Newsgroups: comp.sys.apollo Path: utzoo!utgpu!news-server.csri.toronto.edu!helios.physics.utoronto.ca!alchemy.chem.utoronto.ca!system From: system@alchemy.chem.utoronto.ca (System Admin (Mike Peterson)) Subject: Re: tcpd problems Message-ID: <1991May24.194226.27593@alchemy.chem.utoronto.ca> Organization: University of Toronto Chemistry Department References: Distribution: comp Date: Fri, 24 May 1991 19:42:26 GMT In article neilb@spectrum.cs.unsw.oz.au (Neil Brown) writes: > >We too have had a number of problems with tcpd spinning. >It invariably involves the mbufs being all used up, as in: > >335/336 mbufs in use: > 111/112 ( 80-byte) mbufs used/allocated > 112/112 (1560-byte) mbufs used/allocated > 112/112 (9216-byte) mbufs used/allocated Our DN10000 has always shown 162/336 mbufs in use: 112/112 ( 80-byte) mbufs used/allocated 50/112 (1560-byte) mbufs used/allocated 0/112 (9216-byte) mbufs used/allocated 1187 Kbytes allocated to network (7% in use) 0 requests for memory denied or even 257/336 mbufs in use: 112/112 ( 80-byte) mbufs used/allocated 111/112 (1560-byte) mbufs used/allocated 34/112 (9216-byte) mbufs used/allocated 1187 Kbytes allocated to network (40% in use) 0 requests for memory denied for the last 2.5 years, which may go part way towards explaining its TCP hesitations/delays/hangs (the 0 requests for memory denied is probably because it has just been rebooted after the latest hang). We regularly have 16 telnetd/rlogind sessions active at once, plus X Windows, BSD printing, NFS, rwhod, named, sendmail, NNTP (news), and the Cray satellite station. We are also getting approximately 5 VME Ethernet card resets a day (we are running SR10.3.0.6 which has a patch to kick the VME Enet card when it hangs) which seems to cause TCP sessions to be dropped at random, especially rlogins from other Apollos. Mike. -- Mike Peterson, System Administrator, U/Toronto Department of Chemistry E-mail: system@alchemy.chem.utoronto.ca Tel: (416) 978-7094 Fax: (416) 978-8775