Strange default sorting of all executions

[solveretur]

Describe the problem/error/question

Recently I’ve updated my n8n from version 1.18.4 → 1.49.0 and I noticed that previously when I went to the all executions page it displayed the entries sorted from the newest to the oldest no matter what was the status of the execution. However after the update it displays failed executions at the top, ignoring their dates (so I see executions from 2023 at the very top) and when I want to see my latest, successfully finished executions I need to use filter.

I’d like to ask whether it’s a feature or it’s a bug ? If it’s a feature, are we able to change some settings to bring back the old behavior ?

Before:

After:

Information on your n8n setup

• n8n version: 1.49.0
• Database (default: SQLite): postgres
• n8n EXECUTIONS_PROCESS setting (default: own, main): default
• Running n8n via (Docker, npm, n8n cloud, desktop app): docker
• Operating system: linux

[bartv]

I can’t reproduce this; are you able to see successful executions at all? I wonder if your workflow settings got changed:

If your defaults have changed and you’re self-hosted, you can update them:

Execution data | n8n Docs

[solveretur]

Yes I’m able to see the successful ones but I need to use the filter

My workflows settings are all the same and look like this

It all happened after I did the update. Maybe it’s some bug on the n8n side where you do some NULLS FIRST on the db and that’s why I see the results like that.

[Jon]

Hey @solveretur,

I am also not able to reproduce this on 1.49 with Postgres, It may sound odd but we have seen a few odd issues with browser caching recently. Can you try loading the page in private browsing mode to see if the issue is still there?

[solveretur]

it’s the same, and my teammates have the same issue

We use n8n from version 0.179.0 so maybe there is some bug

[bartv]

It’s also odd that the action button says ‘stop’ for the errorred workflows - it’s supposed to be ‘view’:

[solveretur]

ok I fixed it for now I just deleted them all manually - I guess it’s because those are some old failed executions even from 0.179.0 version so maybe there is some issue while migrating - thanks for help !

[bartv]

Glad to hear it! We’ll keep an eye on this issue.

[Jon]

That is great to hear, It may be worth generally thinking about how long you need to keep logs for as well which will lead to a happier system