Tag Info

Hot answers tagged

16

Here are some of the things you need to understand, there are tons more - but generally these are the ones I'm discussing during my classes: Understand the SharePoint Solution Framework - how WSP's are added to SP and then distributed in the farm SQL Server!!!! Understand how large database affects performance Understand how queries affects performance ...


14

Although there is full support for automatic updates in SharePoint Foundation 2010, we never have automatic updates turned on for production servers. We normally install patches manually, test them then roll them out within two weeks of release. You will still need to schedule a downtime window for updates due to reboots etc. Obviously that requires ...


6

I very recently had the same issue with Nintex Workflow 2010: It wouldn't install properly from my App server because this service was turned off. Other than that, yes it's a good way to turn an application server into a true application server, and prevent it from running and serving the content web applications and IIS sites. Have a look at the ...


6

Focus on packaging up EVERYTHING as Features, Site Definitions, DLLs, etc in a WSP. You will use this portable installer package to deploy to other environments (staging, production, etc). There will most likely be some extra work involved in "solutionising" any custom funcionality that was created with SharePoint Designer. You will need to get familiar ...


6

You don't necessarily require the same number of servers per environment but it does help to validate testing. You could for example have a standalone (single server) machine for dev, 2 UAT server and 2 production servers - this isn't uncommon but you will need to ensure that you have sufficient hardware resources (SP2010 loves RAM and fast disks for ...


5

It is important that you work with your infrastructure team to apply patches on a set, predictable schedule. System updates have caused issues with SharePoint before, and will always have the potential to throw a wrench in the works with any application. This is a production environment and anything that can cause downtime needs to be handled in a ...


5

I would also add basic understanding of DNS and local hosts file SMTP Load Balancing As a developer these are generally out of your remit and in the sysadmins domain - but in reality you need a working knowledge of all of this especially when troubleshooting your apps and persuading the sys admins to 'just take another look at that thing you insist is ...


5

If you are limited by Memory, you can share the same App Pool between several Web applications. This will allow you to save some memory. Sharing AppPools will allow you to pool that "base" memory per app pool. The total amount of requiered RAM cannot be dertermined by just the number of WebApps, but is also affected by many other factors. A valid reason to ...


4

Leave it on, as it provides flexibility. You should already be directing traffic to your WFEs with DNS or a load balancer, so other servers won't have traffic hitting them. But leave it on. Down the road you'll need to perform some sort of troubleshooting, and this will come in handy. "Well, then turn it on when you need to!" I hear you say. The problem ...


4

If the content is in a content database with no other content, you could backup the content db and move it and then attach it in the new farm. If it is not, you can use the command line stsadm.exe -o backup -url [yoursitecollectionurl] -filename [filename] to back the file up on the current server, then copy the file to the new server and use the command ...


3

It's bit hard to answer your question with just a diagram but here we go... If you are looking how to setup all these servers here is an article on SharePoint 2010 ALM with Team Foundation Server, diagram: If you are looking how to configure your environment for something like this check the article I wrote for v2007, the diagram:


3

We are trying to achieve what is explained here. Having a TFS + Build Server + Testing environment all working together is a must, but somehow complex to achieve.


3

You must deploy your WSP solution between each staging. This is done by uploading your WSP file in Central administration of your UAT, Prod environment. So it means you need to install SharePoint 2010 on each environment. I don't know if you understand this detail. Do you need more details ?


3

The thing to remember is that on x64 systems, an app pool can consume a lot of memory, if unchecked, 1GB+. On x86 systems, ASP.NET is capped at 900MB by default. Setting a memory cap is useful, but will cause the app pool to recycle if the app tries to go over it, basically bringing your app down for a few seconds. If bad code (Undisposed SPSite objects ...


3

The following whitepaper details all enterprise-only features and how they are used with SharePoint: http://technet.microsoft.com/en-us/library/cc990273.aspx It also includes a table that relates SQL Server editions with SharePoint functions. The most complete overview of SQL Server features can be found at ...


3

SharePoint 2010 Service Pack 1 does support content databases up to 4TB, if upgrading to Service Pack 1 is one of the things available in your SharePoint road map, let's do it to utilize maximum its capability. I've found a white paper "Managing Multi-Terabyte Content Databases with SharePoint 2010" written by MCM, Sr. Technical Product Manager Bill Baer. ...


