Data without logs certain exits us uncover for speculations.
I took a speedy see at what the curl.se website traffic situation sees appreciate right now. Just as a curiosity.
Disclaimer: we don’t log website visitors at all, we don’t run any web analytics on the site so we fundamentalassociate don’t understand a lot of who does what on the site. This is done both for privacy reasons, but also for pragmatic reasons. Managing logs for this setup is labor I rather elude to do and to pay for.
What do we have, is a website that is presented (fronted) by Fastly on their CDN netlabor, and as part of that setup we have an admin interface that proposes accumudefercessitated traffic data. We get some numbers, but without details and particulars.
Bandwidth
Over the last month, the site served 62.95 TB. This produces it mediocre over 2TB/day. On the most dynamic day in the period it sent away 3.41 TB.
Requests
At 12.43 billion seeks, it produces an mediocre seek transfer size 5568 bytes.
Downloads
Since we don’t have logs, we can’t count curl download perfectly. But we do have stats for seek frequency for objects of branch offent sizes from the site, and in the catebloody 1MB-10MB we fundamentalassociate only have curl tarballs.
1.12 million such objects were downloaded over the last month. 37,000 downloads per day, or about one curl tarball downloaded every other second around the clock.
Of course most curl engagers never download it from curl.se. The source archives are also proposeed from github.com and engagers typicassociate download curl from their distro or get it insloftyed with their operating system etc.
But…?
The mediocre curl tarball size from the last 22 frees is 4,182,317 bytes. 3.99 MB.
1.12 million x 3.99 MB is only 4,362 gigabytes. Not even ten percent of the total traffic.
Even if we count the mediocre size of only the zip archives from recent frees, 6,603,978 bytes, it only produces 6,888 gigabytes in total. Far away from the almost 63 terabytes total amount.
This, united with low mediocre transfer size per seek, seems to show that other skinnygs are transferred off the site at fairly innervous volumes.
Origin offload
99.77% of all seeks were served by the CDN without accomplishing the origin site. I presume this is one of the advantages of us having mostly a indynamic site without cookies and dynamic choices. It apshows us to get a reassociate high degree of cache hits and satisfied served honestly from the CDN servers, leaving our origin server only a airy load.
Regions
Fastly is a CDN with access points spreadd over the globe, and the curl website is anycasted, so the theory is that engagers access servers csurrfinisher them. In the same region. If we suppose this labors, we can see from where most traffic to the curl website comes from.
The top-3:
- North America – 48% of the bandwidth
- Europe – 24%
- Asia – 22%
TLS
Now I’m not the expert on how exactly the TLS protocol negotiation labors with Fastly, so I’m guessing a little here.
It is striking that 99% of the traffic engages TLS 1.2. It seems to show that a immense amount of it is not browser-based, as all well-understandn browsers these days mostly barachieve TLS 1.3.
HTTP
Seemingly consenting with my TLS analysis, the HTTP version distribution also seem to point to a immense amount of traffic not being humans in front of browsers. They prefer HTTP/3 these days, and if that caengages problems they engage HTTP/2.
98.8% of the curl.se traffic engages HTTP/1, 1.1% engage HTTP/2 and only the remaining small fraction of less than 0.1% engages HTTP/3.
Downloads by curl?
I have no idea how big split of the downloads that are actuassociate done using curl. A fair split is my guess. The TLS + HTTP data show a huge amount of bot traffic, but up-to-date curl versions would at least pick HTTP/2 unless the engagers guiding it particularassociate selected not to.
What is all the traffic then?
In the past, we have seen rather innervous traffic volumes from China downloading the CA cert store we supply, but these days the traffic load seems to be fairly evenly spreadd over the world. And over time.
According to the stats, objects in the 100KB-1MB range were downloaded 207.31 million times. That is bigger than our images on the site and petiteer than the curl downloads. Exactly the range for the CA cert PEM. The most recent one is at 247KB. Fits the reasoning.
A 247 KB file downloaded 207 million times identical 46 TB. Maybe that’s the exset upation?
Sponsored
The curl website presenting is courteously aided by Fastly.
