TrailAssembly is a catalogue of specific Salesforce problems: an exception with a name, a Flow that fails silently at 200 records, a permission that does not do what its label says. Each article takes one of them and explains what triggers it and how to fix it.
Why this exists
Search for a Salesforce error and you usually get a forum thread from 2019 with no accepted answer, a documentation page that describes the feature without addressing your problem, and a few posts that restate the documentation in different words. Few of them say what actually went wrong or how to confirm you fixed it. This site tries to be the page you wanted when something broke.
Tested, or written from the documentation
Some articles were reproduced in a Developer Edition org. The error was triggered on purpose against made-up records, so the article can quote the message the platform actually prints. Those carry a badge at the top with the release and the date they were run on.
Most of the catalogue was written from Salesforce’s documentation, release notes and developer guides without being run in an org. Those say so in a note at the top. If one of them does not match what you see, the comment form on the article is the quickest way to get it checked.
How articles are written
- The answer comes first. The fix goes in the opening paragraph, and the explanation follows for readers who want it.
- Primary sources. Official documentation, release notes, developer guides and the Known Issues site, rather than other people’s blog posts. When someone is quoted, it is briefly, in quote marks, with a link.
- Every citation is opened before it is linked. Salesforce’s documentation site returns HTTP 200 for pages that do not exist, so a status code proves nothing. Where a page could not be loaded, the document is named without a link.
- The cost of a fix gets named. Moving work async to clear a governor limit costs you eventual consistency, and an article that recommends it should say so.
- Diagrams are drawn for the article they sit on. Screenshots come from a developer org with made-up data. Header photos are licensed stock credited to the photographer, or official press images credited to the company that issued them. Nothing is taken from another site’s articles, and no customer data appears anywhere.
- Credit for techniques. If an approach came from someone’s talk, answer or repo, they are named and linked.
- No exam dumps. Certification articles are study guidance. They never reproduce exam questions.
Who writes it
Articles without a named author are published under the Trail Assembly name. Outside contributors get their own byline and bio. If you have solved something properly and want to write it up, see Write for Us.
When an article is wrong
Corrections are published rather than patched in quietly. The article says what changed and when, and the reader who reported it is credited unless they would rather not be. If something here no longer matches your org, tell us which release you are on and what you saw.
Independence
This site is not affiliated with, sponsored by, or endorsed by Salesforce, Inc., and it is not connected to any employer of its contributors. Nobody pays for coverage. If that changes, the sponsorship or affiliate link will be disclosed on the page where it appears.