Page 1 of 1 [ 3 posts ] 

Shep
Sea Gull
Sea Gull

User avatar

Joined: 8 Apr 2013
Gender: Male
Posts: 219
Location: Akron, OH

25 Jan 2015, 8:47 pm

Not sure where else to post this, but for some reason, I'm trying to respond to a PM, and any time I try to preview it or save the draft, it's triggering this screen in Chrome:
Image
The page said to make a note of the CloudFlare Ray ID, which in my case is 1ae8f4f9255e10db -- I don't understand what it's trying to do, but I double-checked my BBCode and it all seems accurate, so I'm not sure what's going on. Quite confused here. I didn't use any bad words,images, or anything out of the ordinary. Is anyone else having this issue? And if so, how did you fix it? Did I use bad markup or something? :?

EDIT:
Interesting, even clicking "Cancel" is doing the same thing, ID 1ae9106f83600436 this time around. How can cancelling the formation of a PM cause a security error? Something must be off...



Shep
Sea Gull
Sea Gull

User avatar

Joined: 8 Apr 2013
Gender: Male
Posts: 219
Location: Akron, OH

26 Jan 2015, 7:34 pm

I had time, so I deconstructed the message bit-by-bit until I got to the problem area. To replicate it, copy what's in this quote below and paste it into a new PM, then click preview:

Quote:
" and making
Bam, instant security message.

Is there some kind of wordfilter we have in place that would block that kind of message from being sent? It's morbid curiosity at this point. Interestingly, removing any of the above characters leads to it working fine, so I can't think of any other explanation.



Kiprobalhato
Veteran
Veteran

User avatar

Joined: 25 Mar 2014
Age: 29
Gender: Female
Posts: 29,119
Location: מתחת לעננים

31 Jan 2015, 3:13 pm

Shep wrote:
Quote:
" and making

Bam, instant security message.

yep.

man, that's such a laughable 'trigger word'. i'm guessing the only explanation is a glitch because otherwise i have no idea why this would be happening with such lexically fundamental words.

otherwise i haven't gotten the same message, no...just captchas when writing longer PM's.

wordfilters themselves aren't bad ideas when implemented correctly, like say, keeping an 'eye out' for PMs with numerous references to sketchy pharmaceuticals much like the forum spam we've gotten (however that would be implemented), but these may as well be either spam from bots or messages from legitimate users.


_________________
הייתי צוללת עכשיו למים
הכי, הכי עמוקים
לא לשמוע כלום
לא לדעת כלום
וזה הכל אהובי, זה הכל.