I am using the "SSIS Log Provider for SQL Server" to log events to a table for "OnError" and "OnPostExecute" events of a package. This works as expected and provides a nice clean output on the execution steps of the package.
I am curious as to why I do not see any detail for any/all tasks that fall under the "Data Flow" section of the package though. For instance, on my "Control Flow" tab, I added a "Data Flow" task that simply loads a few tables from a target to destination server. However, there is nothing shown in the logging output. Just that a Data Flow task was initiated. And when I'm configuring this logging under "SSIS-->Logging" in the checkbox area on the left, you cannot "drill into" data flow steps.
Is there a reason why there is no detailed logging for Data Flow tasks? Would getting to that require me to create a custom log provider?
Has anyone come up/determined a generic way to capture and log indicative information within a data flow in SSIS - e.g., a number of rows selected from the source, transformed, rejected, loaded, various timestamps around these events, etc.? I am trying to avoid having to build a custom solution for each of the packages that I will have (of which there will be dozens). Ideally, I'd like to have some sort of a generic component (such as a custom transformation) that will hide the implementation details and provide a generic interface to the package.
It is not too difficult to achieve something similar on the control flow level, but once you get into data flows things get complicated.
Does anyone know how to hook up to data flow pipeline events via custom solution (C#)? I am trying to write code to log start and end times of components(lookup,merge joins etc) in a data flow task. I tried with a class inheriting from the EventsProvider class but it didn't work as this is only for container tasks. Any ideas will be greatly appreciated.
I would like to fetch the data flow component name while package is executing. Since system variable named [System::SourceName] only fetches name of the control flow tasks? Is there a way to capture them?
Hi, I have one data flow with 10 sources and destinations in the flow. For the sources I'm using the datareader for an odbc and for the destination I'm using the ole db destination source. By default these run parallel when executed. Is there a way in the dataflow to run them step by step instead of creating 10 different data flow tasks in the control flow? Or is it better to have 10 different data flow tasks?
I'm creating a SSIS in the designer view of SQL Server BI Dev. Studio (SQL Server 2005)
I need to import a whole table from MS Access into my local SQL Server.(this task will be performed weekly, so once working I'll schedule a job for it)
I've created a 'FILE' connection to MS Access in the 'Connection Managers'.
When I'm on the 'Data Flow' tab I can't find a Data Flow Item to use as a MS Access connection. (available on the 'Data Flow Sources' are only: DataReader, Excel, Flat File, OLE DB, Raw File and XML Sources)
The issue is in the data flow for loading and setting the Fact table dimension keys (the dimensions are all loaded fine). After 16 rather pedestrian Lookup Transformations, I have an escalating problem adding additional Lookup transforms to the Data Flow. The problem is not in execution; the problem is adding more transforms in design mode.
Lookup # Fields in Data Flow Time to validate that lookup <17 47 Sub-second 17 48 2 sec 18 49 4 sec 19 50 8 sec 20 51 16 sec 21 52 32 sec 22 53 64 sec
While I€™m intrigued by the mathematical progression that is forming here, the issue is that I have at least 6 more Lookups to perform. I hope you can see my dilemma.
I have gone to where it takes a little over 4 minutes each to validate the lookup transform and its associated Derived Column transform and Union transform (Total 12 Minutes). Not only does this add up to many idle minutes to each design step, BUT it breaks the debugger as it pre-validates the ENTIRE data flow before it ever switches into debugging mode.
Some notes: 1. It doesn€™t matter what order the Lookup transforms occur in, the timings are exactly the same. 2. I tried many Data Flow execution optimizations, but they don€™t improve the validation times (or even get a chance to improve the execution times!)
I realize this may be somewhat of a unique problem.
I'm just starting off in SSIS and have a question that I can't find an answer to...
I'm loading in a number of files (in separate Data Flows) and performing some transformations on them before merging them back together. What I'm not sure about is what I should be doing with the data at the end of each of my "Import Data From XXXX Flat File" Data Flows. Am I better off using OLE DB Destinations (or SQL Server Destinations) and saving this intermediate data to temporary tables, or am I better off using a Raw File Destinations and saving this intermediate data to files? Or is there, perhaps, a better option that I'm currently unaware of?
If the Raw File Destination is the way to go, then isn't there a maintenance issue with cleaning up all the files created? And will there not be a management issue to ensure that there is sufficient disc space available on the drive you are saving to?
I'm a bit confused and overwhelmed by SSIS at the moment, so any help would be much appreciated!
We're experiencing a problem where intermittently our SSIS packages will hang. There are no log errors or events in the event viewer. It will happen whether the package is executed from the SQL Job Agent or run from BIDs. When running from BIDs it appears to hang inside one of the data flows (several parallel pipes with sorts, merge joins etc...). It appears to hang in multiple pipes within the data flow component. The problem is reproducable, we just kill it and re-run, and it appears to hang in the same places.
