system@alchemy.chem.utoronto.ca (System Admin (Mike Peterson)) (02/01/91)
In article <1991Jan31.003842.14404@engin.umich.edu> jal@acc.flint.umich.edu (John Lauro) writes: >Has anyone experienced any TCP/IP problems on the DN10k? > <description of standard TCP/IP problems deleted to prevent net.boredom> > >Any ideas? Can anyone reproduce this problem with the 10k? >I would like to know how reproducable it is before I call it into Apollo... > >One last note... I removed the -c from tcpd on all hosts. This should be added to the FAQ file (if there is one). We have seen this problem for over 2 years, on SR10.0.p, SR10.1.p, SR10.2.p and SR10.3.p with so many patches I can't even begin to count them all. Please do call Apollo and tell them you see this problem too - what users we have left will thank you. I have more than 10 calls (dating back 1 1/2 years) on this problem, plus a few APR's. The last I heard, the problem is caused by a glitch in the VME Ethernet card, which causes it to lock up. The node then effectively disappears from the network. What relation this has to the dropped/delayed packets I'm not sure. We just removed -c from tcpd, and that makes no difference. This problem alone has caused us to buy no more Apollo workstations; we are getting SGI and IBM RS/6000 boxes, each of which has their own problems, but at least they don't disappear from the network or hang once a week (our average over the last 2 years with the DN10000). -- Mike Peterson, System Administrator, U/Toronto Department of Chemistry E-mail: system@alchemy.chem.utoronto.ca Tel: (416) 978-7094 Fax: (416) 978-8775
rees@pisa.ifs.umich.edu (Jim Rees) (02/01/91)
In article <1991Jan31.161650.2117@alchemy.chem.utoronto.ca>, system@alchemy.chem.utoronto.ca (System Admin (Mike Peterson)) writes:
... [ dn10000 tcpp bugs ]
This should be added to the FAQ file (if there is one).
It's on dabo.ifs.umich.edu. I'll add (almost) anything anyone wants to send
me. I'm always looking for additions.jal@acc.flint.umich.edu (John Lauro) (02/01/91)
In article <1991Jan31.161650.2117@alchemy.chem.utoronto.ca> system@alchemy.chem.utoronto.ca (System Admin (Mike Peterson)) writes: >The last I heard, the problem is caused by a glitch in the VME Ethernet >card, which causes it to lock up. The node then effectively disappears >from the network. What relation this has to the dropped/delayed packets >I'm not sure. That sounds worse than what we are experiencing. Right now all are home directories are on the 10000, and that could really cause problems... The only time it disappears from the net is with TCP/IP based services such as ruptime. It will sometimes drop a connection (tcpip) for no reason too. I'll take your suggestion and call the problem into Apollo. - John