Hello,
There were 12 contiguous iterations when I first reported the
problem. Given the large loop, there was a stopwatch period of time
between iterations printed to the terminal window (seemed fairly even
in duration during output to terminal window; but, I was not
timing). In the sample I pared it down to three. I tried sending a
log file with the -d in the command that ended up over 324 kbytes in
size (exceeding the 150 kbyte limit) containing 14 iterations before
I hit the "control C" and that file choked at access to the list.
The iterations may have been testing dependencies each time; but, the
file was getting quite large and the run time taking quite long.
Yes, I agree with you on the possibility of not being unbounded;
but, the lack of evidence is not running long enough to find out.
Given the corrections to the process via the "sudo port sync"
command, it is unlikely that I can repeat the problem for another
run. We can file this as something to watch for in the future.
Frank
On Dec 16, 2008, at 12:23 AM, Joshua Root wrote:
Frank J. R. Hanstick wrote:
Hello,
It is quite possible that the only way into the infinite loop
is via
a port failure under certain circumstances. Problems usually
arise when
anomalies occur and not when things run as expected because not all
anomalies can be planned for. If not, sorry to disappoint you.
Fair enough, but I would have expected to see a large number of
iterations given the command you used, and we have no evidence that
the
number of iterations would have been unbounded and not just large.
- Josh
_______________________________________________
macports-users mailing list
macports-users@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macports-users
