Peter M's questions

I have just installed MS-SQL 2019 under Ubuntu 20.04 (the first time I have installed this Db under linux). I opened up port 1433 and can connect to it via the server IP address with Management studio from my windows laptop.

However I think that the server browser service isn't running as even though I have opened up port 1434 I can't browse to the database. I also suspect that the service may be disabled by default. I have down a lot of googling but can't seem to find linux related instructions for the browser service.

How can I check that the browser service is/isn't running? And if not, what steps do I need to take in order to get it running under Ubuntu?

I am working at a customer site commissioning a bunch of Windows 7 systems that were just delivered. The systems were imaged with a basic setup that included all of the main software needed for the customer except for a local (non admin) account that is required for this application.

I know that I can run the net user command to programmatically add the desired local account. However this command does not fully set up the local account. You still need to log into it (and see the your desktop is being prepared message) before the complete windows configuration has been performed for that account. But I don't want to do that - I just want to run my powershell script and say "Done!"

So after running net user is there anyway via command line or powershell action to trigger the process that fully fleshes out the account?

Edit

My end goal is to have access to the Startup folder for the account I am creating so that I can copy in some shortcuts. And to do all of this via a powershell script with NO user interaction.

Edit 2

And while I need the solution to work under powershell 2, I am also interested in seeing any solution that works under any later powershell version

Several years ago I built and deployed a system on XP that involved a custom written service that talked to a local 2008 SQL Express instance (that also had a CLR dependency). At the time I set up the custom service to be dependent on the Express service starting up, and all seemed well.

Fast forward to this week where my customer calls me in to debug why things aren't working properly (and apparently hadn't been for a while - a very long story).

What I saw was that when rebooting the machine, 7 out of 10 times everything came up fine. For those 3 out of 10 when it was failing, my custom service was complaining about things like:

• It couldn't connect to the SQL instance
• The SQL instance complained that the CLR subsystem wasn't enabled
• The SQL instance timed out when doing a query

But then when manually restarting my custom service (well after the computer had been rebooted), it came up error free.

So it seems that when XP says that the SQL Express system was up and running, it occasionally wasn't fully up and running, and that there is an unexpected race condition which is killing my assumptions about the XP service dependency order.

So my questions are:

What does it mean when the SQL Express instance announces to the XP service system that it is up and running?

Does it mean all components have been initialized and are ready to work, or is more of a "Hello world" statement?

Finally I also noticed that my client had loaded up the computer with several other pieces of software which probably exposed/caused the race condition that I am seeing. And yes we also started discussing what it would take to move them off XP and onto W7.

I have a client that has several geographically dispersed installations of a 3rd party commercial product that logs data to their own instances of an SQL Express database (GE's CIMPLICITY HMI if you are interested). He now wants to collect the logging data to central DB for long term storage (IE the local installations will keep a 3 month rolling log of their own data, but the central location will keep a 5 year rolling log of all data. And total data generation is probably less than 100MB per month). The table structures will be kept the same in both the remote and local DBs so that the same/similar reports can easily be run on al the data.

If these were full SQL installations then I believe that I could have used publication replication to solve the problem. But I am dealing with Express editions on the remote ends, (though I will have a full edition to play with at the central location).

So I am looking for how best to simulate publication replication on an Express DB using 3rd party tools (and not having to bang out custom code of my own). To balance the low data throughput, 3 of the installations are actually connect back home via a VPN running over DSL, so there is an expectation of link drop outs. This isn't seen as a big problem as long as the remote data eventually gets centrally logged.

(One possible solution is to convert all the express editions to full editions, but I believe that would be $$ prohibitive, and would be an administration headache to boot. So I am not even considering that at the moment)

Thanks for your help

Peter

Anyone????!?!?

Crickets

Edit

I've decided to go with a custom code solution that uses MSMQ to perform queuing on the remote side. I probably could have used SSIS to do the same thing, but for me it was jut as easy to roll my own.

I'm not sure if this a serverfault or superuser question.

A client of mine shipped me a Windows 7 box that I am supposed to work on. Unfortunately the machine is joined to his domain (which I cannot access - legally or physically) and the local administrator account appears to be disabled.

Is there anyway that I can get into the system and enable/create a local account with administrator level rights given that I have physical access? Or is my only choice to ship it back and get my client to enable a usable account and ship it back to me again?

For what its worth this is Windows 7 professional.

Edit

As prompted by Chunkyb2002's answer about caching of domain credentials I forgot to mention that there is a domain user as being the last login. Presumably this person also set this computer up and potentially has admin rights. So given this new info is it likely that if I get the password that Windows 7 will let me log in?

Edit 2

I'm in with cached credentials!

Sorry if this seems to be a stupid question, I'm not sure how to specify what I want to know when checking google.

