On Thu, May 24, 2012 at 08:37:52PM -0500, Jimmy Hess wrote: > On 5/24/12, not common <notcommonmista...@gmail.com> wrote: > [snip > > I am looking for some guidance on full packet inspection at the ISP level. > Aside from any legal issue; there is a "respectable practices" > issue. Even if there is no regulation that prohibits something does > not mean it is OK. Your customers' deserve to be made aware of any > full packet capture practices that may impact traffic to/from network > they own/manage, before packet capture occurs, especially when there > is data retention, or human examination/analysis based on contents of > large numbers of packets; otherwise there is a risk you will be in > trouble, for some definition of "in trouble" that depends on the > circumstances. > > Because your packet interception can put your user at risk; > proprietary information can be disclosed. And most ISP customers > intend to purchase network connectivity service, not "record all my > traffic without telling me" service ..
If you need a call center to handle this just let me know... :) since your call volume is going to spike through the roof. > > > > Are you prepared to explicitly explain to your customers, both > existing, and new ones, > before they are allowed to buy or continue service from you -- under > what circumstances > you intercept full packets, whose packets do you capture, what > packets do you capture, how many packets / how long will you capture > their packets, what do you do with their contents after you capture > them, how long do you keep data, what security controls do you have > in place to prevent unauthorized access to their packets and > ensure timely destruction of sensitive data? > > > If the answer is NO, that you have poor planning, or your privacy > practices are not solid enough to reveal to your customers with > confidence, then save the money on consulting lawyers, by choosing > NOT to implement interception and capture of full packets. > > > > Is there any regulations that prohibit or provide guidance on this? > -- > -JH -- - (2^(N-1))
