Pierre
Pierre
You may want to edit your text and use markdown syntax (especially, you can use backquotes to protect and properly display code and program output).
@bcoles Thanks for your review. I think it should be better now.
Hi @Oborichkin! Could you please rebase against current master for the CI tests?
This PR seems to break some of the tests (apparently Python 2). Could you have a look?
Closing for no activity. Please re-open if this issue also affects the current master branch, and provide a PCAP sample that triggers the crash.
Hi, Can you sniff using tcpdump while your script is running? Do you see the same periods with no packet displayed? Also, could you provide the exact (not redacted) script...
That's interesting! Quick question: how would the IPv4 stack chose a source IP address when no route exists? Based on the interfaces's address? Thanks!
Thanks for your answer!
I'm not sure. Both have advantages. With random, you get more chance to find corner-case bugs. Without, you get stable, predictable CI. Both have their benefits and drawbacks. (I understand...
Anyway, not sure it was a good idea after all, since Python 2.7 lacks `os.PathLike`... Sorry about that :-/