JIRA: add retry/rate limit support

[valentijnscholten]

fixes #13769

Enable JIRA Connection Retries and Rate Limiting Support

Summary

This PR enables configurable retry and timeout settings for JIRA connections to handle rate limiting (HTTP 429) and connection errors gracefully. Previously, JIRA connections had retries disabled (max_retries=0), causing immediate failures on rate limits or transient network issues.

Changes

Configuration

Added three new environment variables:

• DD_JIRA_MAX_RETRIES (default: 3): Maximum number of retry attempts for recoverable errors (429, 503, ConnectionError)
• DD_JIRA_CONNECT_TIMEOUT (default: 10 seconds): Connection timeout for establishing a connection to the JIRA server
• DD_JIRA_READ_TIMEOUT (default: 30 seconds): Read timeout for waiting for a response from the JIRA server

Technical Details

Retry Behavior

The jira library automatically retries on:

• HTTP 429 (Too Many Requests) - Rate limiting
• HTTP 503 (Service Unavailable) - Temporary server errors
• ConnectionError - Network connectivity issues

Migration Notes

No migration required. The new settings use sensible defaults that match the previous behavior for retries (now enabled with 3 retries instead of 0) and add timeout configuration.

Testing

Tested with:

• Management command to push finding 1 finding 500 times with 10 parallel celery workers

celeryworker-10     | [29/Nov/2025 13:30:49] WARNING [jira.resilientsession:381] Request rate limited by Jira. Request should be retried after 1 seconds.
celeryworker-10     | [29/Nov/2025 13:30:49] WARNING [jira.resilientsession:334] Got recoverable error from GET https://xx.atlassian.net/rest/api/2/issue/22871, will retry [1/3] in 1.5314574809826222s. Err: 429 Too Many Requests
celeryworker-6      | [29/Nov/2025 13:30:50] WARNING [jira.resilientsession:381] Request rate limited by Jira. Request should be retried after 1 seconds.
celeryworker-6      | [29/Nov/2025 13:30:50] WARNING [jira.resilientsession:334] Got recoverable error from GET https://xxx.atlassian.net/rest/api/2/issue/22871, will retry [1/3] in 1.8632971142324433s. Err: 429 Too Many Requests
celeryworker-9      | [29/Nov/2025 13:30:49] WARNING [jira.resilientsession:381] Request rate limited by Jira. Request should be retried after 1 seconds.
celeryworker-9      | [29/Nov/2025 13:30:49] WARNING [jira.resilientsession:334] Got recoverable error from GET https://xx.atlassian.net/rest/api/2/issue/22871, will retry [1/3] in 1.635605867089348s. Err: 429 Too Many Requests

Future improvements

If we encounter retry delays of over 60s, we may consider adding another layer of retries in our celery task where we just reschedule the task with a delay. This avoids blocking other waiting tasks, but has other challenges.

[mtesauro]

Approved