Power BI Refresh of TransactionLinesIncremental fails with itgenboe367

Our Power BI daily refreshes keep failing regularly:

Request failed: The operation has timed out.
DataSourceKind = OData.
DataSourcePath = https://bridge-online.cloud/acme-exact-online/odata4/ExactOnlineREST.Incremental.TransactionLinesIncremental@eol
Request failed:
The operation has timed out.
The exception was raised by the IDataReader interface.
Error code: itgenboe367
The data download was cancelled straight away after requested.
Please try again. Contact Invantive Support when this problem persists.

Possibly related to Itgenboe242 op elke OData request (Dutch).

Can you add the request ID from Bridge Online Monitoring (see More insight with new Bridge Online Monitoring)? It is in the title bar when you click on the request.

Request id is:

0HN27CG692FFH:0000002D (HTTP/2)

The download started at 16:07:28 and was considered canceled within 10 ms, going from itgenclr150 to itgenboe367. It returned a HTTP status 499.

The URL visited seems to be /setup, which is a list of available databases. Managed thread ID is 206.

The session has been restarted. This type of problem may occur at rare occassions, but should not occur consistently across multiple days.

Question: did restart of the session solve the problem for now?

In case the problem resurfaces within 7 days, please add a reply to this topic and include the request ID from Invantive Bridge Online Monitoring Details.

The same problem occurred today noon twice (19 maart). The request IDs are:

  • 0HN27GF812EHN:00000015
  • 0HN27GF812EHN:0000000F

The error keeps occurring:

  • 0HN28PUQEV7E2:0000003D (HTTP/2)

Similiar issue on same table on current release:

At approximately what date did this problem start occurring for you?

Is the problem occurring consistently (i.e. on every download of this specific table)?

The weird part is that only some of the users experience the problem.

Since a few days (as of approx 15 March). Occurs almost during every overnights refresh.

Thank you for the additional information!

No root cause has yet been established. A new release with additional logging facilities will be taken into production before Monday, so hopefully this will shed some more light on the possible cause(s).

Are there are any updates? The data is still not refreshing, which is unacceptable for the correct work of our financial team.

The problem is still occurring for a few users. For some consistently and for some with intervals.

The problem has been analyzed, but the root cause is not yet clear. The team is currently working on finding a solution.

This topic will be updated when there is progress.

The problem occurs between Invantive Bridge Online and the UniversalSQL engine. Therefore, a possible temporary workaround is to copy the to an Azure database or text files, such as using the instructions on Elementary Data Replication Module between Exact Online and Azure SQL Server.

Since 16/3 it has not been possible to refresh.
This problem occurs on every download of this specific table.

Is there a filter on partition/company on this table?

So far, the signals are that the problem only occurs when there is a single partition selected.

There was a filter on one partition.
I have removed this filter.
I tried to refresh, it started. Refresh is still running at 13:24 but in monitoring last step was at 12:59:

31 27-03-2024 12:59:27.412 itgenboe152 DataRetrieval Tot zover 6,000 rijen teruggegeven voor query op ‘Invantive.ExactOnlineREST.Incremental.TransactionLinesIncremental@eol’ (geschatte grootte 7,873,940 bytes).

It looks like it stopped downloading at 6000 rows.

Thank you for the additional information. We are uncertain whether it can also occur when multiple companies are selected. However, 6.000 rows in 25 minutes is extremely slow, so probably the issue also occurs for some users with multiple companies.

A new release has been taken into production, which should solve most or all direct cancelations and endless downloads of *Incremental tables on Exact Online. For details refer to Intermittent broken downloads on *Incremental-Exact Online tables.

A small percentage of issues remained. A fix for these has been taken into production as 24.0.114.