one user and for a while they were easier to manage.
Over time, as some orgs got bigger and had to manage lots of permission sets, even these became hard to keep up with. And so, Salesforce introduced the concept of permission set groups.
As a best practice, Administrators should use permissions or permission set groups to grant permissions and not the profile. In a nutshell, profiles should be used to restrict access and permission sets or groups should be used to grant access to specific users.
As a best practice, Administrators should use permissions or permission set groups to grant permissions and not the profile.
When thinking about granting permissions to a user, a best practice involves the principle of least privilege. This means that users should only be given the minimum permissions necessary to do their job. By keeping this in mind when approaching permissions you can be ensured that users are able to do their jobs, while also protecting the integrity of the entire org.
Access to objects are granted through CRUD permissions. CRUD stands for Create, Read, Update and Delete. Salesforce allows you to assign two other permissions to objects (View All and Modify All) that you might like to think of as super power permissions. For this reason, these permissions should be given sparingly and typically only to Administrators.
Fields access is handled with Field-level security or FLS, as it is better known. Where objects can be granted CRUD permissions, fields can only be granted read or edit access. Fields also do not have access to the super power permissions of “view all” or “modify all”.
It is important to realize that object-level access overrides FLS. For example, you cannot remove read permission for an object and then enable it for all that object’s fields. If the user does not have read access to an object, then the user can not see any of the object’s fields.
Controlling Access to Salesforce Records
Access to Salesforce records resolves around the concept of record ownership. Each record or row of data as you might like to think of it can be owned by an individual user or a group of users assigned to a queue.
Org-wide defaults (OWD’s) are the only way you have of limiting record access. All the other tools open up access to users through sharing tools or the role hierarchy.
OWD’s set the default level of access users have to each others records. Each object will be assigned a level (private, public read only or public read/write) and these can be different for internal versus external users. Public read/write transfer is a special level that is only available to lead and case objects since those are the only kinds of records that allow for a transfer.
The role hierarchy opens up access to data records vertically. Access moves from the top of the hierarchy down to the bottom. This allows managers to have access to the records of employees they manage. But peers on the same level, do not have access to each others records.
The role hierarchy opens up access to data records vertically.
Besides the role hierarchy, there are other sharing tools that open up access restricted by OWD’s. Record sharing is a huge topic and for that reason it was given an entire course for this series. I only mention it briefly here so you know where it fits with other Salesforce security access levels.
This course is one of several, designed to prepare you on your journey towards completing the Salesforce Certified Sharing and Visibility Designer Exam. In this course, you will learn how to successfully setup a Territory model using Salesforce’s Enterprise Territory Management.
By the end of the course, you will know how to do the following:
Define Salesforce Enterprise Territory Management
Understand the benefits of using Enterprise Territory Management
Understand how to setup Territory Management in Salesforce
Explain the capabilities of Enterprise Territory Management and its impact on data visibility
Explain how Enterprise Territory Management impacts forecasting in Salesforce
Globomantics is a robotics company that needs to configure Enterprise Territory Management
The fictional company at the center of the story line for this course and the whole series will be Globomantics. The main character of the story will be a Salesforce Architect that has been asked to design the security model for this Salesforce Enterprise customer. The story for this course will be that the Architect is continuing her analysis of their org and is now assisting on building territories to help with account management.
If you are able to check out the course, I would love to hear your feedback. Especially, if you are using this course to help prepare you for the certification exam.
saramorgan.net 