I will have 2 or 3 devices pumping out data on a 100Base-T port. The combined data rate of all devices is about 15MB/S which exceeds the optimal 100Base-T channel capacity (12MB/S), but well within the realms of a 1000Base-T connection. Each device will be sending a burst of data in the form of an FTP transfer to a common, single host computer in a sequential manner ie:

1. Device A establishes FTP connection and transfers data
2. Device B establishes FTP connection and transfers data
3. Device C establishes FTP connection and transfers data

It may be that the A&B, B&C and C&A transfers overlap in the time domain to some extent.

There will be minimal traffic going back from the computer to each device (in general what ever is needed to support the FTP transfers), and the network will be dedicated to transferring data between these devices and the host computer.

Is it possible to use a switch to combine the multiple incoming 100Base-T streams into a single outgoing 1000Base-T stream? if so what features in a switch should I be looking for?

Or would it be better to have 3 physical point-to-point 100Base-T dedicated connections between each device and the host computer? (thus having at least 3 physical Ethernet interfaces on that computer)

Note that I can't change the interface on the devices, but I am free to choose the network and host computer configuration.

Thanks for you help

Peter

Edit Oops .. I meant MB/S not KB/S

Edit 2 I should have mentioned that all the devices and the host computer will be in close proximity - IE easily within a single Ethernet run length.

Edit 3

This is where I start pounding my head against the wall. The back story is that I am involved in a project where the device generates the data which is sent to the computer where it is collated and saved. My part is the computer backend and overall UI and display - not the generating device. Last week I saw that they were specifying 1 single device and trying to get it to pump out 15MB/S over a single 100Base-T port. I pointed out that they would need to break the laws of physics to achieve this. The problem could be parallelized by adding up to 5 more devices in order to bring the peak data rate on any single 100Base-T link to well below that links maximum throughput. This question was part of my research into how the system could be configured and still keep things in the current version of reality. I have since learnt that there is no money in the budget for additional devices. Thus the question is now moot. I have yet to see how they intend to meet requirements with only a single device. Its not really my problem, but it also does not bode well for the quality of computer they bought which will form the basis of the work I will be doing.

And yes I do read TDWTF

Thank-you to everybody who contributed to the discussion

I am in the process of trying to get a PHP PDO MySQL driver up and running under Leopard and I am basically following ideas from Tom Gidden in forcing Apache to run as i386 instead of the Leopard default of x86_64.

As per Tom's ideas I have copied

/System/Library/LaunchDaemons/org.apache.httpd.plist

to

/Library/LaunchDaemons/org.apache.httpd.plist 

And added a flag run Apache in 32 bit mode:

<string>arch</string>
<string>-i386</string>
<string>/usr/sbin/httpd</string>
<string>-D</string>
<string>FOREGROUND</string>

I have also confirmed from the Arch documentation that /Library/LaunchDemons should be processed before /System/Library/LaunchDaemons

When I reboot my mac and check the process by the Activity Monitor, the httpd process is flagged as "intel" - which is what I expect to see.

However if I open up the System Preferences->Sharing and restart Apache by Unchecking and rechecking the "Web Sharing" option, Apache starts up in 64 bit mode (as seen in Activity Monitor as an "Intel (64 bit)" process. This is not what I expected to see!

So my question are:

• Why am I seeing that starting Apache via the System Preferences result in different behaviour than via a reboot?

• Should I remove/rename the /System/Library/LaunchDaemons/org.apache.httpd.plist file, or is there a better way of doing things?

• Or am I just doing things in a plain crazy manner?

Thanks for your help.

Anyone??

I have been experimenting with ESXi and have the following setup on my desk:

Dell Laptop:

• T2500 2GHz, 4Gb RAM
• boots ESXi off USB flash drive,
• has 320Gb 7200rpm internal SATA as the datastore (purchased new for this experiment)
• single W2k3 guest OS installed on a 15Gb virtual disk and 60% is free at the moment.

(and yes, ESXi seems to be running happily on this hardware)

iMac

• 2.16GHz Core 2 Duo, 3Gb RAM
• Leopard 10.5.8
• VMware Fusion (2.0.6) running as a window in OS X, not as full screen.
• Windows XP SP3 running as a VM under VMware Fusion. (512Mb for this VM)

I am running the VMware VIC client under XP on the iMac, and the iMac and the laptop are connected by a gigabit Ethernet hub.

The problem

The problem is when I start up the VIC and open the console to the running ESXi VM image. Prior to opening the console everything seems fine - I can poke around XP and the VIC contents and do anything I want.

