Forum Moderators: open

Message Too Old, No Replies

Hits from Macintosh PC's with no browser in the user agent

         

SumGuy

11:09 pm on Aug 2, 2023 (gmt 0)

5+ Year Member Top Contributors Of The Month



I've noticed today a web hit (human, not bot) where the UA was:

Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko)

No browser. I thought that was strange. I've scanned through the logs looking first for any with "Macintosh" and then removing any with "Safari" and "Firefox". I didn't have to remove any with chrome - because there were none? This results in hits from 14 unique IP's such as from EDU's (Queensland, Missouri, UC Irvine, Providence, Japan) and ISP's (Sunrise GMBH, Brit Telecom, Centurylink, Verizon). Earliest dates to Feb / 2021. These look like legit hits, some are to PDF files, others are page-browsing. In one case, a browsing session requested /null (which got them a 404).

The vast majority have the UA as I've stated above, sometimes the OS X string was 10_14_4 or 10_15_6. Otherwise there was one example of this:

Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_5) AppleWebKit/603.3.8 (KHTML# like Gecko)

And one example of this:

Mozilla/5.0 (Macintosh; Intel Mac OS X) Word/0.0.0

I just thought it strange not to see a browser. Could these have been hits referred to from an embedded link in an email, document or spread sheet? With the application not mentioned for some reason?

dstiles

10:13 am on Aug 3, 2023 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member Top Contributors Of The Month



KHTML is an outdated linux KDE implementation. It's possible Konqueror still uses it. Search for KHTML

AppleWebKit is, of course, a common web browser engine.

Pfui

4:41 am on Aug 7, 2023 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member Top Contributors Of The Month



Did you look at hits immediately after the no-browser instances? A quick skim over here shows what appear to be phone-related four-hit sets:

Example A - Australia - subsequent seconds:
1.) /
Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko)
2.) /apple-touch-icon-precomposed.png
Safari/18615.2.9.11.10 CFNetwork/1408.0.4 Darwin/22.5.0
3.) /
Mozilla/5.0 (iPhone; CPU iPhone OS 16_4 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/16.4 Mobile/15E148 Safari/604.1
4.) /apple-touch-icon-precomposed.png
Safari/15613.3.9.1.16 CFNetwork/1128.1 Darwin/19.6.0 (x86_64)

Example B - Florida - subsequent seconds:
1.) /
Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko)
2.) /apple-touch-icon-precomposed.png
Safari/15613.3.9.1.16 CFNetwork/1128.1 Darwin/19.6.0 (x86_64)
3.) /
Mozilla/5.0 (iPhone; CPU iPhone OS 14_0 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/14.0 Mobile/15E148 Safari/604.1
4.) /apple-touch-icon-precomposed.png
Safari/18615.2.9.11.10 CFNetwork/1408.0.4 Darwin/22.5.0

No clue as to the rapid file redundancy plus four UA changes on the fly. Collecting favicons and such for iCloud-connected bookmarks?

SumGuy

12:09 pm on Aug 8, 2023 (gmt 0)

5+ Year Member Top Contributors Of The Month



I will have a look at what hits happened around the same time. The phenomena I describe below was gleaned from logs where I was looking for hits from a specific IP neighborhood - no such UA-changing is happening from the same IP and no other hits from the neighborhood.

I was going to post the following as a new thread, but since it features this same UA without a browser, I'll tack this on here. I'm seeing an odd file-request pattern from the same IP. The file being requested is a pdf (about 250 kb).

The request happens as 7 individual, sequential requests. Six of these are http code 206 (requesting part of the file) but one is a 200 (requesting the entire file). The 6 requests for part of the file are always requesting the same byte-size sequence, in KB the sequence is 1, 0 (yes - zero), 16, 2, entire file, 136, 32. This happens over the course of a few SECONDS.

I believe that when someone views a PDF file in a browser window, the browser will usually download chunks of the file as the user scrolls through the document, hence the HTTP partial file request (code 206). What I can't figure out here, is why am I seeing this same sequence of partial requests happening from the same IP on different days?

Twice in January, 6 times in June, 6 times in July, and so far 4 times this month. There was one request in June that was actually a 304. (?)

Always the same user-agent:

Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko)

No referrer, and no request for favicon.ico (I often see a request for a pdf file paired with a request for favicon). No previous hits from this IP or the /16 neighborhood. No browsing of my actual website - these are direct hits to the same PDF file.

The extended response headers contain this:

application/xhtml+xml
application/xml;q=0.9
*/*;q=0.8
en-US
en;q=0.9
gzip, deflate, br

Is this someone that keeps multiple tabs open, and each time they restart their browser all content is re-downloaded - instead of cached?

The IP is 67.6.98.X (67-6-98-x.clma.centurylink.net). I have issues in general with some centurylink hits, I think some of their IP's are used as a VPN or web proxy.