Unblock Links Tasks

🔍 Overview

These Celery tasks work together to verify and manage blocked Links:

1. check_blocked_links_via_google_task – checks blocked domains via Google Safe Browsing API, scheduled to run every 15 minutes.
2. unblock_link_task – unblocks links deemed safe after revalidation.

---

📝 Task Definitions

@celery.task
def check_blocked_links_via_google_task():
    logger.info("Checking BLOCKED links via Google Safe Browsing API...")
 
    domain_map = Link.get_domains_map(
        status=LinkStatuses.BLOCKED.value,
        reason_exclude=BlockReasons.DNS_NOT_RESOLVABLE.value
    )
    total = len(domain_map)
    if total == 0:
        logger.info("No blocked domains found to check!")
        return
    # count unique domains
    logger.info(f"Found {total} blocked domains to check...")
 
    for chunk_domains in chunked(list(domain_map.keys())):
        logger.info(f"Checking BLOCKED domains chunk: {chunk_domains}")
        unsafe_domains = GoogleSafeBrowsing.check_links(chunk_domains)
        if unsafe_domains is None:
            continue
        for domain in chunk_domains:
            if domain not in unsafe_domains:
                for obj in domain_map[domain]:
                    logger.info(f"Blocked link {obj.link} is safe!")
                    unblock_link_task.delay(obj.id)
        
        logger.info(f"Processed {len(chunk_domains)} domains, safe domains found: {len(chunk_domains) - len(unsafe_domains)}")
    logger.info("Finished checking BLOCKED links via Google Safe Browsing API.")
 
 
@celery.task
def unblock_link_task(link_id: int, action_by: str = "system"):
    try:
        link = Link.objects.get(id=link_id)
    except Link.DoesNotExist:
        logger.error(f"Link with ID {link_id} does not exist!")
        return
    if link.status != LinkStatuses.BLOCKED.value:
        logger.warning(f"Link {link.link} is not blocked, cannot unblock!")
        return
    if not is_domain_resolvable(link.domain):
        logger.warning(f"Domain {link.domain} is not resolvable, cannot unblock link {link.link}.")
        link.set_block(
            reason=BlockReasons.DNS_NOT_RESOLVABLE.value,
            with_switch=False,
            action_by=action_by
        )
        return
    # bypass of incorrect unblocking by Google Safe Browsing
    if timezone.now() - link.blocked_at < timezone.timedelta(hours=1):
        logger.warning(f"Link {link.link} was blocked less than an hour ago, skipping unblocking.")
        return
    # if current link has an active link - unblock it with return to work else just unblock
    active_link = link.get_main_active_link(search_direction=SearchDirection.FORWARD.value)
    link.set_unblock(with_return_to_work=active_link is not None, action_by=action_by)
    logger.info(f"Link {link.link} unblocked successfully!")

---

📑 Attributes

This table describes the parameters for unblock_link_task:

Name	Type	Description
link_id	int	Primary key of the Link to revalidate and unblock.
action_by	str	Actor requesting the unblock (system by default or other).

---

⚙️ Task Behavior & Flow

1. Blocked Links Revalidation

   • Retrieves all blocked domains excluding those blocked for DNS resolution issues.
   • Splits domains into chunks by 500 domains for batch checking.
   • Calls GoogleSafeBrowsing.check_links() on each chunk.
   • Enqueues unblock_link_task for every domain not flagged as unsafe.

2. Unblocking Operation

   • Fetches the Link record by link_id.
   • If still blocked and blocked for over 1 hour, calls link.set_unblock(...).
   • If the domain is not resolvable, it logs a warning and blocks the link again with a DNS resolution reason.
   • If the link has an active link, it will return to work; otherwise, it will simply unblock.
   • Logs a warning if the link was blocked less than an hour ago or is already unblocked.
   • Logs an error if the Link record does not exist.