2

This isn't one of the operating systems supported for SharePoint 2007 installation by Microsoft (at time of writing). It would be an interesting experiment but if your installation would be unsupported I'm not sure it's a good idea. Update: Tried this out but found running setup.exe did absolutely nothing, presumably because of missing components of the ...


2

The Service Applications are redundant if you make sure that you install and start the services on multiple load balanced application servers. SharePoint 2010 handles the load balancing for you. Regarding the fail-over strategies you have to look into the different services and if they use a database for storage or not. The databases that the SA uses must ...


2

The environment I work on is a bit simpler...somewhat. Every machine is a VM Development works locally on single machines, all on VM's. This allows coding and installing updated WSP's though Visual Studio Integration compiles the builds and deploys to the general Development environment, still single machine Test is a multi-machine environment with a CA, ...


2

Do you have your Alternate Access Mappings setup correctly to be able to use the NLB address? Also, I'm pretty sure you need to have another nic for server to server communication. If you are using Microsoft Windows NLB, typically you won't be able to communicate with the other servers in the NLB through the interface that is load balanced, thus you need ...


2

Without knowing what you need, probably 1 limited deployment for your DEV And 2x small farms for your stage and prod (remembering you want your stage and prod to be the same) http://technet.microsoft.com/en-us/library/cc263044.aspx


2

As a sysadmin who works with Sharepoint Developers, I'd suggest basic understanding of the things you want to interface with, or rely on is what is important; I'm not suggesting you should be an expert on AD (for example) you should know enough about it to understand what it does for SP and how it is used so you can get the best out of it instead of viewing ...


2

The answer to your question is Yes! you would configure you Web Application to use claims based authentication and then enable Anonymous access on the root site collection and then turn it off for the subsite. See the documentation here http://technet.microsoft.com/en-us/library/ee806886.aspx


2

Be careful here - Chris is correct in that 200MB is an absolute minimum amount of RAM that an app pool might use. On the flip side, if you had say 5 very busy app pools you could quite conceivably hit 24GB + RAM usage. I wrote a blog on app pools on this a while back which goes into a little more detail (which includes some input from Steve Smith, MVP). ...


2

You could use WFE's to cache, but ensure you have enough RAM assigned in each of the servers to cater for the needs of App Fabric and SharePoint. Performance will drop if the servers run out of RAM and start paging. The main motivator to run on the same servers should be cost driven. Some possible reasons to run App Fabric caching on seperate servers are in ...


2

Sites Size The biggest factor in choosing to go with sites is size. A single site collection cannot span multiple content databases. Microsoft recommends that a single content database should not grow beyond 200 Gigabytes. This isn’t a hard limit, and with expensive hardware (Disk sub-system performance of 0.25 IOPs per GB. 2 IIOPs per GB) you ...


1

SharePoint Server is built on top of WSS. AS long as your are doing a x86 to x86 upgrade you should be ok following the Technet article. However, if you move to R2 that is x64 only and it requires a clean install. Clean installs are preferred in almost every instance, as they usually will give you less headaches.


1

Yes - Microsoft SharePoint Foundation Web Application is only useful if your server is a FrontEnd - I tend to turn it off on App Server Roles Machine when I can - This way you can ensure that no frontend activity occurs on the app server. The only (potential) issue is with deployment of custom .wsp that will assume that they are deployed on a frontend - you ...


1

It may be worth investing in some outside advice. As Jaap correctly states, for the packaging of the features which will be used to build the site structure you should use WSPs. Also investigate Feature Stapling to correctly attach the features. However, in terms of ensuring you end up with a SharePoint deployment that not only scales well, but that is used ...


1

For development it's a good idea both from a cost and ease-of-use stand-point to use 64-bit Windows 7 machines. You probably already use a 64-Bit W7 laptop for everyday use, just use that. For UAT I recommend using single server Hyper-V instances so you can allocate resources additional resources during UAT. I always have a support instance a production ...



Only top voted, non community-wiki answers of a minimum length are eligible