I was wondering if anyone had a method of running a report to see what a user is doing in Epicor or what they are printing. We are having users report that in the middle of the night, when no one is here at the plant, there are 500 page reports being printed. We are able to see in the print queue who printed what, but the report doesn't match up with anything in our system. We would have for example a report called DailySales.rpt, but in the printer queue it would be something like hb986a87dthr.rpt. Just wondering if anyone else has seen this, or would have a solution that would let me see what a user is printing.
It is not possible to link the print job directly to the SysTask record because neither the print job number, the temp file, nor mac addresses are saved in Epicor for cross referencing. It can be approximated by looking at runtimes and the SysTask record.
You can create a BAQ and BAQ report to display Active and recently completed SysTask information by user. This will give you the report run, start/end times, user, and current status. If you need more detailed information such as the criteria used in the report, you can also join to the SysTaskParam table. Keep in mind that the SysTaskParam table is fully normalized by field name, so you may want to join multiple copies of the table with specific criteria if you need a lot of information. Unfortunately, for "print all pages" jobs, Epicor doesn't know how many pages the report will be until after the data is instantiated and then it is rendered in the reporting software, so you won't be able to get any estimate of number of pages or size.
There are many strategies for mitigating the issue you described. Here are a couple:
You can use criteria within the BAQ to limit the number of records returned for a specific query
You can create a subquery criteria from BAQ parameters to return no data when abnormally open parameters are used for the report (e.g. > 30 days range). You could also use this method with time gates based on the current system time.
Retrain users
Related
I have a report on PowerBI that has many pages/tabs and each one also has alot of data being displayed. As I didn't design this, I'm going through the report to eliminate as much as I can and possibly splitting the report as alot of the data only requries refreshing once a week.
This is where my query comes in, I have information on one page that requires a refresh every two hours over a 12 hour duration, one field of data that requires a daily refresh and two more fields only require refreshing when required.
Is it possible to segment scheduled refreshes throughout a single part of the report, or does scheduled refresh only allow the entire report to be refreshed? (I.E. Sales status is hourly, Outbound status is daily, and sales summary is weekly)
I'd rather avoid having to split reports, as it is very handy to have them on one page; rather than having to open two and view them on multiple monitors.
I am just starting out on PowerBI reports, having been shown enough to get what I need done; but plan to delve further in, this being my first port of call if it is possible.
Thanks for any reponses in advance.
Brian.
No. It's Not Possible.
PowerBI Internal working like Tabular Model.
In Import mode we can not do incremental refresh also.
So other option is you can create Reporting layer and define denormilized with calucaluated columns Reporting tables.( Sales ,summary )
and use Direct query or Refresh and Do ETL for This table.
So you can schedule ETL for specific Tables i.e.Sales or summary.
The company I work for has a parcel-tracking software that we sell to our customers. It uses SQL-database to store the data. And to create reports for the parcels delivery times and other stuff, like exceptions, costs, signatures, we use crystal reports.
One of our customers would like a SLA report, where they want to track the time between a certain location to the time where is has been delivered.
The formula must contain:
Package Status, in this case it's LOCSCAN
Package Location, it's X-ray
And the time that is has been scanned to that location.
I have all the data in the SQL-database, but I cant figure out the formula for this.
A brief description of the workflow:
The parcel arrives at the site.
They scan it in to the system, thus gets an received date/time, location:Goods and status: RECEIVED.
After that they put the parcel trough a x-ray scanner, so it doesnt contain any illegal stuff. Scans the parcel with the new status: LOCSCAN, location:X-ray and a new date/time.
After that, they deliver the parcel, to the receiver, and gets status: DELIVERED with a new date/time.
All the date/time fields are seperated from eachother, stored in individual columns, dependent on the status of the parcel.
I'm using Crystal Reports, where I'm trying to create the formula.
Use the datediff function:
datediff('d', {startingdate}, {endingdate})
First a formula for RECEIVED to LOCSCAN and then LOCSCAN to DELIVERED if I'm understanding correctly.
The 'd' stands for days, you can replace with 'h' for hours or 's' for seconds. The fields between {} are the ones from database.
I've looked all over and cannot find an answer to my question; I can't even determine whether it is possible.
Referring to the attached image, you will notice that this is a statement report with data grouping activated.
1) The report shows all the services invoiced to an account by date.
You can expand the group to see all the transactions that formed part of that service for that day. (You can for instance make use of the same service multiple times per day)
2) This is the detailed layout of the service invoiced. This list is different for each service, but mainly it will show you a summarized transaction list (PK BatchId), which has the "+" symbol next to it to enable drilldown to a detailed report of the batch.
My problem:
When loading the statement report, we are now hitting multiple tables, multiple times to produce the data to be grouped and displayed in #2 (refer to image).
We are trying to avoid this like the plague.
My Question
Is there a way to populate #2 when and only when the user clicks a "+" symbol or an "expand" image where the "+" is currently located in #1.
In other words. We dispose of the group function and populate the statement without detailed information. When the user clicks on #1, we load a sproc, populate a dataset and display the data in #2.
Any thoughts on this?
Drillthrough Reports look like a good solution here. See the link for more information on how these work. So basically you have the report without the detailed information, but when somebody clicks on 1 it opens up a new report with the details behind it.
After testing, I confirmed that subreports are executed even if they are hidden within an element that can be toggled.
So subreports won't answer this problem.
[Edited: previously I thought they could be used. JAT points out that this negative answer may have some value, so I'm leaving it.]
I have a report with 4 seperate sub-reports embedded into it.