Now here's the odd thing: as we simply open and close some of the components in the pipe line after the place it hangs, a subsequent run will go further in the pipeline before hanging. If we open and close all the components after the point it initially hung, the data flow will run fine, from there on out. When I say "open and close" I mean no changes are made, we simply double-click the component, like a merge join, then click 'close.'
To me this does not seem like a memory problem but likely something is wrong with the metadata, where opening a component and closing it somehow alters the metadata to "right it".
This seems to occur intermittently after we make modifications to the package. It's like if you make any mod, even unrelated to the data flow, you then have to go through and open and close every component in your package to ensure it will work. Again, no errors or warnings are fired.
hi in my package, some sql operations need the special user name and admin privilage. so how do i create my ssis package so that when it executes it takes the given username and password from the table in some database.
We have a package that has a connection called Load_DimItem.trc. We don't need this logging enabled for this package anymore. However, if I delete the connection, and delete the log provider (SSIS log provider for SQL profiler), I get errors when trying to close the package after debugging. I get: "Cannot detach from one or more processes. [3172] The object invoked has disconnected from its clients."
I am having the same problems as those in another post. SSIS package works fine when executed in BIDS and through execute package utility but it doesnt work when executed as a step in a job.
The other problem is that the logging also doesnt work when i try executing it as a job. So I have no clue about what to do without knowing what error it is. When I run the job it simply says the step has failed.
I have tried most of the solutions posted in other websites most of them to do with using proxies with credentials but havent hit a solution. I would love to get any input on what to do.
I am logging all the tasks in my SSIS package to SQL Server. For each task I am logging atleast the Pre-Execute, PostExecute, OnError events. For Script tasks, I have custom logging and I am logging the ScriptTaskLogEntry event too.
When I run the package manually from BIDS, the logging works great ! But when I try to run the package from a job or from the command line, the number of events that is logged greatly reduces. For eg. when manually run, I get 104 records in the log table but when run from the command line I get 23 records only. Most of the custom logging messages from the Script Task do appear. Its the pre and post-execute events that are skipped. Any idea why ?
Here is command line from the job. I also use the same command line with "dtexec" from the command prompt.
Hi, I decided to use the SQL Server log provider to store logging data of all my Integration Services packages. I also created some reports about this data for operating purposes. I have a problem occurs the name of the executing package is not always written to the log,but the name of the single task which failed. But that is not very useful information for operating, because I do not see any chance to get the name of the package by the information which is logged in the sysdtslog90 table in the database which I defined for SSIS Logging.
How do I configure the package to always log the package information into the table, too?
In my control flow, I have a container which contains an Execute SQL Task, and then upon success, a Data Flow Task. The SQL Task truncates my datamart table. In the data flow task, I execute a stored procedure (through a variable) that populates that same datamart table. I can execute the stored procedure's select statement in Management Studio with no problems in about ten seconds. However, in the SSIS package, the SQL task completes successfully, and then it hangs indefinitely on the data flow step. In the Data Flow tab, none of the boxes are even turning yellow. Why won't it complete? When I move the Exec SQL Task to another container, the package executes fine, but it should be in the Load Phase container.
In the data flow task, i have done a group by and now i have a single row.... I want to assign the value in this row to a package variable.... Without using the script component .......Any suggestions ??
I need to see inside a SSIS 2012 project a new SSIS installed component, but in the SSDT 2010 I cannot see the SSIS Data Flow Items tab for adding data source/data destination respect to the choose toolbox items pane.
I need to call a stored procedure to insert data into a table in SQL Server from SSIS data flow task. I am currently trying to use OLe Db Destination, but I am not sure how to map inputs to OLE DB Destination to my stored procedure insert. Thanks
I am using the event handling mechanism to do my custom logging. This works fine. Using the OnPreExecute and OnPostExecute my log tables fill up with the start- and enddates of all the containers and tasks in the complete package hierarchy.
However, what I am missing (e.g. in a system variable) is the source ID of the parent container that started the task or container. In other words, in my logging reports, I would like to build up a tree starting with the topmost package, like the progress indication in the IDE.
The built-in Logging features do not log this information either as far as I can tell.
Hi, this is my first post and I'm relatively new to SSIS so please go easy on me.
Without going into too much detail about it, I've set up a simple SSIS package which does this in a nutshell:
Foreach loop picks up all *.xls files in a given folder 1 - Puts the name of the current spreadsheet into a variable 2 - File System Task copies the current spreadsheet ("abc.xls") to a file called "work.xls" 3 - Data Flow task performs data extraction on "work.xls" and puts it into a SQL server database 4 - File System Task moves "abc.xls" into a "success" folder Continues with loop - move onto next spreadsheet
This works fine, so long as the spreadsheets all have the same number of columns.
As soon as one of them has a column missing (believe me, this will happen - we're dealing with users here) the package falls over at step 3.
When the package comes across an erroneous spreadsheet, what I'd like to do is move the offending file to a failure folder (making step 4 either a success or failure file move) and carry on with the next one.
