On a recent project we were having problems configuring the Windows SharePoint Services Adapter onto a SharePoint 2010 farm. This is basically a BizTalk 2010 web service that gets deployed to your SharePoint farm. It allows BizTalk to deposit documents into SharePoint document libraries.

The adapter was deployed and configured without error, but when calling the web service from BizTalk it generated HTTP 401.2 Unauthorized errors. In the Windows System Event log on the BizTalk server the following details were logged:

The adapter “Windows SharePoint Services” raised an error message. Details “The Windows SharePoint Services adapter runtime does not have permissions to invoke the adapter Web service. In order to fix this issue, you have to add the DOMAIN\biztalk_service_account Windows account to the “SharePoint Enabled Hosts” Windows group on the Windows SharePoint Services machine. This operation will allow BizTalk host instances running under DOMAIN\biztalk_service_account Windows account to invoke the adapter Web service in order to send and receive messages to or from SharePoint sites. The group membership will not take effect until you restart the BizTalk host instance.

When you configure the adapter it creates a local security group on the server which is named “SharePoint Enabled Hosts”. The above error indicates that the service account that the BizTalk process is running under needs to be added to this local group. We tried this and tried this and tried this. After several unconfiguration, reconfiguration cycles we were stuck with the same error.

This appears to be a pretty common problem. A lot of people griping about the local security group approach (justifiably!).

Richard recommends modifying the SharePoint web application – we tried that approach and it didn’t work for us.

What solved this in our circumstance was to manually edit the web.config file for the virtual directory that is hosting the BizTalk web service.

The original authorization block looked like this:

<authorization>
<allow roles=”SHAREPOINTWFE\SharePoint Enabled Hosts” verbs=”GET,HEAD,POST”/>
<deny users=”*”/>
</authorization>

Instead of stuffing around with the local group, I changed the allow line to specifically use the BizTalk service account (a domain account), and also removed the deny tag, so it looked like:

<authorization>
<allow verbs=”GET,HEAD,POST” users=”DOMAIN\biztalk_service_account” />
</authorization>

This fixed it for us, no more authorisation problems. It means that the BizTalk service domain account is still the only account that can invoke the web service.

If you have multiple web front end servers in your SharePoint farm, you need to configure the adapter and apply this ‘fix’ on each. Your mileage may vary on this, I’m just sharing what worked in our situation.

This is implemented with the List Component for SharePoint. This is a sandboxed solution that is deployed to the target SharePoint site. When activated it allows CRM 2011 to create documents in SharePoint and also to present those documents to the user in a CRM native UI.

In my experience this component can be tricky to activate on the SharePoint 2010 side. I have seen errors that range from ’401 Not authorized’ to ‘the Sandboxed Code Host Service was too busy to handle the request’

This has usually been caused by the fact that the SharePoint machine does not have direct access to the internet. Durng the activation process SharePoint (well the .NET Framework) is trying to contact the host crl.microsoft.com. Because this cannot be contacted within the allowed timeout period activation fails.

Tbe solution is to ensure that the SharePoint server can contact this site. If that is not possible (security reasons, working on a standalone machine etc) then you can add the
following entry to the HOSTS file on the machine:

127.0.0.1   crl.microsoft.com

This ensures that the address resolves and then the process works. You can then activate the solution.

For more details on this workaround see my earlier blog post.

Microsoft SQL Server named instances use dynamic TCP ports and rely on the SQL Browser service to provide the port number to the client in order to establish a connection. In a lot of cases I have seen people trying to connect to their new SQL Server named instance (e.g. \sql01) and assuming that SQL Server was listening on port 1433. They open the Windows Firewall to allow inbound TCP traffic in on port 1433 but still cannot connect. Some even specify the port in the connection string if they can.

There are two options – rely on the SQL Browser service or change the named instance to operate on a fixed port.

Using the SQL Browser service
You need to be able to connect to the SQL Browser service (which is often not running) which operates on UDP port 1434. Even then, if your local Windows Firewall on the server hosting SQL Server does not allow inbound traffic on whatever TCP port is being used, then you won’t be able to connect remotely.

Make sure the SQL Browser service is running also, which you can configure in the SQL Server Configuration Manager.

Configuring SQL Server to listen on a fixed port
On the SQL Server, open up the SQL Server Configuration Manager, and navigate to SQL Server Network Configuration > Protocols for <named instance> and right-click on TCP/IP and select Properties

In that dialog, click on the IP addresses tab. You can scroll to the bottom and clear the value for TCP Dynamic Ports and specify a fixed port of your choice.

The only bugbears for me have been the WiFi dropping out in my house. The tablet seems to have the weakest radio of all the other WiFi devices in my house.

The only issue to complain about would be the insane rate of fingerprint collection on the screen. I’ve recently bought a Samsung Galaxy S II and my wife has an iPhone, so I have a new appreciation of the benefits of Gorilla glass which seems to resist fingerprints really well.

Apart from that it’s been great. I take it to meetings and take notes, I spend a lot of time browsing, Tweeting and Facebooking via this device. I love the form factor and the ‘always on’ concept – so I hate to admit it, but I’d be happy with an iPad in that regard too.

Does your server have access to the internet? No? This will cause delays when a signed piece of .NET code runs

Every time some managed .NET code tries to run, the .NET Framework is trying to contact crl.microsoft.com. Now if your server is offline (think VM on a laptop on a plane), or your server doesn’t have access to the internet at all, or possibly through a proxy server, then these attempts to contact crl.microsoft.com will add a delay to start up time of things like SharePoint web apps as the connection attempt times out.

This affects all .NET apps since version 2.0 of the Framework. I’ve seen weird things trying to activate a CRM 2011 SharePoint List Component that just wouldn’t work until we applied one of these workarounds.

