I too saw this message during some FTPs during my little 3c509 (non-b)
test using 1.3.72. They happened several times (up to 3 times during 1
1MB transfer, gets only I think), slowing the transfer down. But they
went away after a few minutes and I saw no segfaults.
After the tests I went back to my cheap AMD PCNET card, which beat the
3c509 in the tests by a few percent, and which has never had these
messages despite heavy use under 1.3.71 (4 days) and 1.3.72 (1 day). This
is rather unsurprising as the PCNET card is busmastering, plus has a much
bigger buffer onboard.
I also got the "socket destroy delayed" messages at boot time back in the
late 1.3.60s, but whatever they meant they did not seem to hurt anything
and they have gone away since then.
David.Niemi@mail.li.org---niemidc@clark.net---703-810-5538---Reston, VA USA
A bear won't treat you so! You're satisfied to know
When he chews you up he still respects you. -- Al Stewart