Purging specific assets not being purged at some edge locations


I’m using “Cache Everything”, with what was until today a very long TTL. The idea was that a purge would automatically be issued for affected assets when modified (the purge always occurs after origin has been updated), and cloudflare can store them for however long they feel like outside of that. This has worked for years without issue.

Today, I was approached by a user who had for some reason been receiving a cached version from the 31st of May. 10 updates and subsequent purges has happened since the version he was seeing. All other users (as far as I can tell) were correctly and instantly seeing the new fresh version for each of these updates, showing that the purge commands did go through (my logs do not show any errors in the API calls either). He however saw every asset as it were on May 31st. The request headers provided by him corroborate that he was indeed getting a HIT straight from Cloudflare, serving him a very old version of all of my assets. It appeared as if that specific edge node had simply ignored all the asset-specific purge commands.

Manually purging everything immediately fixed this issue, and I’ve since lowered Edge TTL significantly, down to its minimum at 2 hours, to somewhat mitigate it should it happen again. But I still wonder: Why did this happen? How do I prevent it from happening again?

Hi @cloudomares,

Is this generally reproducible or was it just a one-off? If you can reproduce it, a sample URL that a purge doesn’t appear to affect and /cdn-cgi/trace and the cache headers would be useful. I suspect you will need to open a support ticket with this info.

1 Like

Hi @domjh,

Thanks for your reply. The user could request the assets as many times as he wanted, so not a one-off in that regard, and the assets did apparently survive 10 purges. However, I’ve never seen it happen before, and the purge-all has fixed it for now, so I have no URL to demonstrate it on. In that sense, it does kind of seem like a one-off for that specific version of the asset.

I will send a ticket to see if CF can make sense of it. Thank you for your suggestion.

1 Like

Thanks for the extra info. Can you post your Ticker number here for any staff? Would also appreciate a message here if you get an answer on the ticket as it may help other users with a similar problem.

Ticket #2180614, and I will update you in this thread if a resolution is found.

1 Like

FYI, official response is that it is “hard to say without more diagnostic info”. I’m guessing a non-free user might get a more thorough investigation, but you get what you pay for :wink:

Thank you, yes it is difficult when the issue is not reproducible. Thanks for following up here.

This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.