Task retry and error handling in Django - Step-by-Step Execution

Concept Flow - Task retry and error handling

Start Task

↓

Try to execute task

↓

Task succeeds?

Yes→Finish Task

No↓

Check retry count < max retries?

Yes↓

Wait or schedule retry

↓

Retry task

Back to Try to execute task↓

Handle failure (log, alert)

↓

Finish Task with error

↩Back to Try to execute task

The task tries to run. If it fails, it checks if retries remain. If yes, it retries. If no, it handles the error and stops.

Execution Sample

Django

from celery import shared_task

@shared_task(bind=True, max_retries=3)
def send_email(self, email):
    try:
        # send email logic
        pass
    except Exception as exc:
        raise self.retry(exc=exc, countdown=5)

A Celery task tries to send an email, retries up to 3 times on failure with 5 seconds delay.

Execution Table

Attempt	Action	Exception Raised?	Retry Count	Next Step
1	Try sending email	Yes: ConnectionError	0	Schedule retry after 5s
2	Retry sending email	Yes: TimeoutError	1	Schedule retry after 5s
3	Retry sending email	Yes: SMTPServerError	2	Schedule retry after 5s
4	Retry sending email	Yes: SMTPServerError	3	Max retries reached, handle failure
End	Stop retries	N/A	3	Log error and alert

💡 Max retries (3) reached, task stops retrying and handles failure.

Variable Tracker

Variable	Start	After Attempt 1	After Attempt 2	After Attempt 3	After Attempt 4
retry_count	0	1	2	3	3
exception	None	ConnectionError	TimeoutError	SMTPServerError	SMTPServerError
task_state	Pending	Retrying	Retrying	Retrying	Failed

Key Moments - 3 Insights

Why does the task retry only 3 times and then stop?

Because max_retries is set to 3, so after the third retry (see execution_table row 4), the task stops retrying and handles the failure.

What happens if the task succeeds before max retries?

If the task succeeds, it finishes immediately without retrying (not shown in this trace but would skip retries).

Why do we use 'self.retry' inside the except block?

'self.retry' raises a retry exception that tells Celery to retry the task later, increasing retry_count and scheduling the next attempt.

Visual Quiz - 3 Questions

Look at the execution table, what is the retry_count after the second attempt?

A 0

B 2

C 1

D 3

💡 Hint

Check the 'Retry Count' column in row for Attempt 2 in execution_table.

At which attempt does the task stop retrying and handle failure?

A Attempt 4

B Attempt 2

C Attempt 3

D Attempt 5

💡 Hint

Look at the 'Next Step' column where it says 'Max retries reached' in execution_table.

If max_retries was set to 5, how would the execution_table change?

A The task would retry only 3 times as before.

B The task would retry 5 times before failure.

C The task would never retry.

D The task would retry only once.

💡 Hint

max_retries controls how many retries happen before stopping, see concept_flow and execution_table.

Concept Snapshot

Task retry and error handling in Django with Celery:
- Use @shared_task(bind=True, max_retries=N) to set retries.
- Wrap task code in try-except.
- On exception, call self.retry(exc=exc, countdown=seconds).
- Task retries up to max_retries times.
- After max retries, handle failure (log, alert).
- Helps recover from temporary errors automatically.

Full Transcript

This visual trace shows how a Django Celery task handles retries and errors. The task tries to send an email. If it fails, it raises an exception caught by the except block. The task then calls self.retry to schedule a retry after 5 seconds. The retry count increases each time. After 3 retries, the task stops retrying and handles the failure by logging or alerting. Variables like retry_count and exception change with each attempt. This pattern helps tasks recover from temporary problems by retrying automatically, improving reliability.

Practice

1. What is the main purpose of using task retry in Django background tasks?

easy

A. To automatically try the task again if it fails temporarily

B. To stop the task immediately when an error occurs

C. To speed up the task execution by running it multiple times

D. To log the task output without retrying

Solution

1. Step 1: Understand task retry concept

   Task retry is used to handle temporary failures by trying the task again later.

2. Step 2: Identify the purpose in Django tasks

   It helps tasks recover from temporary errors without manual intervention.

