On Jan 10, 2008 12:14 PM, Mike Fischer <email@hidden> wrote:
Like the header of the crash log says: 10.5.1 (9B18)
Ah yes.
Should I report this as a bug on BugReporter?
I would *always* report bugs in Apple's frameworks, even if you have a
workaround, or you know it's already been reported by someone else.
It took me a while but I have finally decided to report this as a bug:
<rdar://5702001>
At first I thought my code was somehow at fault. But after completely
rewriting the code to use the lower level NSUrlConnection instead of
NSURL resourceDataUsingCache: I am still seeing exactly the same type
of crashes. I'm usually not seeing them right away. In fact often it
takes a couple of hours and several thousand downloads to trigger the
bug. But I am now pretty sure that it's not my fault.
To me it looks like a multi-threading/multi-access bug in the Apple
Keychain access code. My app querying the Keychain very often only
increases the likelyhood of triggering the bug I believe.
BTW: There is one workaround: Including the name and password in the
url (http://<name>:<password>@<host><path>) will prevent the Url
loading system from asking the Keychain, thus avoiding the bug.
Thanks for you comments!
Mike
--
Mike Fischer Softwareentwicklung, EDV-Beratung
Schulung, Vertrieb
Note: I read this list in digest mode!
Send me a private copy for faster responses.
