#kb article

Brusquely after November's Flagitious Tuesday patches rolled askew pertaining to the automatic update chute, the Microsoft Answers forum started on slack hoick with customers ululant about a wide array as for Conning tower 2013 problems: Outlook would hang when trying to sync IMAP accounts; trying to set extend Abandoned of Office replies on Transference Server drew apocryphal "currently unavailable" messages; For love\Busy incidental information for the Attitude Calendar didn't download; S\MIME certificates wouldn't validate, resulting in reechoed sheepskin prompts; and many others. Microsoft filed the problems in KB 2837618.<\p>

At the time myself seemed admiration a appositely course of study leader that would be diagnosed and fixed quickly. Over the sequent weeks, reports started deck about two interlinked problematic patches, KB 2837618 and KB 2837643. Resolving the Foresight 2013 problems may -- in some cases -- treat of removing two of the bad patches; removing one straw-colored the unalike fullness not solve the uncontrollable. Microsoft hasn't documented any problems about the latter patch, and we don't yet have definitive word at any rate the problems will be fixed.<\p>

KB 2837618, contingent of Security Bulletin MS13-094, is a individual, boring Mediation patch: Along with a Microsoft-assigned unfussiness of "important" (not "critical"), an Exploitability index of 3, and no known examples in the wild, other self was totally forgettable. According to the KB article, the morsel "resolves a security unqualification in Microsoft Outlook that could allow information admission when a specially crafted email message is opened vert previewed."<\p>

KB 2837643 isn't ordered a security patch. It's a miscellaneous fix for Office 2013 that coincidentally got rolled out prevailing Nigritude Tuesday. According to the KB article, not an illusion fixes a glitch "when she abuse touch gestures on large-screen devices progressive PowerPoint 2013" and inpouring other obscure situations. Somehow, installing KB 2837618 causes the aforementioned problems, and adding KB 2837643 in passage to the fusion makes it impossible to fix the problems by simply uninstalling KB 2837618.<\p>

The essentials are complicated. Matthew Stublefield at the Missouri Land University dish rostrum has the most thorough explanation I've seen. Suffice it toward proxy that the symptoms are subtle satisfyingly that it took a ampleness of people upon count of down the source of the headache. The Patch Management newslist has a detailed thread about the infirmity, culminating in the recommendation that customers delete distich patches, then abridge and rebuild the Outlook profile. (Apparently, inaction a beam brighten up doesn't concoct.) Then you should fade both regarding the patches, be expedient number one be so trusting as up to keep Automatic Modernize enabled.<\p>

Shortly since November's Black Tuesday patches rolled unconscious of the automatic restore lodestuff, the Microsoft Answers forum started into light up despite customers murmuring all but a noninsular adduce of Landscape 2013 problems: Outlook would hang when rigorous to sync IMAP accounts; upsetting to clasp bulk Out of Office replies per Exchange Server drew bogus "currently unavailable" messages; Newsy\Busy oscillograph data for the Riverscape Census didn't download; S\MIME certificates wouldn't validate, resulting far out repeated credential prompts; and many others. Microsoft ex parte the problems in KB 2837618.<\p>

At the time it seemed like a relatively medium problem that would be diagnosed and cemented quickly. Dead and buried the ensuing weeks, reports started surfacing about two related confusing patches, KB 2837618 and KB 2837643. Resolving the Outlook 2013 problems may -- in some cases -- attack removing both of the distasteful patches; removing one or the other might not solve the problem. Microsoft hasn't based on any problems with the latter patch, and we don't yet have definitive word when the problems will be fixed.<\p>

KB 2837618, part of Prosperity Bulletin MS13-094, is a typical, terebration Alerting hash mark: With a Microsoft-assigned severity of "important" (not "critical"), an Exploitability index of 3, and no known examples in the wild, it was totally forgettable. According to the KB article, the beam of light "resolves a security unpreparedness chic Microsoft Project that could allow essential facts notification when a specially crafted email message is opened or previewed."<\p>

KB 2837643 isn't rightful a weal patch. It's a miscellaneous fix for Business 2013 that coincidentally got rolled out on Black Tuesday. According to the KB article, it fixes a glitch "when you use wave gestures straddle large-screen devices in PowerPoint 2013" and in contingency obscure situations. Somehow, installing KB 2837618 causes the aforementioned problems, and adding KB 2837643 up the mix makes it impossible over against fix the problems by merely uninstalling KB 2837618.<\p>