I know that you can have an error path (the red line) from any step within the dataflow task, but this doesn't help me because the error lies in the structure of the spreadsheet and not the contents.
I've already come up with a work around whereby each file is moved into the failures folder just after step 2, then moved from the failures folder into the success folder at step 4.
This almost gives me what I want, although of course the package still falls over whenever it encounters a dodgy looking spreadsheet.
Is there any way that I can get the package to do what I'm after?
In several threads there has been discussion regarding adding connection managers to a package's data flow, etc. My challenge is that I have a large solution that contains many packages, and I need to change the connection manager linked to the data flow in all of the packages. When the solution was initially designed, data sources were used, and it has become a tedious maintenance issue to keep those in sync. We want to use a standard OLEDB connection manager, but adding a connection manager to each package and editing the corresponding data flow tasks in each package to use that new connection manager is a daunting task. I've coded a .Net module to access the packages, remove the old connection manager (data source) and add the new OLEDB data source. However, as I traverse the objects in the package hierarchy, when I come to the data flow object, the innerobject is not a dts object, but rather a _com object.. I can't seem to find any documentation/examples as to how to iterate the tasks within a data flow and change the connection manager. If you have any information, that would be quite helpful. If you reply with a code sample, if you would be so kind as to relate it to one of the sample packages provided with SSIS so I can run it, that would be great.
I have Data Flow task that contains 50 components.
My computer configuration: 1 GB RAM Microsoft Windows Server 2003
Periodicaly when i try to save package after making some changes Out of memory ... exceptions message box appears , and soon after this Not fatal error occurs ... message box shows . If i close solution and open it again all my 50 components disappears --instead I see clear list, and all my work losen.
Such "Not fatal errors" making hell out of job -- every time I need to change package i must add package to archive!!!
There is a table with a column that contains Xml documents. For each record from my Data Flow Source, I want to pass in the Xml document and the node to interrogate, and return the value contained in the node. Like the Crm component, this is probably one I will have to write from scratch in C#, but I would like to avoid having to create the custom component if it already exists in the public arena.
Does anyone know of any Xml Ssis Data Flow Components that are downloadable for free?
I was working all day making changes to my 3MB package. I was adding a large number of transforms that were copied-and-pasted from elsewhere in the same data flow task.
All was going well. I even took the time to have SSIS lay out the task again (1/2 hour). Suddenly I started receiving some strange errors:
After the layout, I noticed two stray components 'way off in the upper right corner. I found that one of them had a duplicate name to a component which had been added hours ago. Even after deleting it, I got "duplicate name" errors.
I copied three components in one selection, and when I tried to paste them, got the error "can't initialize component on paste". I tried them one at a time, but got the same error.
I got errors about COM failures due to marshalling to another thread I then exited Visual Studio and started it again. To my great surprise, the data flow task I was working on was still there, but was completely empty.
Comparing what I'm left with to my last version in source control, I find that the entire pipeline element is missing from the DTS: ObjectData element!
I'm developing a real love/hate relationship with SSIS. It varies from one day to the next. Guess what kind of day this is!
I have a Data Flow Script Component(Destination Type) and in the properties I have a read/write variable called User::giRowCount
User::giRowCount is populated by a Row Count Component previously in the Data Flow.
After reading http://forums.microsoft.com/MSDN/ShowPost.aspx?PostID=97494&SiteID=1 it is very clear that you can actually only use variables in the PostExecute of a Data Flow Script Component or you will get an error "Microsoft.SqlServer.Dts.Pipeline.ReadWriteVariablesNotAvailableException: The collection of variables locked for read and write access is not available outside of PostExecute."
What I need to do is actually create a file in the PreExecute and write the number of records = User::giRowCount as second line as part of the header, I also need to parse a read/write variable such as gsFilename to save me hardcoding the path
-they must go in the PreExecute sub --workarounds please-here is the complete script component that creates a file with header, data and trailer --Is there any workaround
Public Class ScriptMain Inherits UserComponent 'Dim fs As FileStream Dim fileName As String = "F:FilePickUpMyfilename.csv" 'Dim fileName = (Me.Variables.gsFilename.ToString)
Dim myFile As FileInfo = New FileInfo(fileName) Dim sw As StreamWriter = myFile.CreateText Dim sbRecord As StringBuilder = New StringBuilder
Public Overrides Sub PreExecute()
sbRecord.Append("RECORD_START").Append(vbNewLine)
End Sub
Public Overrides Sub ParsedInput_ProcessInputRow(ByVal Row As ParsedInputBuffer)
Public Overrides Sub PostExecute() sbRecord.Append("RECORD_COUNT").Append((vbTab)) sbRecord.Append(Me.Variables.giRowCount.ToString).Append(vbNewLine) sbRecord.Append("RECORD_END").Append(vbNewLine) 'Now write to file before next record extract sw.Write(sbRecord.ToString) 'Clear contents of String Builder sbRecord.Remove(0, sbRecord.Length)
