How to avoid running status?

Trung_Nguy_n · September 27, 2023, 7:19am

When I run workflows, they don’t go to completion status, but stay in running status. How to avoid the above situation?

MutedJam · September 27, 2023, 7:31am

Hi @Trung_Nguy_n, without the information requested in the question template (especially your n8n version, and an example workflow to reproduce this) it’s hard to say what’s happening here I am afraid.

That said, we do have a fix coming up to ensure the success status is correctly written in certain cases. Perhaps you want to follow the GitHub pull request and upgrade your n8n version once this is released?

github.com/n8n-io/n8n

fix(core): Prevent executions from displaying Running status incorrectly

n8n-io:master ← n8n-io:help-338-large-number-of-long-running-executions-for-nadjalemlist

opened 02:46PM - 26 Sep 23 UTC

krynble

+1 -0

Github issue / Community forum post (link here to close automatically): https:/…/linear.app/n8n/issue/HELP-338/large-number-of-long-running-executions-for-nadjalemlist#comment-18d1fc96 After investigating this issue with @ivov and @flipswitchingmonkey we've identified this missing assignment of execution status. This is the only inconsistency we've found that could cause executions to continue displaying as `Running` even after finished.

Trung_Nguy_n · September 27, 2023, 7:36am

If I stop the running state. Then I see that my node has stopped at No Operation, due to nothing.

Version : 1.3.1

shrey-42 · October 14, 2023, 6:29am

Hey @MutedJam ,

i’m facing a similar situation:

All of the above shown executions have already completed (in around 4-5 mins. each)
The workflow is Webhook (immediate response) based
The above screenshot has been taken after restarting the n8n docker container, multiple times!
Stop Execution seems to have no effect as well

The execution_entity table from the database:

The concerned workflow id is - FeIJmu5MSLjyneKm

My environment:

n8n version: 1.9.3
Database (default: SQLite): Postgres
n8n EXECUTIONS_PROCESS setting (default: own, main): main
Running n8n via (Docker, npm, n8n cloud, desktop app): Docker
Operating system: Windows

yukyo · October 14, 2023, 11:59am

Hi everyone, just my cents to add more details. I have started to see this behavior since I updated from 1.5.1 to 1.9.3 where execution shows as running, but actually, they completed:

As you can see in the left panel it says running for 1:04m, but actually that workflow is super simple two-steps nodes that should be ok in 1–3 secs.

2º pic still running:

ID of that execution: 2822 (also in the image, another workflow shows as running, but it’s not)

Let’s look at ID 2822, completed in 2.5s

Moreover, I have been seeing this error for a few days (not sure if it’s related or not):

2023-10-14 11:44:42.168 UTC [10211] ERROR: duplicate key value violates unique constraint "pk_workflow_statistics"

2023-10-14 11:44:42.168 UTC [10211] DETAIL: Key ("workflowId", name)=(ScHW0PpO2sVw7BW0, data_loaded) already exists.

2023-10-14 11:44:42.168 UTC [10211] STATEMENT: INSERT INTO "public"."workflow_statistics"("count", "latestEvent", "name", "workflowId") VALUES ($1, $2, $3, $4)

2023-10-14 11:44:42.354 UTC [10222] ERROR: duplicate key value violates unique constraint "pk_workflow_statistics"

2023-10-14 11:44:42.354 UTC [10222] DETAIL: Key ("workflowId", name)=(PMdUTwEZ9ZcjW2hq, data_loaded) already exists.

2023-10-14 11:44:42.354 UTC [10222] STATEMENT: INSERT INTO "public"."workflow_statistics"("count", "latestEvent", "name", "workflowId") VALUES ($1, $2, $3, $4)

2023-10-14 11:44:43.187 UTC [10211] ERROR: duplicate key value violates unique constraint "pk_workflow_statistics"

2023-10-14 11:44:43.187 UTC [10211] DETAIL: Key ("workflowId", name)=(ScHW0PpO2sVw7BW0, data_loaded) already exists.

2023-10-14 11:44:43.187 UTC [10211] STATEMENT: INSERT INTO "public"."workflow_statistics"("count", "latestEvent", "name", "workflowId") VALUES ($1, $2, $3, $4)

2023-10-14 11:44:43.847 UTC [10223] ERROR: duplicate key value violates unique constraint "pk_workflow_statistics"

2023-10-14 11:44:43.847 UTC [10223] DETAIL: Key ("workflowId", name)=(ScHW0PpO2sVw7BW0, data_loaded) already exists.

2023-10-14 11:44:43.847 UTC [10223] STATEMENT: INSERT INTO "public"."workflow_statistics"("count", "latestEvent", "name", "workflowId") VALUES ($1, $2, $3, $4)

2023-10-14 11:44:44.073 UTC [10211] ERROR: duplicate key value violates unique constraint "pk_workflow_statistics"

2023-10-14 11:44:44.073 UTC [10211] DETAIL: Key ("workflowId", name)=(ScHW0PpO2sVw7BW0, data_loaded) already exists.

2023-10-14 11:44:44.073 UTC [10211] STATEMENT: INSERT INTO "public"."workflow_statistics"("count", "latestEvent", "name", "workflowId") VALUES ($1, $2, $3, $4)

n8n version: 1.9.3
Database (default: SQLite): Postgres
n8n EXECUTIONS_PROCESS setting (default: own, main): queue
Running n8n via (Docker, npm, n8n cloud, desktop app): Docker

MutedJam · October 16, 2023, 8:18am

Hi @yukyo, these are two separate issues I am afraid. The duplicate key queries made by n8n are still sitting in the engineering backlog waiting for a closer look and fix, but one possible cause of “endless” executions has been fixed with [email protected]: fix: Prevent undefined issues when restoring binary data by krynble · Pull Request #7419 · n8n-io/n8n · GitHub

Perhaps you can try upgrading your n8n instance and confirm if you’re still seeing this behaviour for new executions started after he upgrade?

yukyo · November 1, 2023, 1:08pm

@MutedJam Hi, sorry didn’t have a chance to report it, I’m running in the version 1.11.2, and I still see this infinite execution issue:

running for +2 minutes and If I stay on this page it could stay running for hours, sometimes I go AFK and I leave the executions page open and when I’m back, I see hundreds of yellow execution running with +300m executions, but those are fake.

In the above image, execution ID: 144845 actually was completed in: 40 sec

Also, I still see this error in the postgresql logs:

I’m planning to update to version 1.14.2 tonight. Do you think those issues have been addressed on that version?

Thanks.