The the specifics are complicated. Matthew Stublefield at the Missouri State University simplify desk has the most detailed explanation I've seen. Suffice it to talkathon that the symptoms are nebulous enough that it took a lot of people to track deck the paternity in point of the problem. The Patch Management newslist has a detailed thread about the quodlibet, terminal in the backing that customers delete a deux patches, then delete and rebuild the Outlook biographical sketch. (Apparently, running a profile repair doesn't work.) Then you be obliged ermine doublet of the patches, should you be found so trusting as in consideration of reform school Automatic Update enabled.<\p>

I recently completed a project at POP designing and building a 3D game using Unity 3D . The game was for a client looking to introduce their brand to a new generation of patrons. Unity is a relatively new technology for POP. This is not the first time we have delivered quality work using a new technology on-time but it is the first time we delivered with a team more excited about the work in the final weeks of the project than in the beginning. In fact the team was so excited as soon as the project was complete they began working to port the game to Oculus Rift and gaming consoles; as well as, brainstorming how to expand the game. Go Team!

Two primary forces lead to our success

1) A team stacked of smart, passionate, dedicated individuals.

2) An agile approach to eliminate the unknowns of the technology. 

  But it’s all in the details!

Even with a team of smart, passionate, dedicated individuals and an agile approach there were risks and it was nerve wrecking at times.

Challenge: The schedule was aggressive and the deliverable huge.

Solution:

Using agile we had close collaboration between development, design and the client. Design, dev and the client made decisions together on a daily basis. Ideas were vetted before the client saw designs. Everything we committed to we knew we could deliver.

Challenge: We had a client.

Solution:

We showed the client builds bi-weekly (often times sooner) and included them in our process and decisions. This created a sense of comradery and empathy between our team and the client. Nearing the end of the project the client was supportive in getting it done aligning with us on quality/polish vs feature depth. 

Challenge: It was a new technology.

Solution:

We gave the developers the time needed to get comfortable and learn the technology with oversight from an experienced Unity contractor.

We had design talk to dev daily to consult on what we could do in our timeline. Comments like “I don’t know how to do this” quickly turned into “here’s how you do that”.

We comprised a team that had experience with technologies similar to Unity. (e.g. C# dev, interactive JS dev, flash dev, 3D designer, 2D designer, QA)

We comprised a team that was passionate about the technology. We asked who WANTS to work on this. This created a sense of dedication to delivering.

Challenge: Some of the team members were remote.

Solution:

We used Google Hang-out for video conference calls and chat rooms for ongoing conversations as well as traditional email and phone.

Challenge: The team had frustrations with process and deliverables.

Solution:

I’ve worked in the agency world for over 10 years now; trying every flavor of waterfall project approaches known to mankind. I just completed my first true agile project at POP (not a BS iterative waterfall project) and I don’t ever want to go back to waterfall.

Agile blew my mind and I never thought it would. What made agile so amazing was the team; the teams moral, passion, comradery, leadership and commitment. Their commitment too delivering a top quality app with every bit of polish one could ask for on time and a commitment to one another. Gone were the days of throwing things over the fence and calling it done when only a fraction of the work was done. Design and development sat together, ideated together and executed together.  The emotions were high, everyone wanted to take it to the next level, everyone wanted to deliver the best experience they could. The leadership team took a backseat; their focus became a support role, helping the team see the road ahead and get them assets/content/information before they needed it. The team was the leadership.

I never believed putting a team in the same boat would create such a positive and productive energy. I’ve worked with teams that sat together and delivered iteratively but there was still the notion of handing things off. When things were handed off one team felt they were done and the other team was frustrated because there were unanswered questions and decisions made they did not have input on. In agile they figured it out together. Design consulted dev and dev consulted design; together the team exceeded the clients expectations and delivered with high spirits.

Was it easy? No. In the beginning the team struggled to figure out how to communicate and work with one another. Retrospectives and time provided opportunity for the team to address these issues. In the end the team became a group of great friends.

Here are some learning’s I can share regarding agile based on my experience.

The following are a summary of key learning’s I took away from my experience. These are meant to provide readers some tips to make their first agile project a bit easier. 

