Repeated purge_cache_queue_write errors

Hi there,

We’re experiencing an issue with Super Page Cache that becomes very visible during load/stress testing, and we’d really appreciate your guidance.Issue we’re seeing

When running a stress test, after around 1 minute we start seeing the following log entry repeating continuously:

[cache_controller::purge_cache_queue_write] Queue file not writable. Sleep 1 second

This line repeats many times per second and appears to coincide with increased CPU usage on the server.

We have attached:

• The relevant debug log showing the repeated messages https://share.drijen.digital/rRux5K1J
• The purge_cache_queue/cache_queue.json file at the time this occurs https://share.drijen.digital/4gup4vZ7

What seems to trigger it

From our investigation, this appears to be triggered when tickets are created via The Events Calendar / Event Tickets plugin.

Specifically, URLs of the following pattern are involved:

/?post_type=tribe_rsvp_attendees&p=90910

(where 90910 and similar values are ticket / RSVP IDs).

Each time one of these tickets is created, Super Page Cache appears to:

• Trigger a cache purge
• Attempt to write to the purge queue
• Enter a retry loop when the queue file is not writable

During stress tests (with many concurrent requests), this behaviour becomes continuous.What we’re trying to achieve

We would like to stop Super Page Cache from purging cache when these RSVP/ticket-related URLs are created or updated.

Ideally:

• The tribe_rsvp_attendees post type (and its query-string URLs) should be ignored by SPC
• No cache purge or queue write should occur for these URLs, as they are dynamic and not meant to be cached anyway

Our questions

1. Is there a supported way to exclude a specific post type (e.g. tribe_rsvp_attendees) from triggering cache purges?
2. Is there a setting or filter to disable purge-on-create / purge-on-update for this post type?
3. Is SPC expected to retry queue writes in a tight loop like this if the queue file cannot be written, or is this behaviour indicative of a configuration issue?

We have already temporarily disabled the preloader, which helped reduce background load, but this purge queue behaviour still occurs under stress.

Any guidance on the correct configuration for high-volume, dynamic post types like Event Tickets would be hugely appreciated.

Thanks in advance for your help,