1 |
On Fri, Jun 6, 2008 at 6:26 PM, Kevin Faulkner <kevlar.kernel@×××××.com> wrote: |
2 |
> Miguel Sousa Filipe wrote: |
3 |
>> |
4 |
>> There is no such thing has a TCP timestamp: |
5 |
>> http://freebie.fatpipe.org/~mjb/Drawings/TCP_Header.png |
6 |
>> |
7 |
>> so, that doesn't make any sense... |
8 |
> |
9 |
> Your right and I'm wrong. |
10 |
> Its not in the header, its thrown on at the end.... |
11 |
|
12 |
...at the end of the tcp header, it's a tcp option. (I didn't |
13 |
understood that the first time I read this email). |
14 |
It might be good to disable all tcp optional headers.. |
15 |
Also there might be important to look at issues such has: |
16 |
- mtu size |
17 |
- tcp window |
18 |
- set the don't fragment flag (this can offload the routers and |
19 |
optimize the mtu for the whole connection path) |
20 |
- use the BIC algorithm (from what I've read, that's my default choice |
21 |
nowadays... but for some specific workload there might be better |
22 |
algorithms) |
23 |
|
24 |
|
25 |
|
26 |
> try doing |
27 |
> cat /proc/sys/net/ipv4/tcp_timestamps |
28 |
> |
29 |
> http://www.networksorcery.com/enp/protocol/tcp/option008.htm |
30 |
> http://www.securiteam.com/securitynews/5NP0C153PI.html |
31 |
|
32 |
kind regards! |
33 |
|
34 |
-- |
35 |
Miguel Sousa Filipe |
36 |
-- |
37 |
gentoo-performance@l.g.o mailing list |