Cloudflare, Litespeed, crawlers, and Google core web vitals

crawl with cloudflare cdn activated, is this normal?

it could , if CF blocked the crawler requests , please try go to General -> server IP , set your server IPv4 , then go to crawler -> sitemap -> drop domain to ON , this should force crawler to directly access to your origin IP

How does the cache get updated when the Instgram plugin updates and cloudflare integration?

if the instgram plugin does things like a post edit , then LSCWP will have hook fired when post status changed , LSCWP does not touch anything on CF

if it’s doing iframe thing , then the real content is still on remote , you don’t need to purge anything from your site.

Should I leave quic.cloud can on/off?

there isn’t really much necessity to enable double CDN

The cloudflare plugin, should that be utilized?

Should cloudflare page rule “cache everything” be implemented?

Should cloudflare APO be considered for my application ?

some LSCWP feature , such as guest mode , JS minify , JS combine , CSS minify , CSS combine , mobile cache, webp replacement …etc will be broken with CF page cache , either done by cache everything , or by APO

Should I complete integration in quic.cloud?

at moment of this replying , you have fully integrated QC already

@qtwrk, thank you for your reply.

Ive turned off Cloudflare CDN and using cloudflare DNS connected through API in plugin and also cloudflare “integrated” in quic.cloud.

I delayed and defer JS with Perfmatters. My page scores are so good with litespeed that I dont think I need any of the Quic.cloud based services. Even mobile cache and guest mode seem to not provide any significant benefit as my Core web vitals suggest near 100%.

I do not have unique mobile content. So I turned off mobile cache and left guest mode on because speed vitals, page speed insights, and lighthouse did not show much benefit to having mobile cache activated.

Is this a logical setup?

With a growing audience internationally, should I look at using the Cloudflare CDN? Or is the DNS for cloudflare with quic.cloud CDN going to provide equal speed benefit in pursuit of LCP internationally?

I thought it may be prudent to go the cloudflare CDN direction, just not sure who much upside there is, above the cloudflare DNS integration. And I know using cloudflare will likely introduce some problems to be troubleshot.

Thank you in advance for any suggestions.

• This reply was modified 1 year, 8 months ago by RL.

yes, that sounds logical , mobile cache is only needed when you have some kind of mobile-specific content

frankly each site is unique to others, so my experience probably does not always make sense or give best result , thus I would suggest you to try different approaches and use the one you see as best result

@qtwrk thank you for your continued feedback. ?? My TTFB shows 500-700ms on the first pull, and 2nd pull immediate after is <100MS domestic and internationally. (Speed vitals multi locations).

Crawler/crawl interval is set at 10,000 seconds with hopes that it rapidly see updates with Instagram feeds, and MOST important keeping cache warm for MOBILE speed on 1st visit.

Is the 10,000 second crawl interval hurting site speed for visitors first page load on first visit?

What can I do to make sure mobile visitors 1st page load on a first visit is as fast as possible with a warm cache?

report # PMXJHQJW

10.000 means the crawler will run every 2.7-ish hours , if you don’t have resources restriction , this should suffice

probably nothing else than crawler

Ok. But what about the fact that when testing 10 locations the TTFB is 800ms average on first pull, and then immediate after that the average drops 1000% to 80 milliseconds.

How do I get the cache to hit on the first try?

unfortunately you can’t

the first access is that the cache is stored on your origin server

so browser -> CDN -> origin cache = 800 ms

on repeated access from same tester , the cache will be stored on CDN node on previous request, so it’s like

browser -> CDN cache = 100 ms

but at this moment , crawler on CDN node directly is not supported by plugin , you can use “server IP” to set crawler to run on one sepcific node , but it won’t help others

cache sync and replication is on the to-do list though

if you known thing or two about program , then certain you can do a script or something to request through the CDN nodes, but this will multiply your bandwidth usage