Hello,

I have some disconnection issues after 5 minutes of inactivity, with Firefox only, after opening a QS app passing by the NLB (F5) address.

If I try to connect directly to the address of one of my nodes, I don't have this disconnection issue.

In the AuditActivity_Proxy log, a specific message is visible each time i'm disconnected :

I found some articles about Websocket ping intervall depending on browser, but I really think that it comes more from a parameter at the load balancer level. However, I don't understand why this issue only concerns a particular browser (Firefox) in my case.

Session inactivity timeout is fixed on 30 minutes on each node :

Keep alive timeout is fixed on 10 seconds :

Can you give me some ideas to explore in order to resolve this issue?

Thanks,

Emmanuel

I have a master measure (let's call it "B"), which references another master measure (let's call it "A"). Master measure "B" works fine.

Then when I create a new master measure "C" (which references master measure "B"), I get the error referenced measure, which Qlik has documented here. https://community.qlik.com/t5/Official-Support-Articles/Error-in-expression-Error-referenced-measure/ta-p/1845945

I've made sure my master measure names are unique and non-overlapping with other names. Is nesting master measures more than one level deep not supported by Qlik?

I have a table with 100 columns

I want to check for column 1 = 'ID001' if the other 99 columns has a certain date value say '2025-01-01'

But my code is throwing the error fieldname not found in the for loop...

what i did

// first filter column 1 for ID001

[TEST]: NoConcatenate LOAD Distinct * Resident[MAIN_TABLE] where column1='ID001';

// Loop to check for date value =  '2025-01-01' if have flag as 1

For f = 1 to NoOfFields('TEST')
LET F1 = FieldName($(f), 'TEST');
[Fields]:
Load distinct FieldName($(f), 'TEST') as Field, If(Date(Floor($(F1)))=Date(Floor(' 2025-01-01')),1,null()) as check
AutoGenerate 1;
Next f
DROP Table[TEST];

Any idea how to solve? Thanks in advance!

Prezados, 

Tenho um gráfico combinado 1 dimensão e 2 Medidas, porem preciso alterar as cores.

Dimensão: Mês/Ano

1º Medida: Relacionada ao valor de uma medição realizada mensalmente.

2º Medida: Meta, cujo a medição tem que alcançar.

A 1º Medida está em barra, preciso pintar a barra caso estiver abaixo da meta em vermelho, e igual ou acima em verde, já consegui fazer pela opção de cores por expressão. Porem a 2º medida que seria a meta, preciso deixar em outra cor, mas como estou usando a opção de cores por expressão o qlik por padrão deixa a 2º medida que está em linha da mesma cor.

Existe alguma maneira de deixar a minha 2º Medica com outra cor?

Segue print do gráfico combinado.

Hi everyone,

I am facing an issue as I am trying to build a flat file to import data in a software (standard format defined by the software editor). For each of my records, I have to go through several outputs I created and this order must be respected. At the moment, my job goes through the first output for all the records and then switches to the next output.

In my case, I would like it to go through each output, one by one, with one record before it switches to the next record.

How can I do this ? Please find attached a screenshot of my job.

Thanks.

Hi!, reading documentation about Qlik Web storage provider 'Google Cloud Storage', it says that you only can put data into a bucket in Qlik SAAS, are there any plans to have this in Qlik Sense Enterprise?

Storing data from your Qlik Sense app in Google Cloud Storage

You can store table data into your Google Cloud Storage bucket in the Data load editor. You can either create a new load script or edit the existing script.

https://help.qlik.com/en-US/connectors/Subsystems/Web_Storage_Provider_Connectors_help/Content/Connectors_StorageProvider/Storage%20provider%20connectors/Google-Cloud-storage-provider-connector.htm

Hi everyone,
I have this bar chart of the number of requests per quarter year.
There is a clear ascending linear trend that I want to visualize with a trend line. The "naive" built-in trendline from Qlik Sense is biased by the incomplete data of the current, incomplete quarter year as shown in the figure. The reasonable thing to do is to base that trend line on the years before YearStart(Today()).
I guess this is a pretty common thing, but I can't figure out how to implement this. Do you have a hint?

What I tried so far:

• Checking similar topics here and there.
• To calculate the intercept and slope of the regression line, I use Linest_B() and Linest_M().
• As X values, I use numeric value of the quarter year: (Num#(Left([Reception YearQuarter],4))+(Num#(Mid([Reception YearQuarter],7,1))-1)/4))
• As Y values, I use Aggr(Count(RequestID), [Reception YearQuarter]) over the set of Reception Dates from previous years {1<[Reception Date]={"<$(=Date(YearStart(Floor(Today()))))"}>}, or as a complete expression: Aggr(Count({1<[Reception Date]={"<$(=Date(YearStart(Floor(Today()))))"}>} RequestID), [Reception YearQuarter]).
• In a table showing Reception Year-Quarter (both text and numeric) and Count(RequestID), I succeed in correctly calculating Linest_B(Y values, X values) and Linest_M(...) as additional measures, but not Linest_B(...) + Linest_M(...)*X-value
• As a reference line this does not work either.

AggregatedSampleData:
LOAD * INLINE [
Reception Year-Quarter, Count
2021-Q1, 17871
2021-Q2, 17515
2021-Q3, 16443
2021-Q4, 19041
2022-Q1, 19297
2022-Q2, 18020
2022-Q3, 16881
2022-Q4, 19582
2023-Q1, 21232
2023-Q2, 19139
2023-Q3, 18494
2023-Q4, 21002
2024-Q1, 21784
2024-Q2, 1790

];