3. Final Answer:

   To automatically try the task again if it fails temporarily -> Option A

4. Quick Check:

   Task retry = automatic retry on failure [OK]

Hint: Retry means try again automatically after failure [OK]

Common Mistakes:

• Thinking retry stops the task immediately
• Confusing retry with speeding up tasks
• Assuming retry only logs errors

2. Which of the following is the correct way to enable retry inside a Django task using Celery?

easy

A. @app.task(bind=True)\ndef my_task(self): self.retry(countdown=10)

B. def my_task(self): retry()

C. @app.task()\ndef my_task(): retry(countdown=10)

D. def my_task(): self.retry(countdown=10)

Solution

1. Step 1: Recognize the need for bind=True

   To use self.retry(), the task must be bound with bind=True.

2. Step 2: Check correct syntax for retry call

   The retry method is called on self inside the bound task function.

3. Final Answer:

   @app.task(bind=True)\ndef my_task(self): self.retry(countdown=10) -> Option A

4. Quick Check:

   bind=True + self.retry() = correct retry syntax [OK]

Hint: Use bind=True to access self.retry inside task [OK]

Common Mistakes:

• Not using bind=True and calling self.retry
• Calling retry without self or decorator
• Missing parentheses or wrong function signature

3. Given this task code snippet, what will happen if the task raises an exception on the first run?

@app.task(bind=True, max_retries=3)
def fetch_data(self):
    try:
        # code that may fail
        raise ValueError('Temporary error')
    except Exception as exc:
        raise self.retry(exc=exc, countdown=5)

medium

A. The task retries once and then stops

B. The task fails immediately without retrying

C. The task retries infinitely every 5 seconds

D. The task retries up to 3 times with 5 seconds delay between tries

Solution

1. Step 1: Analyze max_retries parameter

   max_retries=3 means the task will retry up to 3 times after failure.

2. Step 2: Understand retry call with countdown

   self.retry is called with countdown=5, so retries wait 5 seconds before next try.

3. Final Answer:

   The task retries up to 3 times with 5 seconds delay between tries -> Option D

4. Quick Check:

   max_retries=3 + countdown=5 = 3 retries with 5s delay [OK]

Hint: max_retries limits retries; countdown sets delay [OK]

Common Mistakes:

• Assuming infinite retries without max_retries
• Thinking retry happens immediately without delay
• Confusing max_retries with number of total runs

4. Identify the error in this Django Celery task code that tries to retry on failure:

@app.task(bind=True)
def process_data():
    try:
        # risky operation
        pass
    except Exception as e:
        self.retry(exc=e, countdown=10)

medium

A. retry method called outside except block

B. No max_retries set, so retry won't work

C. Missing self parameter in task function definition

D. Incorrect exception handling syntax

Solution

1. Step 1: Check function signature for bound task

   With bind=True, the task function must accept self as first parameter.

2. Step 2: Verify usage of self.retry

   self.retry is called, but self is undefined because function lacks self parameter.

3. Final Answer:

   Missing self parameter in task function definition -> Option C

4. Quick Check:

   bind=True requires self parameter [OK]

Hint: bind=True means add self parameter to task function [OK]

Common Mistakes:

• Forgetting self parameter with bind=True
• Calling retry outside except block
• Assuming max_retries is mandatory for retry

5. You want a Django Celery task to retry only on network errors but fail immediately on other exceptions. Which approach correctly implements this behavior?

hard

A. Set max_retries=0 and catch all exceptions to call self.retry

B. Use try-except to catch network errors and call self.retry; re-raise other exceptions

C. Call self.retry unconditionally in except block for all exceptions

D. Use a decorator to retry on all exceptions automatically

Solution

1. Step 1: Differentiate exception types in except block

   Catch only network-related exceptions to retry, others should raise immediately.

2. Step 2: Use self.retry only for network errors

   Call self.retry inside except for network errors; re-raise other exceptions to fail fast.

3. Final Answer:

   Use try-except to catch network errors and call self.retry; re-raise other exceptions -> Option B

4. Quick Check:

   Retry selectively by exception type using try-except [OK]

Hint: Retry only inside except for specific exceptions [OK]

Common Mistakes:

• Retrying on all exceptions without filtering
• Setting max_retries=0 disables retries
• Using decorators without exception control