Workaround:

Hack your hosts file to add an entry for 127.0.0.1 crl.microsoft.com

Use this to quickly validate if this is part of your performance problem.

If doing this decreases your startup time (i.e. saves you about 90 seconds) and you want to make a more ‘proper’ change, then you need to add an entry to the machine.config file on your server.

<runtime>
	<generatePublisherEvidence enabled="false"/>
</runtime>

See http://msdn.microsoft.com/en-us/library/bb629393.aspx for more info about this setting.

For a more detailed discussion of this issue, see the TechNet wiki entry

As far as I can tell this is a vanilla install of Honeycomb with some pre-installed apps courtesy of Acer – some games, Acer clear.fi client and some other boring things.

The full specs are here, but the gist of my model is: 1Ghz Tegra 2 CPU, 16GB RAM, Wi-Fi, GPS, Full USB, micro USB, 10.1″ screen with 1280 x 800 resolution, HDMI out

Hardware

Hardware positives

• The build quality of the Acer is great, I’m loving the brushed aluminium back.
• The weight is OK, the same as an iPad 1.
• Responsiveness of the screen is excellent
• Battery life is also great – probably a solid 8 hours of heavy use before a recharge is needed. More than the 2 hours my laptop gives me anyway!
• USB host connectivity works for a keyboard and FAT32 flash drives, but not for digital camera plugged directly in.
• The performance is excellent, showcased by some games like Need for Speed and Air Attack. But even for browsing websites it makes my Windows 7 laptop appear sluggish.

Hardware negatives

• The power charger is a proprietary adapter type that occupies 2 outlets because of it’s dumb sideways design. It’s also got a short cable.
• The client USB port is micro USB, not mini USB, so I can’t reuse all the existing cables of that type. One is supplied in the box tho
• Screen shows fingerprints really badly when the screen is off. Not noticeable when the screen is on though.
• HDMI out requires a micro (or is it mini?) cable or adapter, so I haven’t been able to test this out yet on the big screen. Other users have reported that the screen is mirrored to the big screen just like the iPad
• My Wi-Fi connection drops out semi-regularly – this could be software related. It just seems to have a weaker signal than some of my other devices on the same network. I do get dropouts on other devices though, so could be related to interference at my house, a crappy access point or all of the above.
• This is the annoying power charger (photo taken with the 5MP camera on the Iconia):

Honeycomb

Honeycomb positives

• It works out of the box, no need to tether to a PC with iTunes – the true post PC era Steve Jobs!
• Flash – great for websites like iView, but crap for websites heavy with ads like smh.com.au
• The status bar and notifications, including the groovy way to switch between apps.
• Widgets of course
• Text selection for cut, copy and paste works well.
• Standard browser is great. Multi tab browsing works a treat.

Honeycomb negatives

• Force closes – some apps just don’t run others FC in certain functions. For example, Facebook app FCs if I try to search for an event.
• Small number of apps targeted specifically at Honeycomb. Most apps just go full screen and look OK. I’ve on,y seen one stay small and that is Battery Graph, its still functional though.
• Still lots of little quirks and bugs to iron out. I won’t list them all but some odd behaviour with windows in the Market as an example.
• The Market doesn’t really distinguish between apps written for Tablets versus those for phones like the iOS world which is a minor niggle.
• No screenshot apps without root access which I don’t have yet?

Overall

Overall I’m thrilled with it and I have the patience to wait for some of the software things to be fixed. It works great for me as a replacement to a laptop sitting on the lounge browsing the web in front of the TV. Now if only I could get the recorded TV files from my Windows 7 HTPC to natively play on this device…

If you already use a lot of Google apps and services then you should consider a Honeycomb device over an iPad. The lack of tablet specific apps is probably the biggest drawback if you were considering a device like this versus an iPad. Time will correct this of course. And you might need to be a bit tolerant of the occasional app crash though!

Let me know if you’ve got this device and how you got on, or if you’re considering this or something similar and want to ask questions.

I will try and post a review of the tablet with more details later on today hopefully – been brilliant so far, but a few hiccups along the way.

• Have you ever been pulled over for a random yawn test?
• Do police arrest you if you buy too many energy drinks?
• What about the guy driving at 110 km/h with the window down and the music blaring?
• And what do they actually do if they find someone who is a bit tired, as far as I know it’s not against the law?

Once again we’re tackling the big issues here – let me know your experiences below. Have you ever been pulled over and forced to have a powernap before you could continue your journey?

If you want to download it, visit this thread on XDA Developers.

For another guide on getting Froyo running on a HTC Hero with optimal performance, see this thread on XDA developers also. It’s a lot more granular (i.e. lots more control and time consuming).

Personally I just opted for the Floyo ROM because it was quick and simple. I now have about 100MB of RAM free on my phone after re-installing most my apps – important ones like Angry Birds of course!

I’ve found the performance, especially web browsing to be improved over my previos ROM – Sense Hero based on Android 2.1.

The only downside to the Floyo ROM is that it doesn’t have HTC Sense. This is HTC’s custom UI and other bits on it. This means I can’t run the HTC Sync software on my work computer and synchronise my contacts and calendar from there. I still find it pretty amazing that Android doesn’t support this out of the box – that the preferred method is to sync these other data sources to the cloud. Well that’s just not going to work in my situation.

Anyway, if you’re running a HTC Hero and you’re keen to get Android 2.2 on it – check out Floyo 1.1.9 it is tops. As far as I am aware HTC will not be bringing out an official Froyo ROM for this phone so a custom ROM will be the only way to get it. If you’re really hanging out for a ROM with HTC Sense then maybe hang off a bit – I’m sure there will be one ported from a similar phone like the HTC Legend in the fullness of time.

So what’s your experience – anyone else using the Floyo ROM leave your comments below.