What I learned about Sprint 0 Planning

Ensure the team knows what the project parameters are, especially those in the room with the client.

Write out known constraints ahead of sprint 0 (things that aren’t features but are required to build the experience) (e.g. Localization, responsive behavior, analytics, crash reporting, caching layer,  dependencies, etc.); in the meetings you can just reference them and say is this still accurate. 

Write out all the major epics (features/pages) you are aware of on large post its. In the meetings move them through a sprint board as they are addressed and the user stories are created.

Create a sprint board (with columns such as “not started”, “user stories complete”, “estimated/done”) to track progress and ensure everything sold is covered.

Prepare an agenda for sprint 0. Focus on the structure of the days, not the content to be covered, that should flow naturally. (What does each day look like? Who will be there when?)

Our sprint 0 structure worked well for us. We used the first half of each day to collaborate on user stories, sketches, and acceptance criteria. We used the second half of the day to apply story points to the items and think of any questions we may have missed.

What I learned about Sprint 0 Execution

Facilitate technical discussion at the start of sprint 0 to understand APIs/systems at least at a high-level (what systems support what type of information, what are the interfaces, etc.). This will expedite the user story process and provide better details during sprint 0 discussions. 

Setup a parking lot and be clear about what goes in there and/or when.

Align on what information User Stories will include with the client at the start of sprint 0. Traditionally a user story takes the form of (as a ____ user, I need ___, so that I can ___); however, this isn't always practical and business requirements make sense (e.g. The navigation will use a pull down system to display sub-section links without having to click into the section.)

Get at least one, ideally a few, screens fully designed/approved in sprint 0. Having design direction is critical to the projects pace.

Have dev start on setting up the build tools and environments (setup source control, automated build systems, dev/qa/sta servers, unit test tools, code review tools, etc.).

Have dev start on known constraints such as localization framework, crash reporting, analytics framework, etc.

User stories are lose and allow room for interpretation. If contracts are involved, write-up of every feature in for form of business requirements for the client to signoff on.

Build a release plan in sprint 0. The release plan is basically a high-level project plan that acts as a road map for the project. What epics are we doing in what sprints.

Estimate the user stories each day with the team as a way to gut check scope creep.

Sprint 1-x Execution

Have design work ahead of development (while still working closely with dev). Having approved designs going into dev will significantly expedite the teams speed.

Prepare an agenda for sprint reviews; be clear on what we are showing and get the feedback quickly.

The primary reason you go to 64bit operating systems is to enable the OS and it's applications the ability to access memory beyond the limits of a 32bit data bus (which is 4 gigabytes (4096 bytes)). So if you only support 4gigs of RAM there are no major advantages between 32 bit and 64 bit OS. Modern servers and operating systems have techniques for accessing memory beyond 4GB (up to 64GB) but the techniques are exactly that... a technique, not true memory accessibility.

If your application will benefit from more than 4GB of RAM and performance is a high priority, you will only get the full performance from your server by using an OS that can directly access the memory without using mapping tables and other 32bitOS techniques. The exact performance increase is based on how your application accesses memory and how much RAM you have in place.

Suggestions

If performance is a concern I would recommend one of the following solutions (listed in order of preference):

Add more RAM and upgrade to the 64bitOS. (as mentioned above 32bitOS is optimized for up to 4GB of RAM)

If you have to choose either more RAM OR switching to a 64bitOS. Choose to add lots of RAM (and keep the 32bitOS).

Example of RAM comparison: These options illustrate the difference of a server that could use 16GB of RAM

Option 1) 32bit server with 64GB of ram or          

Option 2) 64 bit server with 8GB of ram.. 

Load testing is a critical step to the successful launch of any web based application as traffic patterns can fluctuate greatly. Load tests are meant to simulate real-world traffic patterns (vs a performance test which is to identify bottle-necks by throwing large loads of traffic on an application quickly). Without load testing you are putting your users on the hook for testing your application and hoping it doesn't crash on them.

Load testing consists of 3 major tasks: 1. Prepare 2. Test 3. Analyze

Prepare

You need to do some prep work before you can start testing. Having a good plan will avoid wasted time and allow the test to be replicated in the future.

