Search Support

Avoid support scams. We will never ask you to call or text a phone number or share personal information. Please report suspicious activity using the “Report Abuse” option.

Learn More

Do "Race Cache With Network (RCWN)" requests identify themselves to servers?

  • 1 cavab
  • 1 has this problem
  • 3 views
  • Last reply by cor-el

more options

This question was asked a while ago: https://support.mozilla.org/en-US/questions/1267945 asking what the "(raced)" values were inside the Transferred column of the Network tab in the Developer Tools.

In short, it sounds like when the browser detects a slow pull of a value from the local cache, it "races" the cache by actually making the network request to see which comes back first to provide the fastest loading possible for the end user.

While conceptually I totally get this, as a Web Developer I'm now curious if this particular network request identifies itself in some way via HTTP Header values?

We actively seek out customers of our web applications to see which ones have "bad" cache settings (no cache, too small a cache, or flushed constantly) to reach out to them to resolve this (and/or find out the reason why they were disabling their cache). TL;DR - I think these RCWN requests may give us a false positive when trying to detect users with bad cache settings... so I'd like to be able to "ignore" these RCWN requests in our analysis (I don't want to block or otherwise interfere with these requests).

On a secondary note, is there any info for the end user on why a given resource was raced? It appears that all of the "(raced)" resources I'm seeing on my frequently visited sites are less than 2.5KB (often under 1 KB) - just wondering if the algorithm to detect "slow" resources is properly scaled for small resources.

This question was asked a while ago: https://support.mozilla.org/en-US/questions/1267945 asking what the "(raced)" values were inside the Transferred column of the Network tab in the Developer Tools. In short, it sounds like when the browser detects a slow pull of a value from the local cache, it "races" the cache by actually making the network request to see which comes back first to provide the fastest loading possible for the end user. While conceptually I totally get this, as a Web Developer I'm now curious if this particular network request identifies itself in some way via HTTP Header values? We actively seek out customers of our web applications to see which ones have "bad" cache settings (no cache, too small a cache, or flushed constantly) to reach out to them to resolve this (and/or find out the reason why they were disabling their cache). TL;DR - I think these RCWN requests may give us a false positive when trying to detect users with bad cache settings... so I'd like to be able to "ignore" these RCWN requests in our analysis (I don't want to block or otherwise interfere with these requests). On a secondary note, is there any info for the end user on why a given resource was raced? It appears that all of the "(raced)" resources I'm seeing on my frequently visited sites are less than 2.5KB (often under 1 KB) - just wondering if the algorithm to detect "slow" resources is properly scaled for small resources.
Attached screenshots

All Replies (1)

more options

Note that users can look at the about:networking#rcwn page to see some RCWN stats. I would assume that this is a normal request, so I think there is no way a website can detect the cause of the request.