Messages in this thread |  | | Subject | Re: [ircu-development] Slow on high-MTU (local host) connections? | From | Entrope <> | Date | 09 Nov 2001 11:04:23 -0500 |
| |
Following up on this: Carlo and I both tried disabling Nagle on the sockets (both ends). We still saw the same slow sending pattern.
On one end (the "ircu" process in Carlo's original mail), the send and receive buffers (SO_SNDBUF, SO_RCVBUF) had been set to 8 KB each. If these are increased to be larger than the loopback MTU, it makes the problem go away. (On the other, "srvx" end, the buffer sizes were kept at the default.)
Is this expected behavior? If so, why? (The references I've found on the web suggest that the send and receive buffer sizes would only come into play if Nagle were still enabled.)
-- Entrope - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
|  |