Artsize still not quite right in CURRENT
davidsen at tmr.com
Sat Apr 14 16:11:23 UTC 2001
In article <20010414141949U.kondou at inn.do.mms.mt.nec.co.jp>,
Katsuhiro Kondou <kondou at nec.co.jp> wrote:
| In article <20010414010330.A25065 at opentransit.net>,
| Fabien Tassin <fta at sofaraway.org> wrote;
| } > Worse than that, I found that sm is just hanging on some articles, not
| } > ever returning anything, and accumulating a small amount of CPU. I'll
| } > try to get a trace of whats happening.
| } same here. sm and innxmit from CURRENT are both eating all the memory
| } and spending most of their time in D state because of this.
| } strace -tt sm @0302425546303100000000AE4B9F00000007@
| There may still be a bug in updated cnfs code. Does that
| also happen with older sm?
Not that I can see. I'm using the 20010222 sm, innxmit and innfeed with
the 20010412 release (CURRENT) and it's working. Not that I like running
stuff like that, and not that I want to run CURRENT in production, but
the speed of STABLE in accepting articles doesn't keep up on my
So currently the 0222 can't save articles because of the ov size bug,
and the 0412 can't retrieve. And STABLE is a factor of two slower. This
is not a comfortable combination of choices!
Since it doesn't crash and dump, or even hang in a tight loop, it's not
easy to debug. The process just hangs, using just a little CPU. At least
it's common to all three parts, so it has to be up in the library code.
bill davidsen <davidsen at tmr.com>
CTO, TMR Associates, Inc
Doing interesting things with little computers since 1979.
More information about the inn-workers