decnet/ultrix/sun problems

CURCI%FSU.MFENET at lll-mfe.arpa CURCI%FSU.MFENET at lll-mfe.arpa
Thu May 1 08:09:58 AEST 1986



I do not remember the details, but a few months ago our VAX 11/780 ULTRIX
system (connected to both DECNET and a SUN network) had some big problems.
The main problem was that the response time was terrible between the SUNs
and VAX (rlogin, ftp, rcp, rsh), and frequently caused the comlink to crash.
Typically the link would come back up by itself, but only after 10-20 minutes.
I the people at SUN were able to fix the problem over the phone.  Aparently
SUN UNIX and ULTRIX did not see eye2eye, and somebody was sending extraneous
packets.  A second bug found was with the ULTRIX rcp command.  I do not un
understand the particulars, but aparently it allocates some kind of virtual
circuit when used, that is never released.  After 30 or so rcp commands, the
system has no more virtual channels to allocate, and the DLOGIN command
(and also DCP) commands stop working, giving an error message something
like:  "INSUFFICIENT SYSTEM RESOURCES AVAILABLE"
We are still waiting for a fix on the rcp problem, and have been using FTP
instead for right now.  One interesting side effect that the first problem
had, was that our IMAGEN 300 (on the same ethernet with the SUNS + ULTRIX)
would periodically display error messages on the console, something like:
"Packet abc.def.xyz: ignoring; unknown destination;"



More information about the Comp.unix.wizards mailing list