On 03/ 6/12 11:17 AM, [email protected] wrote:
On Tue, Mar 06, 2012 at 10:08:00AM +1300, Tim Foster wrote:
I've got an incremental, and new webrev at:
https://cr.opensolaris.org/action/browse/pkg/timf/poisoned-cache-v2/
On lines 1286-1298 and 1887-1899 you're not recording the fact that
you've seen a content error, which is going to skew the repository
statistics for future choices.
In general, I'm not really thrilled with this approach. It might be
simpler to set the cache-clearing headers on any subsequent attempt to
connect to a repository that has already given us a content error.
Ok, I've had another go at this, and have a webrev at:
https://cr.opensolaris.org/action/browse/pkg/timf/poisoned-cache-v3
I'm still trying to track down an intermittent problem with the new test
I'm introducing here, where Apache seems to not cache a url fast enough
when we're running lots of tests at the same time[1], resulting in the
cache returning good data when we're expecting corrupt data (I'm using
urllib2 as the client to verify the data is corrupt, so it's not the pkg
code at fault here)
Apart from that though, comments welcome,
cheers,
tim
[1] it's possible it hasn't had a chance to flush its cache to disk, and
so my code to corrupt the cache doesn't take effect
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
