On 09/18/14 15:26, Vlad Yasevich wrote:
On 09/18/2014 10:21 AM, Jamal Hadi Salim wrote:
[..]
As Michael pointed out, we can't do that.
So while i agree with the principle Michael described:
does reliability make any sense in the case where the previous event
is obsoleted by current event? There is no new information. You just
accumulated a million events for one socket all saying "the
connect failed".
I think it is a reasonable default policy.
My main concern - and i cannot claim i have ever run into this in
practise - is the case where events take over all the socket
buffers at the expense of data.
I will dig it up. But ss would be better...
I'll see what I can cobble together. Problem is that it would have to be /proc based,
but that's better then nothing.
Why /proc based? ss does direct netlink.
cheers,
jamal
--
To unsubscribe from this list: send the line "unsubscribe linux-sctp" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
