Hi Odoo Community, I'm facing an issue while importing large datasets (e.g., thousands of records) using a cron job in Odoo, and this is specifically on *Odoo.sh*. 🧪 Scenario: I upload a large CSV file (with 5000+ records). A scheduled cron job runs every 5 minutes to process and create records in the database. After some time (usually mid-process), the cron crashes with this error: psycopg2.InterfaceError: cursor already closed 🔍 Observations: The issue does *not occur* with small files (und

Odoo is the world's easiest all-in-one management software.
It includes hundreds of business apps:

CRM
e-Commerce
Contabilidad
Inventario
PoS
Proyecto
MRP

All apps

Todas las publicaciones Personas Insignias

Etiquetas (Ver todo)

odoo accounting v14 pos v15

Acerca de este foro

Ayuda

Cursor closed error during large CSV import via cron job in Odoo.sh

database cronjob odoo18 odoo-development odoo-customization

1 Responder

2053 Vistas

Rushik Pitroda

Hi Odoo Community,

I'm facing an issue while importing large datasets (e.g., thousands of records) using a cron job in Odoo, and this is specifically on Odoo.sh.

🧪 Scenario:

I upload a large CSV file (with 5000+ records).
A scheduled cron job runs every 5 minutes to process and create records in the database.
After some time (usually mid-process), the cron crashes with this error:

psycopg2.InterfaceError: cursor already closed

🔍 Observations:

The issue does not occur with small files (under 500 records).
It seems to be related to long-running loops, likely causing the DB session or cursor to close unexpectedly.
This happens consistently during large data operations on Odoo.sh and also in Local.

❓ My Questions:

Why does the “cursor already closed” error occur during long cron executions in Odoo?
Since I’m using Odoo.sh, I cannot modify server timeouts or configurations — what are the alternatives?
What’s the best practice to safely process thousands of records in a cron job on Odoo.sh? and also in Local

Any help would be highly appreciated 🙏

Thanks in advance!

Chris TRINGHAM

If you are using a shared server on odoo.sh then you can't run long batch jobs. As a dedicated odoo.sh server is very expensive you probably need to look at other hosting options which allow you to have more control.

Christoph Farnleitner

You need to perform your import in smaller chunks.

We usually do this by just fetching a more reasonable number of lines from a large file and keep track of the last processed line once that chunk has been processed.

Now, by keeping track of the last record, you can utilize the ir.cron.trigger model. In here you would store the cron_id and a call_at datetime allowing you to continue with the next set of lines, once it's its turn.

This way you don't have to deal with manual cursor creation since each iteration of a cron task would just work on a limited number of records, safely process it, stop gracefully and the awaits the next trigger to run the cron task (starting from the last processed line + 1). This is logic you will have develop.

¿Le interesa esta conversación? ¡Participe en ella!

Cree una cuenta para poder utilizar funciones exclusivas e interactuar con la comunidad.

Inscribirse

Publicaciones relacionadas	Respuestas	Vistas
How to Set Decimal Accuracy for a Custom Float Field odoo18 odoo17 odoo-development odoo-customization	1 jun 25	1902
Schedule Action code is not working in odoo 18 Resuelto cronjob odoo18	1 nov 25	7499
🔍 Odoo 17 : How to Search Compute Fields Without store=True in Python (Backend)? Resuelto computed-fields odoo-development odoo-customization	1 jun 25	2073
Alternative to `numbercall` for Automated (Cron) Jobs (Scheduled Action) in Odoo 18 Resuelto cronjob AutomatedActions ScheduledActions odoo18	1 oct 25	6022
Is it safe to remove a module by deleting it from the addons directory along with all dependent modules? database addons uninstall ir_model_fields odoo18	1 ene 25	1867

Se marcó esta pregunta

🧪 Scenario:

🔍 Observations:

❓ My Questions:

¿Le interesa esta conversación? ¡Participe en ella!