4.7.* to 4.8.* Migration Issues

Without going into too much detail, from what I can tell of the changes, upgrading from v4.7 to v4.8 is going to require me to debug all of my custom controller logic, and re-think/re-write a bunch of it. I don’t know what my question is, but I guess it would have been helpful to have some sort of flag that there is some breaking changes here. I understand why it’s necessary, but for me at least the upgrade won’t be simple, and I’ll probably wait for a few weeks until more info on the changes can be disclosed. A migration guide would be helpful at that point too.

System Information
  • Strapi Version: 4.8.1
  • Operating System: Linux
  • Database: Postgres
  • Node Version: 16.*
  • NPM Version:
  • Yarn Version:

I don’t think there is any breaking changes to it. There where some hotfixes from 4.8 to 4.8.2. So upgrading from 4.7 to 4.8.2 should be pretty straight forward.

Without going into the details I can’t say or answer if there is a “breaking change” of course.

I can confirm that I experience similar issues after updating from 4.7.1 to 4.8.2.

I submitted a bug report with details: Update from 4.7.1 to 4.8.2 results in missing parts in the response · Issue #16137 · strapi/strapi · GitHub

I guess it depends on your definition of breaking change I guess. I do know that certain logic I was implementing in controllers will not work after upgrading. Can discuss pros/cons once the details are released.

I would say that if you are looking for release details to have a look at the releases
This lists each one :slight_smile:

I am not sure what just happened but after I migrated from v4.6 to 4.8.2 now randomly my DB get erased in Production Env AWS EC2 when I restart it using PM2

1 Like

This happened at my site too. I had a development instance I was still working on our future schema for. We went from 4.7.x to 4.8.2 because of the security issue but I didn’t check things right away. We discovered that all of our schema was missing from the DB (postgres, AWS Aurora) and unfortunately this environment didn’t have a long enough backup interval on the DB.

There’s bits and pieces of content still there (some references to files uploaded) but all of our models are gone and have to be rebuilt from scratch.

I wish I had some way to diagnose what happened here but so fare there’s nothing that indicates what went wrong.

I have unexpected issue after migrating from 4.7.1 to 4.8 (and I tried even higher with same issue).
In firs look everything ok. but When I go to Market place …/admin/marketplace in admin panel it says an error

Compiled with problems:
ERROR
Cannot read properties of undefined (reading 'rankings')
TypeError: Cannot read properties of undefined (reading 'rankings')
    at matchSearch (webpack-internal:///23928:58:73)
    at MarketPlacePage (webpack-internal:///23928:143:31)
    at renderWithHooks (webpack-internal:///73116:14985:18)
    at updateFunctionComponent (webpack-internal:///73116:17356:20)
    at beginWork (webpack-internal:///73116:19063:16)
    at HTMLUnknownElement.callCallback (webpack-internal:///73116:3945:14)
    at Object.invokeGuardedCallbackDev (webpack-internal:///73116:3994:16)
    at invokeGuardedCallback (webpack-internal:///73116:4056:31)
    at beginWork$1 (webpack-internal:///73116:23959:7)
    at performUnitOfWork (webpack-internal:///73116:22771:12)

And on role page … /admin/settings/roles

Compiled with problems:
ERROR
(0 , match_sorter__WEBPACK_IMPORTED_MODULE_3__.default) is not a function
TypeError: (0 , match_sorter__WEBPACK_IMPORTED_MODULE_3__.default) is not a function
    at useSortedRoles (webpack-internal:///39466:60:79)
    at RoleListPage (webpack-internal:///39466:216:7)
    at renderWithHooks (webpack-internal:///73116:14985:18)
    at mountIndeterminateComponent (webpack-internal:///73116:17811:13)
    at beginWork (webpack-internal:///73116:19049:16)
    at HTMLUnknownElement.callCallback (webpack-internal:///73116:3945:14)
    at Object.invokeGuardedCallbackDev (webpack-internal:///73116:3994:16)
    at invokeGuardedCallback (webpack-internal:///73116:4056:31)
    at beginWork$1 (webpack-internal:///73116:23959:7)
    at performUnitOfWork (webpack-internal:///73116:22771:12)
1 Like

I am unable to replicate it as well, as it happens randomly after deployment with no errors, and drops all the database records… now I have to make a DB backup before deploying :frowning: , I hope it got sorted it is very critical as data is the key.

I’ve gotten this error on both my prod server and my local dev environment. I believe it stems from db migrations on postgres but I haven’t been able to find a cause or solution. On my prod server the problem fixed itself after some time. This could potentially point to a primary key indexing error (which is an ongoing issue when working with postgres and an import/export plugin).

I just wanted to add a +1, this bug definitely exists somewhere.