jr@amanue.UUCP (Jim Rosenberg) (04/18/88)
I'm in the process of installing a new Altos 2000. As far as process loading
is concerned the system is just about unloaded -- cron and lpsched are
running, but I haven't put up any crontabs yet. I can't remember exactly, but
I think I have about 10 getty's running. On various days I have gotten what
appears to be a driver error message on /dev/console:
Out of C-Lists
I'm not completely sure yet, but I *think* this is correlated with Kermiting
*to* the Altos at 9600 baud. It only shows up *once per boot*; Kermit at the
non-Altos end has not shown me any huge error rate.
Has anybody else seen this? A friend of mine reported seeing the same message
on (I believe) a 1086 or 2086. If I had to put money on the table I would bet
(without really knowing) that this is a driver bug in which the error message
is in fact spurious and nothing is really wrong.
I'm not a kernel hacker but have read Bach and have a reasonable idea what a
C-List is. Is it possible that when I Kermit to the Altos at 9600 baud, the
driver keeps requesting new C-Lists before reusing an old one, and that when
it finally has to reuse a C-List this message is erroneously printed even
though the system is really not, in fact, out of C-Lists?
I hope it's something like this. If just running Kermit at 9600 baud to an
unloaded system can genuinely exhaust C-Lists, something pretty awful is wrong
here!
Even if the problem is only a spurious console message this is extremely
irritating. The last thing in the world I want to do is train my operators to
ignore console messages!! Is there a uucp address at Altos I can mail such
questions to??? No sense consuming net bandwidth if I can get the answer
straight from the horse's mouth.
--
Jim Rosenberg
CIS: 71515,124 decvax!idis! \
WELL: jer allegra! ---- pitt!amanue!jr
BIX: jrosenberg uunet!cmcl2!cadre! /