Jon Bailey's questions

I've got a Windows 2008R2 terminal server with ~30 users on it. It's joined to a Samba-based domain. During the login script, we connect directly to the LDAP server to pull out certain profile information. This used to work just fine. Now, it doesn't, but only for non-local-admin accounts. Local admins work fine.

As a non-local-admin:

• Connection to ports 389 or 636 just terminate (wireshark on the LDAP server reveals no connection attempt)
• Connection to other ports on the same server work fine
• Same thing on multiple LDAP servers
• Windows firewall is disabled
• Can't find any other rules/policies that may block this

I suspect since this used to work, it came down during an update, but for the life of me, I can't find what.

EDIT: I just ran Wireshark on the machine and didn't see anything when connecting to the LDAP server in question (or any LDAP server for that matter).

I can, however, see traffic when I connect to that server on another port.

We're running a VDI environment with vSphere 4.0 and Oracle VDI 3.2.2 and are having a bit of a problem with users that connect to an IPSec VPN from within their VM. For some reason, once connected to the VPN, the VMWare API returns GuestInfo.ipAddress as the VPN IP rather than the primary IP of the only NIC on the system. The IP address shown in net[0].ipAddress is the correct address and is what vSphere client is reporting.

Is there any way to get VMWare tools to report the net[0].ipAddress as GuestInfo.ipAddress?

Below is sample output from the guestinfo.pl script. 172.16.1.2 is the example "bad" VPN address that our VDI software is seeing.

VMXFLEX01 guestFamily: windowsGuest
VMXFLEX01 guestFullName: Microsoft Windows XP Professional (32-bit)
VMXFLEX01 guestId: winXPProGuest
VMXFLEX01 guestState: running
VMXFLEX01 hostName: VMXFLEX01
VMXFLEX01 ipAddress: 172.16.1.2
VMXFLEX01 toolsStatus: VMware Tools is running and the version is current.
VMXFLEX01 toolsVersion: 8194
VMXFLEX01 Screen - Height: 600
VMXFLEX01 Screen - Width: 800
VMXFLEX01 Disk[0]: Capacity 42935926784
VMXFLEX01 Disk[0]: Path : C:\
VMXFLEX01 Disk[0]: freespace : 33272619008
VMXFLEX01 net[0] - connected : 1
VMXFLEX01 net[0] - deviceConfigId : 4000
VMXFLEX01 net[0] - macAddress : 00:50:56:95:1f:c9
VMXFLEX01 net[0] - network : VM Network
VMXFLEX01 net[0] - ipAddress : 10.0.0.2

For some reason, after users do the initial authentication, the system logs in (or unlocks) and locks the screen again within a few seconds (without dropping the RDP session). I've seen this on XP SP3 and Server 2k3. Clients are both major versions of the MS client, and rdesktop.

Update: We've seen the issue on a range of machines from 2k3 server (no logon/post logon scripts on servers) to XP SP3. Multiple machines have the same issue - even coming from a fresh and fully patched install.

I'm running a ISC DHCPd server for my network serving a number of subnets. One of the things I would like to do is assign a specific range of IPs to hosts with a common MAC prefix (ex. 00:01:02). Also, the assignments have to be able to be over-ridden by assignments with fixed-address. I've googled for it, but haven't found anything definitive.

Bonus if I can put the statement within a subnet stanza of my dhcpd.conf (it would fit better with my management software).

We're looking at implementing a VDI solution where I work. Currently, we run a mostly Linux shop with Windows clients (Samba PDC too). We've been evaluating a VDI solution for use and was wondering what "real use" comparisons people have done with what I consider the three main VDI solutions out there. There is scant information available on the web when it comes to comparing these solutions.

What are your experiences?

Assume the following drive setup (with ZFS):

Controller 1    Drive1    Drive4    Drive7    Drive10
Controller 2    Drive2    Drive5    Drive8    Drive11
Controller 3    Drive3    Drive6    Drive9    Drive12

VDEV setup:

vdev1: drive1, drive2, drive3 vdev2: drive4, drive5, drive6 vdev3: drive7, drive8, drive9 vdev4: drive10, drive11, drive12

Is it better (for reliability) to add all vdevs to the same zpool, or have separate zpools (one for each vdev)? Also, if we lost a single vdev, would we lose the entire array? We don't need all the storage capacity in one place - the smaller zpools would be find from the storage standpoint.

Update: For 3dinfluence's question about the boot pool, that will be on a RAID1 set. I don't like mixing the OS and my multi-terabyte RAID arrays.

We're in the process of building a new SQL server, and would like some recommendations on hardware. Currently, we build all systems in house. With the information below, what would you recommend for drives and an external SAS cage. Also, in this workload, how beneficial would SAS be vs. SATA2?

Basic specs so far:

• AMD Phenom II X4 955 CPU
• 8GB RAM
• 2TB RAID6 storage (thinking 450GB Seagate Cheetah 15K.6 drives)
• 1TB RAID1 storage (why below) (1TB Seagate Barracuda ES.2)
• 3ware 9690SA SAS Controller

Software:

• Windows 2003 or 2008 server (haven't yet decided)
• SQL Server 2005 or 2008 (will depend on application compatibility)

Workload:

This server is fairly read/write heavy for most applications (about 1TB) (financials, internal apps, etc). We also have GIS data on the server that is used by ArcGIS. Our plan is to have our vector data on the main RAID6 array with the raster data (about 750GB) being on the RAID1 array (since it's use isn't as frequent).