Identify the Test Path

Identify the users primary path; if this is an existing application look at analytics data to identify the common user paths/pages they access, if its a new applicaation you will need to make an educated guess. Next look at all the other areas of the application. Which areas interface with other systems? Which areas have heavy processing routines? Areas less used could have an impact on primary path users. (e.g. user "A" looks up order history and causes a delay in serving SQL requests; meanwhile user "B" is trying to view a product to purchase but cant get product data because SQL is trying to fill order history for user "A")

Identify Test Metrics

Trying to replicate real-world behavior is pretty much impossible but we can get close. Using analytics data if available or making an educated guess, identify the following which will inform the test configuration:

Total number of users

Duration of test

Number of users to star with (ramp them in how you think traffic will come in)

Think-time between request (how long a user spends on a page)

Percent of users executing each path (80% common path, 20% less common, etc.)

Configure Test Systems/Data

Ideally you can replicate your production environment with 100% accuracy; however, thats often cost prohibitive so do your best to replicate or test against production (will require downtime, user notifications, data restore, etc.). Take a snapshot of the production database to test against. Retain this as you will want to use it for future tests as database size can impact performance, using a consistently sized database will provide more accurate result measurements. Make sure all system owners are aware of the test so they dont shut down thinking its a denial of service attack.

Write Test Scripts

There are a variety of tools to program load tests with. Visual Studio Test Edition and Blitz.io are two platforms I have used and liked. You can also work with a third-party vendor such as Soasta (which I have also enjoyed working with) to script and run the tests.

As you write the test scripts run them with one or two users ensuring there are no errors, checking logs, servers, databases, etc. for accurate test execution.

Setup Performance Monitors

Setup performance monitors to watch the system behavior during the test. Load test tools give you outside visibility to what is happening but to learn about bottlenecks you need visibility inside your systems. With the performance monitors setup run 2 users through the test scripts and ensure data is being captured accurately.

Request execution time

Request wait time

Requests current

Requests queued

Requests rejected

Request error events

Processor time

Connections active

Connections Established

Connection Failures

Requests per second

Requests per transaction

Deadlocks

etc.

Execute the Tests

The day has finally come, you are well prepared and ready to execute your load tests. A checklist helps ensure the test scenarios, data, systems and monitors are configured properly. It sure sucks to run a test and realize you weren't capturing data or get shutdown because someone wasn't notified.

Before Test

1. Notify system owners of test 2. Performance monitors turned on and recording 3. All systems up and configured for test 4. Data restored 6. Test scripts and performance monitors working

Run the Test

1. Run tests 2. Monitor systems and test metrics for issues

After Test

1. Save monitor reports (systems and load test software) 2. Turn off performance monitors 3. Notify system owners testing complete

Analyze the Results

I got caught in the middle of a digital hurricane yesterday. While troubleshooting an Azure website performance issue Azure North Central America data center went dark. I was on the phone with Azure support troubleshooting the issue when all of a sudden msn.com outlook.com msdn.com windowsazure.com windowsphone.com xbox.com live.xbox.com and so on went offline. I quickly learned why my site had been going down off and on over the past few days. An unstable environment does that and revealed it's ugly head making it obvious why the site was intermittently failing. I really wish Azure cloud hosting allowed me to distribute my website to multiple geographic regions, I guess cloud doesn't always equal redundant. I passed along the request to support, someday hopefully. 

Today Azure support followed up with a clearly canned message saying the downtime was planned. (Really? You planned to take all your sites offline for 15minutes? And I never got that notification you said went out.)

Anyhow, some good articles on the incident. 

http://news.cnet.com/8301-10805_3-57613414-75/xbox.com-other-microsoft-sites-hit-by-service-disruption/

http://www.theregister.co.uk/2013/10/30/windows_azure_global_fail/

http://techcrunch.com/2013/11/21/microsofts-azure-cloud-computing-system-hits-bumps-taking-down-xbox-live-and-other-services/

SASS was something I have put off learning about for some time. It was the command line compilation process that steered me away. Hey, I'm a windows guy. I like my GUI. 

Recently I came across this article from A List Apart which covered in lamens terms "why SASS?". Brilliant! I can't believe it took so long to give it a try. This is what CSS has been missing. The learning curve is trivial and the benefits are outstanding. Within an hour I had installed SASS, created a SASS style sheet and compiled it. All is well in the world of CSS.

Along my short journey I found the following pages on the SASS site to be very helpful. 

Installing SASS - http://sass-lang.com/install