On 12/09/2014 03:43 AM, Per Jessen wrote:
It's more a case of - programmers test their code to see that it works, testers test the code to see where it breaks. It's all about one's perspective.
Even so, a tester can't test everything. Back when I was at IBM, the first time, I tested desktop systems, to ensure everything worked properly. If I ever came across a problem, I added it to my list of things to be checked next time around. However, I find I often push things beyond common use. As an example I recently came across a problem with a WiFi access point. I have had this AP for almost 3 years and it worked well with both IPv4 and IPv6. A few months ago, I added a 2nd SSID for guests, which used a VLAN and only connected users to the Internet. That worked well with IPv4. A couple of weeks ago, I decided to add IPv6 to it. I soon discovered that IPv6 would fail as devices were getting router advertisements from both the main and guest subnets. After some testing, I discovered that the access point was allowing a lot of broadcasts & multicasts from the native LAN to the 2nd SSID. I would imagine that the original tester only verified VLANs and additional SSIDs worked with IPv4 and not IPv6. I've also spent the past few days educating the support person on how VLANs worked. He seemed to think it was normal for native LAN traffic to appear on a VLAN. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org