Once I open the ESXi VM console and click on anything inside of it the mouse starts behaving in a funny way. I can mouse over the OS X part of my system, but as soon as I mouse back over the XP VM, the mouse pointer has an affinity to the guest OS in the ESXi console in the VIC. In addition the mouse acceleration within the console also has a strange profile - but that simply might be the W2k3 setup

I cannot do anything in the XP VM except interact with the W2k3 VM in the ESXi console until I tell VMware Fusion to send a Ctrl-Alt-Del to the XP VM. This causes the ESXi VM to throw up the standard W2k3 response to a Ctrl-Alt-Del and seems to break the affinity between the mouse pointer and the ESXi console (also VMware Fusion throws up a warning dialog box about how I should be sending Ctrl-Alt-Del). Now I can mouse around the XP VM and do what I want. But as soon as I click into the ESXi console then the cycle starts all over again.

Can anyone explain this behavior or better still suggest ways to improve things?

I am dealing with some software that "wants" to be run from a system that has a desktop (call it the primary package). This software invokes other packages that also run as user level applications on the desktop (call them the secondary packages).

Most of these secondary applications are windows based programs that require their gui in order to run, and the primary package programatically pushes buttons on the secondary packages in order to achieve the desired results. One of the secondary packages is a DOS based application that uses the button pushing technique, while another can actually be invoked by OLE automation and driven directly. (For reference the secondary programs analyze data and generate reports that the primary package will email out)

The desire is to move this configuration to a VM and perhaps use something like SrvAny to shoehorn the first application into a Windows service. However I believe that when the VM starts that there will be no desktop by default, so my questions are:

1. How will the gui based programs react?

2. Will they likely crash and burn when the primary program tries to invoke them but finds no desktop?

3. Will windows automagically support the secondary programs?

4. Or do I have it totally wrong and running the VM does create a default desktop?

The guest OS will most likely be W2k3 server, hosted on ESX 3.5

BTW the reason for running headless is security policies that disallow unattended systems to be left running.

Edit

Thanks for the answers - I wasn't sure of the mechanics of the VMs virtual graphics system.

As to the security policy, in hindsight I didn't explain it in the best way possible. What I really meant to say was that an alternative solution was proposed to me as being:

1. Start the VM

2. RDP to the VM to get a desktop

3. Start the software

4. Disconnect from the VM (not logging off) so that the software keeps running in that session

The leaving a disconnected but running RDP session is seen as a no-no, not normal VM functions.

However

That does raise the question of is it possible to connect to whatever virtual desktop exists that the software will be running under? (so that I might be able to interact with it) (This is starting to sound like a "Have my cake and eat it too" request!)

I am an independent contractor and as part of my work setup I am currently using VMware Workstation to isolate test configurations and software installs between several customers. Currently I am running this on a Windows XP laptop with 3Gb of memory. My general workflow is

1. Fire up VMware when I need to do a test install
2. Clone a pre-existing base model Guest VM
3. Give it the latest updates
4. Install the test software
5. Test it on and off over the course of a project
6. Put it away for later reference

However I have decided that I want to offload this task to a separate desktop machine that will sit in my office with me. So I have a blank piece of paper on which to design a practical VMware machine that meets my requirements (both hardware and software):

Case: I want to keep the case a physically small as possible and I like the idea of a shuttle style machine. I would also want to minimize noise as much as possible. I have no experience with liquid cooling but I am willing to give anything a go. I have also briefly toyed with the idea of a rack mount solution.

I see the case supporting an internal drive for the Host O/S and two other hard drives: One for storing the VM's and the second for backing up the VMs. I also see the backup drive being mounted in a removable caddy so I can easily slot in multiple backup drives for easy offsite storage. I can also see the VM drive being in a caddy as well (but not necessary).

Motherboard/CPU: I don't know if I should be going 2 or 4 core on the CPU, but it should be capable of supporting the Host and 2 or 3 Guest O/S at the same time in a usable manner. However I do know that I want something that supports more than 3Gb of RAM for the host O/S. I also need a scattering of USB ports, gigabit ethernet and good 2d graphics support. I don't do any 3d work and I am not interesting in playing 3d games.

Drives: Good solid (but not screaming) performance. I probably need 500+Gb for Guest VM storage and another 100+Gb for collected file storage. Host O/S probably needs 50-100Gb alone. I am not sure if any sort of RAID configuration is necessary (or if so the most suitable). I assume 3.5" over 2.5" . Also is Flash a viable competitor here?

Host O/S: I am not tied to any Host O/S*, but I do know that I will need to run both 32 and 64 bit guest OS's. The Host O/S will also act as a file share for various documents etc associated with the guest VMs. ( * But I can't spring for a Mac Pro running OS X)

VMware product I only have experience with VMware workstation on my laptop. I don't know if another VMware product would better suit my needs. Or even another competitors product ;-) However I would like to be able to sling the odd Guest VM onto my current laptop and run it from there.

So who wants to help me design my server!

Edit

Budget Oops I left that off the initial question. I'd say up to US$1200 would be a reasonable one. But I want to try and keep it below $1000.