Messages in this thread | | | Date | Wed, 6 Oct 1999 22:21:55 +0100 | From | Steve Dodd <> | Subject | Re: [2.2.{12,13p15}] PPP uploads fizzle out and die (tcpdump info incl.) |
| |
On Wed, Oct 06, 1999 at 11:05:24AM -0500, J. Scott Berg wrote: > On Wed, 6 Oct 1999, Steve Dodd wrote:
> > I'm stuck. FTP downloads over my PPP link are fine, they pretty much > > saturate the connection when the remote host can send that > > fast. Uploads are a different matter: regardless of the remote host > > I'm uploading to, the xfer rate withers away to sod-all and a huge > > gob of data remains resolutely stuck in the send-Q.
> I've seen this before, long ago. I fixed the problem by specifying an > mtu of 1500. What happened was that my ISP specified <mru 1524> in > the initial negotiation, indicating that I could set my mtu to that; > pppd did so, but the 1524 size packets were being dropped by a router > somewhere (at least that's what I surmised, and had some minimal > evidence to back that up).
OK, I looked into this and it seems some Ascend NASs will advertise a max mru of 1524 when they really mean 1500, and will throw the packets on the floor silently, without fragmenting them (or sending ICMP_FRAGNEEDED if DF is set). I /was/ talking to such a box, but fixing the mtu at 1500 locally doesn't solve my original problem.
> Your mtu appears to be at 1500 already; try setting it to 576 (I think > this is the magic "always works" number) and see if that fixes the > problem.
I tried this too and it seems to help, but I think it's masking the problem rather than curing it. I'll go grab some more info (tcpdumps and the RetransSegs count from /proc/net/snmp) and see..
-- Fortune: You will be attacked next Wednesday at 3:15 p.m. by six samuri sword wielding purple fish glued to Harley-Davidson motorcycles.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |