Search Results

Search found 58774 results on 2351 pages for 'ssis data tranformations'.

Page 26/2351 | < Previous Page | 22 23 24 25 26 27 28 29 30 31 32 33  | Next Page >

  • Truly understand the threshold for document set in document library in SharePoint

    - by ybbest
    Recently, I am working on an issue with threshold. The problem is that when the user navigates to a view of the document library, it displays the error message “list view threshold is exceeded”. However, in the view, it has no data. The list view threshold limit is 5000 by default for the non-admin user. This limit is not the number of items returned by your query; it is the total number of items the database needs to read to calculate the returned result set. So although the view does not return any result but to calculate the result (no data to show), it needs to access more than 5000 items in the database. To fix the issue, you need to create an index for the column that you use in the filter for the view. Let’s look at the problem in details. You can download a solution to replicate this issue here. 1. Go to Central Admin ==> Web Application Management ==>General Settings==> Click on Resource Throttling 2. Change the list view threshold in web application from 5000 to 2000 so that I can show the problem without loading more than 5000 items into the list. FROM TO 3. Go to the page that displays the approved view of the Loan application document set. It displays the message as shown below although I do not have any data returned for this view. 4. To get around this, you need to create an index column. Go to list settings and click on the Index columns. 5. Click on the “Create a new index” link. 6. Select the LoanStatus field as I use this filed as the filter to create the view. 7. After the index is created now I can access the approved view, as you can see it does not return any data. Notes: List View Threshold: Specify the maximum number of items that a database operation can involve at one time. Operations that exceed this limit are prohibited. References: SharePoint lists V: Techniques for managing large lists Manage large SharePoint lists for better performance http://blogs.technet.com/b/speschka/archive/2009/10/27/working-with-large-lists-in-sharepoint-2010-list-throttling.aspx

    Read the article

  • Big DataData Mining with Hive – What is Hive? – What is HiveQL (HQL)? – Day 15 of 21

    - by Pinal Dave
    In yesterday’s blog post we learned the importance of the operational database in Big Data Story. In this article we will understand what is Hive and HQL in Big Data Story. Yahoo started working on PIG (we will understand that in the next blog post) for their application deployment on Hadoop. The goal of Yahoo to manage their unstructured data. Similarly Facebook started deploying their warehouse solutions on Hadoop which has resulted in HIVE. The reason for going with HIVE is because the traditional warehousing solutions are getting very expensive. What is HIVE? Hive is a datawarehouseing infrastructure for Hadoop. The primary responsibility is to provide data summarization, query and analysis. It  supports analysis of large datasets stored in Hadoop’s HDFS as well as on the Amazon S3 filesystem. The best part of HIVE is that it supports SQL-Like access to structured data which is known as HiveQL (or HQL) as well as big data analysis with the help of MapReduce. Hive is not built to get a quick response to queries but it it is built for data mining applications. Data mining applications can take from several minutes to several hours to analysis the data and HIVE is primarily used there. HIVE Organization The data are organized in three different formats in HIVE. Tables: They are very similar to RDBMS tables and contains rows and tables. Hive is just layered over the Hadoop File System (HDFS), hence tables are directly mapped to directories of the filesystems. It also supports tables stored in other native file systems. Partitions: Hive tables can have more than one partition. They are mapped to subdirectories and file systems as well. Buckets: In Hive data may be divided into buckets. Buckets are stored as files in partition in the underlying file system. Hive also has metastore which stores all the metadata. It is a relational database containing various information related to Hive Schema (column types, owners, key-value data, statistics etc.). We can use MySQL database over here. What is HiveSQL (HQL)? Hive query language provides the basic SQL like operations. Here are few of the tasks which HQL can do easily. Create and manage tables and partitions Support various Relational, Arithmetic and Logical Operators Evaluate functions Download the contents of a table to a local directory or result of queries to HDFS directory Here is the example of the HQL Query: SELECT upper(name), salesprice FROM sales; SELECT category, count(1) FROM products GROUP BY category; When you look at the above query, you can see they are very similar to SQL like queries. Tomorrow In tomorrow’s blog post we will discuss about very important components of the Big Data Ecosystem – Pig. Reference: Pinal Dave (http://blog.sqlauthority.com) Filed under: Big Data, PostADay, SQL, SQL Authority, SQL Query, SQL Server, SQL Tips and Tricks, T SQL

    Read the article

  • SSIS ForEachLoop Container

    - by Leonard Mwangi
    I recently had a client request to create an SSIS package that would loop through a set of data in SQL tables to allow them to complete their data transformation processes. Knowing that Integration Services does have ForEachLoop Container, I knew the task would be easy but the moment I jumped into it I figured there was no straight forward way to accomplish the task since for each didn’t really have a loop through the table enumerator. With the capabilities of integration Services, I was still confident that it was possible it was just a matter of creativity to get it done. I set out to discover what different ForEach Loop Editor Enumerators did and settled with Variable Enumerator.  Here is how I accomplished the task. 1.       Drop your ForEach Loop Container in your WorkArea. 2.       Create a few SSIS Variable that will contain the data. Notice I have assigned MyID_ID variable a value of “TEST’ which is not evaluated either. This variable will be assigned data from the database hence allowing us to loop. 3.       In the ForEach Loop Editor’s Collection select Variable Enumerator 4.       Once this is all set, we need a mechanism to grab the data from the SQL Table and assigning it to the variable. Fig: Select Top 1 record Fig: Assign Top 1 record to the variable 5.       Now all that’s required is a house cleaning process that will update the table that you are looping so that you can be able to grab the next record   A look of the complete package

    Read the article

  • Data Masking for Oracle E-Business Suite

    - by Troy Kitch
    E-Business Suite customers can now use Oracle Data Masking to obscure sensitive information in non-production environments. Many organizations are inadvertently exposed when copying sensitive or regulated production data into non-production database environments for development, quality assurance or outsourcing purposes. Due to weak security controls and unmonitored access, these non-production environments have increasingly become the target of cyber criminals. Learn more about the announcement here.

    Read the article

  • Extracting Data from a Source System to History Tables

    - by Derek D.
    This is a topic I find very little information written about, however it is very important that the method for extracting data be done in a way that does not hinder performance of the source system.  In this example, the goal is to extract data from a source system, into another database (or server) all [...]

    Read the article

  • Importing Excel data into SSIS 2008 using Data Conversion Transformation

    Despite its benefits, SQL Server Integration Services Import Export Wizard has a number of limitations, resulting in part from a new set of rules that eliminate implicit data type conversion mechanisms present in Data Transformation Services. This article discusses a method that addresses such limitations, focusing in particular on importing the content of Excel spreadsheets into SQL Server.

    Read the article

  • Design patterns to avoiding breaking the SRP while performing heavy data logging

    - by Kazark
    A class that performs both computations and data logging seems to have at least two responsibilities. Given a system for which the specifications require heavy data logging, what kind of design patterns or architectural patterns can be used to avoid bloating all the classes with logging calls every time they compute something? The decorator pattern be used (e.g. Interpolator decorated to LoggingInterpolator), but it seems that would result in a situation hardly more desirable in which almost every major class would need to be decorated with logging.

    Read the article

  • Ma este Oracle Data Mining újdonságok webcast!

    - by Fekete Zoltán
    2010. május 12-én szerdán 18 órakor a böngészonkkel kapcsolódva a következo roppant érdekes eloadást hallgathatjuk meg az Oracle BIWA keretében: BIWA SIG TechCast Series - May 12 - Data Mining Made Easy, az eloadó Charlie Berger, az Oracle adatbányászati vezetoje. Könnyen elvégezheto adatbányászat! Az Oracle Data Miner 11g Release 2 új "Work flow" grafikus felületének bevezetése. Csatlakozni az Oracle BIWA-hoz a ezen a linken ingyenesen lehet. Itt találhatjuk meg, hogyan lehet meghallgatni ezt a konferenciát: www.oraclebiwa.org

    Read the article

  • Google doesn't always show rich snippets when the site uses structured data [duplicate]

    - by Sam Se
    This question is an exact duplicate of: Google Structured Data [on hold] 1 answer I'm so tired of the Google structured data recipe. After some days, it loses the image and the extra information. Then I test it again, and it shows again. Some other days in the future it might go away even if it is still showing in test tool. What i can do? I tried with RDFa and schema.org microdata.

    Read the article

  • Import SSIS Project in Denali CTP1

    For years Analysis Services has had the ability to take an existing database from a server and reverse engineer it into a BIDS project.  This is extremely useful when all you have is the running instance of the database and the project that created it has long since disappeared.  Reverse engineering has never been a feature of SSIS until now. Let me walk you through the simple steps. The first step is that you obviously have to have a project deployed to an SSIS Catalog.  I will do a video on this soon but in case you can’t wait then my good buddy Jamie Thomson has written it up here As you can see I have a project called imaginatively “Denali1” with one package “Package.dtsx” The next thing we need to do is fire up BIDS and choose the right project type (Integration Services Import Project) Now we just follow the wizard.  We make sure we specify on which server to find the Catalog and in which folder to look for the project. Next the setting are validated and we are greeted with the familiar review screen before the creation of our new project from the deployed project happens Hit Import and away we go The result is just what we wanted.

    Read the article

  • SSIS Send Mail Task and ForceExecutionValue Error

    - by Kevin Shyr
    I tried to use the "ForcedExecutionValue" on several Send Mail Tasks and log the execution into a ExecValueVariable so that at the end of the package I can log into a table to say whether the data check is successful or not (by determine whether an email was sent out) I set up a Boolean variable that is accessible at the package level, then set up my Send Mail Task as the screenshot below with Boolean as my ForcedExecutionValueType.  When I run the package, I got the error described below. Just to make sure this is not another issue of SSIS having with Boolean type ( you also can't set variable value from xp_cmdshell of type Boolean), I used variables of types String, Int32, DateTime with the corresponding ForcedExecutionValueType.  The only way to get around this error, was to set my variable to type Object, but then when you try to get the value out later, the Object is null. I didn't spend enough time on this to see whether it's really a bug in SSIS or not, or is this just how Send Mail Task works.  Just want to log the error and will circle back on this later to narrow down the issue some more.  In the meantime, please share if you have run into the same problem.  The current workaround is to attach a script task at the end. Also, need to note 2 existing limitation: Data check needs to be done serially because every check needs to be inner join to a master table.  The master table has all the data in a single XML column and hence need to be retrieved with XQuery (a fundamental design flaw that needs to be changed) The next iteration will be to change this design into a FOR loop and pull out the checking query from a table somewhere with all the info needed for email task, but is being put to the back of the priority. Error Message: Error: 0xC001F009 at CountCheckBetweenODSAndCleanSchema: The type of the value being assigned to variable "User::WasErrorEmailEverSent" differs from the current variable type. Variables may not change type during execution. Variable types are strict, except for variables of type Object. Error: 0xC0019001 at Send Mail Task on count mismatch: The wrapper was unable to set the value of the variable specified in the ExecutionValueVariable property.   Screenshot of my Send Mail Task setup:

    Read the article

  • Messages do not always appear in [catalog].[event_messages] in the order that they occur [SSIS]

    - by jamiet
    This is a simple heads up for anyone doing SQL Server Integration Services (SSIS) development using SSIS 2012. Be aware that messages do not always appear in [catalog].[event_messages] in the order that they occur, observe… In the following query I am looking at a subset of messages in [catalog].[event_messages] and ordering them by [event_message_id]: SELECT [event_message_id],[event_name],[message_time],[message_source_name]FROM   [catalog].[event_messages] emWHERE  [event_message_id] BETWEEN 290972 AND 290982ORDER  BY [event_message_id] ASC--ORDER BY [message_time] ASC Take a look at the two rows that I have highlighted, note how the OnPostExecute event for “Utility GetTargetLoadDatesPerETLIfcName” appears after the OnPreExecute event for “FELC Loop over TargetLoadDates”, I happen to know that this is incorrect because “Utility GetTargetLoadDatesPerETLIfcName” is a package that gets executed by an Execute Package Task prior to the For Each Loop “FELC Loop over TargetLoadDates”: If we order instead by [message_time] then we see something that makes more sense: SELECT [event_message_id],[event_name],[message_time],[message_source_name]FROM   [catalog].[event_messages] emWHERE  [event_message_id] BETWEEN 290972 AND 290982--ORDER BY [event_message_id] ASCORDER  BY [message_time] ASC We can see that the OnPostExecute for “Utility GetTargetLoadDatesPerETLIfcName” did indeed occur before the OnPreExecute event for “FELC Loop over TargetLoadDates”, they just did not get assigned an [event_message_id] in chronological order. We can speculate as to why that might be (I suspect the explanation is something to do with the two executables appearing in different packages) but the reason is not the important thing here, just be aware that you should be ordering by [message_time] rather than [event_message_id] if you want to get 100% accurate insights into your executions. @Jamiet

    Read the article

  • Big data: An evening in the life of an actual buyer

    - by Jean-Pierre Dijcks
    Here I am, and this is an actual story of one of my evenings, trying to spend money with a company and ultimately failing. I just gave up and bought a service from another vendor, not the incumbent. Here is that story and how I think big data could actually fix this (and potentially prevent some of this from happening). In the end this story should illustrate how big data can benefit me (get me what I want without causing grief) and the company I am trying to buy something from. Note: Lots of details left out, I have no intention of being the annoyed blogger moaning about a specific company. What did I want to get? We watch TV, we have internet and we do have a land line. The land line is from a different vendor then the TV and the internet. I have decided that this makes no sense and I was going to get a bundle (no need to infer who this is, I just picked the generic bundle word as this is what I want to get) of all three services as this seems to save me money. I also want to not talk to people, I just want to click on a website when I feel like it and get it all sorted. I do think that is reality. I want to just do my shopping at 9.30pm while watching silly reruns on TV. Problem 1 - Bad links So, I'm an existing customer of the company I want to buy my bundle from. I go to the website, I click on offers. Turns out they are offers for new customers. After grumbling about how good they are, I click on offers for existing customers. Bummer, it goes to offers for new customers, so I click again on the link for offers for existing customers. No cigar... it just does not work. Big data solutions: 1) Do not show an existing customer the offers for new customers unless they are the same => This is only partially doable without login, but if a customer logs in the application should always know that this is an existing customer. But in general, imagine I do this from my home going through the internet service of this vendor to their domain... an instant filter should move me into the "existing customer route". 2) Flag dead or incorrect links => I've clicked the link for "existing customer offers" at least 3 times in under 5 seconds... Identifying patterns like this is easy in Hadoop and can very quickly make a list of potentially incorrect links. No need for realtime fixing, just the fact that this link can be pro-actively fixed across my entire web domain is a good thing. Preventative maintenance! Problem 2 - Purchase cannot be completed Apart from the fact that the browsing pattern to actually get to what I want is poorly designed, my purchase never gets past a specific point. In other words, I put something into my shopping cart and when I want to move on the application either crashes (with me going to an error page) or hangs or goes into something like chat. So I try again, and again and again. I think I tried this entire path (while being logged in!!) at least 10 times over the course of 20 minutes. I also clicked on the feedback button and, frustrated as I was, tried to explain this did not work... Big Data Solutions: 1) This web site does shopping cart analysis. I got an email next day stating I have things in my shopping cart, just click here to complete my purchase. After the above experience, this just added insult to my pain... 2) What should have happened, is a Hadoop job going over all logged in customers that are on the buy flow. It should flag anyone who is trying (multiple attempts from the same user to do the same thing), analyze the shopping card, the clicks to identify what the customers wants, his feedback provided (note: always own your own website feedback, never just farm this out!!) and in a short turn around time (30 minutes to 2 hours or so) email me with a link to complete my purchase. Not with a link to my shopping cart 12 hours later, but a link to actually achieve what I wanted... Why should this company go through the big data effort? I do believe this is relatively easy to do using our Oracle Event Processing and Big Data Appliance solutions combined. It is almost so simple (to my mind) that it makes no sense that this is not in place? But, now I am ranting... Why is this interesting? It is because of $$$$. After trying really hard, I mean I did this all in the evening, and again in the morning before going to work. I kept on failing, But I really wanted this to work... so an email that said, sorry, we noticed you tried to get a bundle (the log knows what I wanted, where I failed, so easy to generate), here is the link to click and complete your purchase. And here is 2 movies on us as an apology would have kept me as a customer, and got the additional $$$$ per month for the next couple of years. It would also lead to upsell on my phone package etc. Instead, I went to a completely different company, bought service from them. Lost money for company A, negative sentiment for company A and me telling this story at the water cooler so I'm influencing more people to think negatively about company A. All in all, a loss of easy money, a ding in sentiment and image where a relatively simple solution exists and can be in place on the software I describe routinely in this blog... For those who are coming to Openworld and maybe see value in solving the above, or are thinking of how to solve this, come visit us in Moscone North - Oracle Red Lounge or in the Engineered Systems Showcase.

    Read the article

  • Classic vs universal Google analytics and loss of historical data

    - by iss42
    I'm keen to use some of the new features in Google Universal Analytics. I have an old site though that I don't want to lose the historical data for. The comparisons with historical data are interesting for example. However Google doesn't appear to allow you to change a property from the classic code to the new code. Am I missing something? I'm surprised this isn't a bigger issue for many other users.

    Read the article

  • Classic vs universal and loss of historical data

    - by iss42
    I'm keen to use some of the new features in Google Universal Analytics. I have an old site though that I don't want to lose the historical data for. The comparisons with historical data are interesting for example. However Google doesn't appear to allow you to change a property from the classic code to the new code. Am I missing something? I'm surprised this isn't a bigger issue for many other users.

    Read the article

  • Referencing Entity from external data model - Core Data

    - by Ben Reeves
    I have a external library which includes a core data model, I would like to add a new entity to this model which has a relationship with one of the entities from the library. I know I could modify the original, but is there a way to without needing to pollute the library? I tried just creating a new model with an entity named the same, but that doesn't work: * Terminating app due to uncaught exception 'NSInvalidArgumentException', reason: 'Can't merge models with two different entities named 'Host''

    Read the article

  • SSIS Debugging Tip: Using Data Viewers

    - by Jim Giercyk
    When you have an SSIS package error, it is often very helpful to see the data records that are causing the problem.  After all, if your input has 50,000 records and 1 of them has corrupt data, it can be a chore.  Your execution results will tell you which column contains the bad data, but not which record…..enter the Data Viewer. In this scenario I have created a truncation error.  The input length of [lastname] is 50, but the output table has a length of 15.  When it runs, at least one of the records causes the package to fail.     Now what?  We can tell from our execution results that there is a problem with [lastname], but we have no idea WHICH record?     Let’s identify the row that is actually causing the problem.  First, we grab the oft’ forgotten Row Count shape from our toolbar and connect it to the error output from our input query.  Remember that in order to intercept errors with the error output, you must redirect them.     The Row Count shape requires 1 integer variable.  For our purposes, we will not reference the variable, but it is still required in order for the package to run.  Typically we would use the variable to hold the number of rows in the table and refer back to it later in our process.  We are simply using the Row Count as a “Dead End” for errors.  I called my variable RowCounter.  To create a variable, with no shapes selected, right-click on the background and choose Variable.     Once we have setup the Row Count shape, we can right-click on the red line (error output) from the query, and select Data Viewers.  In the popup, we click the add button and we will see this:     There are other fancier options we can play with, but for now we just want to view the output in a grid.  WE select Grid, then click OK on all of the popup windows to shut them down.  We should now see a grid with a pair of glasses on the error output line.     So, we are ready to catch the error output in a grid and see that is causing the problem!  This time when we run the package, it does not fail because we directed the error to the Row Count.  We also get a popup window showing the error record in a grid.  If there were multiple errors we would see them all.     Indeed, the [lastname] column is longer than 15 characters.  Notice the last column in the grid, [Error Code – Description].  We knew this was a truncation error before we added the grid, but if you have worked with SSIS for any length of time, you know that some errors are much more obscure.  The description column can be very useful under those circumstances! Data viewers can be used any time we want to see the data that is actually in the pipeline;  they stop the package temporarily until we shut them.  Also remember that the Row Count shape can be used as a “Dead End”.  It is useful during development when we want to see the output from a dataflow, but don’t want to update a table or file with the dataData viewers are an invaluable tool for both development and debugging.  Just remember to REMOVE THEM before putting your package into production

    Read the article

  • inserting new relationship data in core-data

    - by michael
    My app will allow users to create a personalised list of events from a large list of events. I have a table view which simply displays these events, tapping on one of them takes the user to the event details view, which has a button "add to my events". In this detailed view I own the original event object, retrieved via an NSFetchedResultsController and passed to the detailed view (via a table cell, the same as the core data recipes sample). I have no trouble retrieving/displaying information from this "event". I am then trying to add it to the list of MyEvents represented by a one to many (inverse) relationship: This code: NSManagedObjectContext *context = [event managedObjectContext]; MyEvents *myEvents = (MyEvents *)[NSEntityDescription insertNewObjectForEntityForName:@"MyEvents" inManagedObjectContext:context]; [myEvents addEventObject:event];//ERROR And this code (suggested below): //would this add to or overwrite the "list" i am attempting to maintain NSManagedObjectContext *context = [event managedObjectContext]; MyEvents *myEvents = (MyEvents *)[NSEntityDescription insertNewObjectForEntityForName:@"MyEvents" inManagedObjectContext:context]; NSMutableSet *myEvent = [myEvents mutableSetValueForKey:@"event"]; [myEvent addObject:event]; //ERROR Bot produce (at the line indicated by //ERROR): *** -[NSComparisonPredicate evaluateWithObject:]: message sent to deallocated instance Seems I may have missed something fundamental. I cant glean any more information through the use of debugging tools, with my knowledge of them. 1) Is this a valid way to compile and store an editable list like this? 2) Is there a better way? 3) What could possibly be the deallocated instance in error? -- I have now modified the Event entity to have a to-many relationship called "myEvents" which referrers to itself. I can add Events to this fine, and logging the object shows the correct memory addresses appearing for the relationship after a [event addMyEventObject:event];. The same failure happens right after this however. I am still at a loss to understand what is going wrong. This is the backtrace #0 0x01f753a7 in ___forwarding___ () #1 0x01f516c2 in __forwarding_prep_0___ () #2 0x01c5aa8f in -[NSFetchedResultsController(PrivateMethods) _preprocessUpdatedObjects:insertsInfo:deletesInfo:updatesInfo:sectionsWithDeletes:newSectionNames:treatAsRefreshes:] () #3 0x01c5d63b in -[NSFetchedResultsController(PrivateMethods) _managedObjectContextDidChange:] () #4 0x0002e63a in _nsnote_callback () #5 0x01f40005 in _CFXNotificationPostNotification () #6 0x0002bef0 in -[NSNotificationCenter postNotificationName:object:userInfo:] () #7 0x01bbe17d in -[NSManagedObjectContext(_NSInternalNotificationHandling) _postObjectsDidChangeNotificationWithUserInfo:] () #8 0x01c1d763 in -[NSManagedObjectContext(_NSInternalChangeProcessing) _createAndPostChangeNotification:withDeletions:withUpdates:withRefreshes:] () #9 0x01ba25ea in -[NSManagedObjectContext(_NSInternalChangeProcessing) _processRecentChanges:] () #10 0x01bdfb3a in -[NSManagedObjectContext processPendingChanges] () #11 0x01bd0957 in _performRunLoopAction () #12 0x01f4d252 in __CFRunLoopDoObservers () #13 0x01f4c65f in CFRunLoopRunSpecific () #14 0x01f4bc48 in CFRunLoopRunInMode () #15 0x0273878d in GSEventRunModal () #16 0x02738852 in GSEventRun () #17 0x002ba003 in UIApplicationMain () solution I managed to get to the bottom of this. I was fetching the event in question using a NSFetchedResultsController with a NSPredicate which I was releasing after I had the results. Retrieving values from the entities returned was no problem, but when I tried to update any of them it gave the error above. It should not have been released. oustanding part of my question What is a good way to create this sub list from a list of existing items in terms of a core data model. I don't believe its any of the ways I tried here. I need to show/edit it in another table view. Perhaps there is a better way than a boolean property on each event entity? The relationship idea above doesn't seem to work here (even though I can now create it). Cheers.

    Read the article

  • Proper Use Of HTML Data Attributes

    - by VirtuosiMedia
    I'm writing several JavaScript plugins that are run automatically when the proper HTML markup is detected on the page. For example, when a tabs class is detected, the tabs plugin is loaded dynamically and it automatically applies the tab functionality. Any customization options for the JavaScript plugin are set via HTML5 data attributes, very similar to what Twitter's Bootstrap Framework does. The appeal to the above system is that, once you have it working, you don't have worry about manually instantiating plugins, you just write your HTML markup. This is especially nice if people who don't know JavaScript well (or at all) want to make use of your plugins, which is one of my goals. This setup has been working very well, but for some plugins, I'm finding that I need a more robust set of options. My choices seem to be having an element with many data-attributes or allowing for a single data-options attribute with a JSON options object as a value. Having a lot of attributes seems clunky and repetitive, but going the JSON route makes it slightly more complicated for novices and I'd like to avoid full-blown JavaScript in the attributes if I can. I'm not entirely sure which way is best. Is there a third option that I'm not considering? Are there any recommended best practices for this particular use case?

    Read the article

  • Auto-Configuring SSIS Packages

    - by Davide Mauri
    SSIS Package Configurations are very useful to make packages flexible so that you can change objects properties at run-time and thus make the package configurable without having to open and edit it. In a complex scenario where you have dozen of packages (even in in the smallest BI project I worked on I had 50 packages), each package may have its own configuration needs. This means that each time you have to run the package you have to pass the correct Package Configuration. I usually use XML configuration files and I also force everyone that works with me to make sure that an object that is used in several packages has the same name in all package where it is used, in order to simplify configurations usage. Connection Managers are a good example of one of those objects. For example, all the packages that needs to access to the Data Warehouse database must have a Connection Manager named DWH. Basically we define a set of “global” objects so that we can have a configuration file for them, so that it can be used by all packages. If a package as some specific configuration needs, we create a specific – or “local” – XML configuration file or we set the value that needs to be configured at runtime using DTLoggedExec’s Package Parameters: http://dtloggedexec.davidemauri.it/Package%20Parameters.ashx Now, how we can improve this even more? I’d like to have a package that, when it’s run, automatically goes “somewhere” and search for global or local configuration, loads it and applies it to itself. That’s the basic idea of Auto-Configuring Packages. The “somewhere” is a SQL Server table, defined in this way In this table you’ll put the values that you want to be used at runtime by your package: The ConfigurationFilter column specify to which package that configuration line has to be applied. A package will use that line only if the value specified in the ConfigurationFilter column is equal to its name. In the above sample. only the package named “simple-package” will use the line number two. There is an exception here: the $$Global value indicate a configuration row that has to be applied to any package. With this simple behavior it’s possible to replicate the “global” and the “local” configuration approach I’ve described before. The ConfigurationValue contains the value you want to be applied at runtime and the PackagePath contains the object to which that value will be applied. The ConfiguredValueType column defined the data type of the value and the Checksum column is contains a calculated value that is simply the hash value of ConfigurationFilter plus PackagePath so that it can be used as a Primary Key to guarantee uniqueness of configuration rows. As you may have noticed the table is very similar to the table originally used by SSIS in order to put DTS Configuration into SQL Server tables: SQL Server SSIS Configuration Type: http://msdn.microsoft.com/en-us/library/ms141682.aspx Now, how it works? It’s very easy: you just have to call DTLoggedExec with the /AC option: DTLoggedExec.exe /FILE:”mypackage.dtsx” /AC:"localhost;ssis_auto_configuration;ssiscfg.configuration" the AC option expects a string with the following format: <database_server>;<database_name>;<table_name>; only Windows Authentication is supported. When DTLoggedExec finds an Auto-Configuration request, it injects a new connection manager in the loaded package. The injected connection manager is named $$DTLoggedExec_AutoConfigure and is used by the two SQL Server DTS Configuration ($$DTLoggedExec_Global and $$DTLoggedExec_Local) also injected by DTLoggedExec, used to load “local” and “global” configuration. Now, you may start to wonder why this approach cannot be used without having all this stuff going around, but just passing to a package always two XML DTS Configuration files, (to have to “local” and the “global” configurations) doing something like this: DTLoggedExec.exe /FILE:”mypackage.dtsx” /CONF:”global.dtsConfig” /CONF:”mypackage.dtsConfig” The problem is that this approach doesn’t work if you have, in one of the two configuration file, a value that has to be applied to an object that doesn’t exists in the loaded package. This situation will raise an error that will halt package execution. To solve this problem, you may want to create a configuration file for each package. Unfortunately this will make deployment and management harder, since you’ll have to deal with a great number of configuration files. The Auto-Configuration approach solve all these problems at once! We’re using it in a project where we have hundreds of packages and I can tell you that deployment of packages and their configuration for the pre-production and production environment has never been so easy! To use the Auto-Configuration option you have to download the latest DTLoggedExec release: http://dtloggedexec.codeplex.com/releases/view/62218 Feedback, as usual, are very welcome!

    Read the article

< Previous Page | 22 23 24 25 26 27 28 29 30 31 32 33  | Next Page >