Explanation of bugs

[snaslund]

Can someone from HP clarify the two bugs cited below for E-series controllers?  These are from the 5.5.3 release notes.

102860 Clients roaming to untagged VLANs cannot communicate with the network

This is a very serious issue that should have prevented the software release if it is as general as the text implies.  Are there more specifics on when this happens?  This will prevent us from upgrading to this release or any other that contains this bug unless it is narrower than it sounds.

53704 Wireless clients fail to be authenticated when the controller is configured to use an
Active Directory server with a NetBIOS name that differs from the Windows domain
name. To avoid this problem, make both names the same.

Are you guys working on this or do you assume we should just change the domain name?  Just to let you know, this is a very common domain configuration especially on older, larger networks because Windows domain naming has different naming requirements than NetBIOS did.  Often both names must continue to exist for quite a long time due to legacy applications and servers that are using the NetBIOS name instead of the Windows domain name.  A question to HP would be why do you care if they are different and why do you even need to use both?  This is another bug that will prevent users from moving forward off of buggy software.

You have two other undocumented bugs that I have reported several times and keep being told is not a bug or cannot be duplicated. 

1 . The issue is with AD authentication on teamed controllers.  There is only the web option to join the domain only at the team or team manager controller levels.  The problem is that any non-manager controllers are NOT joined to the domain and any access points that are connected to the non-manager controller cannot authenticate off AD.  The work-around is to join the non-manager controllers from the CLI interface.  Then all users authenticate just fine.

2.  It seems that after every controller reboot, it is necessary to rejoin the controllers to the domain.  We don't understand why this is the case.  Most of the time the controllers web interface shows them as joined when they actually are not.  Entering the data and rejoining the domain seems to fix the issue.  This requires us to intervene every time a controller gets bounced which is not good.

I know you are going to tell me to contact support again.  The reason you are seeing the post here is that I am tired of HP support telling me you "never heard of that" when I report issues that later show up in release notes or on this forum.  It is really frustrating to be ignored and then see that many other people have experienced the same thing.

Steven Naslund

[ISoliman]

HP is aware of the untagged VLAN issue and they are working on it, the other issue i'm not sure, you can simply log a case with the support and if they say anything ask for a business escalation ;)