Examined one of the new features in the 9.4.7 BETA:
- Added a new Active Directory console session type
But I can't really say that what happened is a bug. When you add the AD console session, you have to tell RDM what domain controller to use. When I click on the browse button next to the "Domain Controller" text box so that I can browse computer objects on the domain, it provides me a list of all computer objects on the domain (identical to what it looks like if I was performing an import from AD). I would think that it should only show me available domain controllers and not all computer objects.
But I'd like to also make a suggestion on this feature as well that may help here. I'm not entirely sure of the internal workings on how RDM is querying AD, but it seems to be a lengthy process (it took maybe 30 or 40 seconds after I clicked the browse button to display the list of computer objects). Question is whether you can leverage Microsoft's domain design to make this process easier for you and quicker for us. When you have a Windows AD domain, you can browse the FQDN and almost instantly an available domain controller is selected for you. I know some of how the process works in AD, but maybe that type of design can be integrated into RDM (why have to choose a DC at all)? If you create an AD console in RDM, why limited yourself to a single domain controller (because it may be unavailable or unreachable), then the console session won't work? So I wonder if RDM can use the same process that Windows uses to query a domain and have Windows provide RDM with an available DC to use.
Example - if I was Wile E. Coyote (Genius) and my domain FQDN was "acme.com", I could open Windows Explorer (or a command prompt, etc.) and browse to "\\acme.com" and it will instantly show me the shares available on the domain (i.e., NETLOGON, etc.), even though I never told Windows which DC to use. Anyway, the thing is, if one DC is down, Windows will quickly redirect my request to another available DC in the domain (if there is one). If that design could be integrated into RDM, a user of this feature wouldn't limit themselves to a single DC, and they since they wouldn't need to choose a specific DC, it wouldn't take 30-40 seconds to load after clicking the browse button to list computer objects. As for performance, my speed is 1GB network cards, we have 8 (eight) domain controllers, and a few thousand computer objects in this domain. This one isn't a very large domain.
Anyway, might not be doable, or it might be too much for a single feature to implement, but just maybe something to consider.
As always, thanks and good work
edited by STGdb on 6/12/2014
- Added a new Active Directory console session type
But I can't really say that what happened is a bug. When you add the AD console session, you have to tell RDM what domain controller to use. When I click on the browse button next to the "Domain Controller" text box so that I can browse computer objects on the domain, it provides me a list of all computer objects on the domain (identical to what it looks like if I was performing an import from AD). I would think that it should only show me available domain controllers and not all computer objects.
But I'd like to also make a suggestion on this feature as well that may help here. I'm not entirely sure of the internal workings on how RDM is querying AD, but it seems to be a lengthy process (it took maybe 30 or 40 seconds after I clicked the browse button to display the list of computer objects). Question is whether you can leverage Microsoft's domain design to make this process easier for you and quicker for us. When you have a Windows AD domain, you can browse the FQDN and almost instantly an available domain controller is selected for you. I know some of how the process works in AD, but maybe that type of design can be integrated into RDM (why have to choose a DC at all)? If you create an AD console in RDM, why limited yourself to a single domain controller (because it may be unavailable or unreachable), then the console session won't work? So I wonder if RDM can use the same process that Windows uses to query a domain and have Windows provide RDM with an available DC to use.
Example - if I was Wile E. Coyote (Genius) and my domain FQDN was "acme.com", I could open Windows Explorer (or a command prompt, etc.) and browse to "\\acme.com" and it will instantly show me the shares available on the domain (i.e., NETLOGON, etc.), even though I never told Windows which DC to use. Anyway, the thing is, if one DC is down, Windows will quickly redirect my request to another available DC in the domain (if there is one). If that design could be integrated into RDM, a user of this feature wouldn't limit themselves to a single DC, and they since they wouldn't need to choose a specific DC, it wouldn't take 30-40 seconds to load after clicking the browse button to list computer objects. As for performance, my speed is 1GB network cards, we have 8 (eight) domain controllers, and a few thousand computer objects in this domain. This one isn't a very large domain.
Anyway, might not be doable, or it might be too much for a single feature to implement, but just maybe something to consider.
As always, thanks and good work
edited by STGdb on 6/12/2014