arla-0.35.9 FreeBSD crash

Harald Barth haba at pdc.kth.se
Tue Sep 3 13:39:00 CEST 2002


> At a guess, perhaps what happened here is that the volume got moved
> between the collectstats_start() and _stop() calls?

The code seems to make the assumption that the host of the connection
matches the one in the volume entry. That was not the case, so the
assert should go. The question is if it is enough to just ignore that
data point in the statistics or if we should do something more with
the volume entry now that we know that the host has changed.

Harald.





More information about the Arla-drinkers mailing list