Jeff Morriss
changed
bug 10620
| What |
Removed |
Added |
| Status |
RESOLVED
|
CONFIRMED
|
| Resolution |
FIXED
|
---
|
Comment # 6
on bug 10620
from Jeff Morriss
(In reply to Alberto Leiva from comment #4)
> Because of the way ICMP interacts with the contraption I'm working on, this
> is a fair solution (maybe ideal, even) for me. I'm a little concerned with
> the rationale behind it, though.
>
> By your logic, it sounds like you should also applyr36193 on other protocols
> as well, such as TCP and UDP.
> Disabling validation of inner UDP and TCP checksums when the inner packet is
> complete would wreck many of my integration tests.
>
> What are your thoughts on this?
Interesting; I would have thought TCP/UDP checksums were also disabled in ICMPs
but they're not.
I don't have time at the moment to think about it; reopening the bug so it's
not forgotten (even though the patch was merged).
You are receiving this mail because:
- You are watching all bug changes.