On the main report, I have a parameter to filter by group and I have it set to filter by group unless you choose All for the parameter, then it shows all groups.
If {?Main Group Name} = "ALL" then {Incident.Group Name} like "*"
else {Incident.Group Name} = {?Main Group Name}
This filter setup works fine on the individual sub reports, however if I run it selecting "all" as part of the main group, it causes the sheet to loop indefinitely and doesn't show the data properly.
The data is in all Group header 1 and group footer 1, so ideally the same page(as below) should repeat for every group.
When I run it for all, It runs for an abnormally long time, and has to be manually stopped, otherwise it just keeps going. After manually stopping, I get the following. It almost looks like it's trying to render all the groups on top of each other.
Anyone have a clue what could cause this?
Where are your subreports located in the main report (group header)? What are you doing with the record details of the main report (suppressing, I'm guessing)? What do the record selection formulas look like, FROM the subreports? How many tables are in the subreports? If multi-table, are your joins correct?
I have seen occurrences when subreports have a given Record Selection, and then have links added from the main report, where the resulting subreport query is totally hosed as Crystal Reports "helps" by modifying the subreport record selection on top of what the developer intended. True Story (a 17 second report was updated and ran for 20 hours before anyone noticed it had never finished a scheduled run). It sounds (heh, in inter-text) like you don't have correct relationships set up in the subreports, and may actually have cartesean products being formed.
Where are you using your code shown in your question? In the Record Selection? In an #Formula? What kind of link are you using to the subreport? What "should" be linked to the subreports is the {Incident.Group Name} -- as that is what changes in the main report group.
If you break out the subreports, and run them individually, manually setting the {?Main Group Name} to "ALL" and then to sample group names, what is the result? If the individual subreports still run on, you can at least watch the record counts (status bar) and when you manually kill the report, you can look at the performance information, and at the subreport Show Query.
Lastly, if you try running the main as "ALL" using a Top 10 (or 2, or whatever number you want) and then watch the subreport performance, that might give some clues. If there are selection-hosation byproducts, this would at least limit the effects a bit. You'd still be passing "*" to the subreport, but only doing it "N" cycles.
Sigh, and one last: If you really want "all", you might be better served if you use something like:
If {?Main Group Name} = "ALL" then TRUE
else {Incident.Group Name} = {?Main Group Name}
Sorry to be so long... I didn't have enough time to shorten it.
OK, really, really LAST one: If I have misunderstood your post, assumed too many disparate themes, or am just talking out of my hat: Sorry. Missed a meal, picked up several interruptions, and doubled up on the coffee. I'm sure there's a smiley for that, somewhere.
[Inserting add'l answer, as it is distinct enough to warrant it's own entry.]
I think I understand the Group Header/Foot Charts.
Did you try running Main as a Top 10 to limit the # groups?
How big is the (chart) record set?
AND, is the report set up for asynchronous queries?
What IS going on in the subreports:
Complex query?
Simple query?
Suppressing all Details and just showing the Chart?
Going to a different table/database?
Table joins?
Run for "Top 10" Groups in main, wait until it ends, then check for the Main and each chart sub (for at least a few groups):
Report / Performance Information:
Total Number of Records -- (Main should be "Top 10" )
Performance Timing:
Run the Query,
Read Database Records,
Time to Format All Charts
or
Time to Format 1st Page.
If you have several millions of records, even with "top", it could take a while to read and calculate the top. If you are calling charting, even tho' it is for a single Group per chart, if the subreport still has to apply a linked query, parse millions of records, perform add'l calculations, create a chart (that still accounts for all records), etc., it can just be a long time running.
You may find that a small "All" Top "n" will show a lot of time being spent behind the scenes, and that allowing the whole report to run that the add'l Comparison of Ticket Counts is displayed. And if not, we can look at that later (mainly by opening up the corpse and looking inside).
I am trying to ascertain regular behaviour, and I can't get a requirements doc for review. I can only make assumptions and ask questions and make suggestions.
I am using Crystal Reports 8.0.1.0 and have an OLEDB connection set as "favourite".
We have several reports being run against a few stored procedures on the database.
Our database is an SQLExpress 2005 named instance.
One of the reports deals with balance, i.e. we have to provide a balance report for a specific company or set of companies.
The customer, however, has specifically asked us to have a separate section to be the last one for a specific economic account typology and in order to do this, we have to separate this one from the others which are instead displayed using two parallel subreports.
Hence our structure should be like follows:
GENERAL HEADING
PATRIMONIAL STATE
...data...
ECONOMIC COUNTS
...data...
THIRD-PARTY EFFECTS
...data...
The first two sections below the general heading are achieved with two parallel subreports, so now we have to display the third section.
Is it clear up to now? Well, as you may well be aware, CR subreports require you to set the database location, regardless if you have set one up already for main report.
The problem, however, is that if I try to set the location pointing to our stored procedure(which returns ALL the data for the balance report), the first time it will display the message in subject, and the second time again, but this time it'll also crash.
Our stored procedure has a parameter and both previous subreports(the ones in parallel) are set to the same SP and are working flawlessly.
This crash makes it impossible for me to work on the third section and I'm pretty much at loss as to what would be the best approach to solve the issue at hand.
Any ideas/suggestions?
I would love to receive your answers and I hope my problem was clear enough(I'm very willing to further delve into it for clarification if ever needed).
Thank you so much for your time,
Andrea Raimondi
Geez... changing the connection from OLEDB to SQLServer it correctly fetches the rowset and doesn't crash!
I am speechless.
Andrew