Task reload failure

[#1]

Hello,

Few of the recurring tasks fail at random in my environment and when checked for the Log folder, i could not find file which captured the failure. What could be the reason for this behaviour?

However i have got the error from the tasks list and it is pasted below.

2017-02-21 11:00:39 UTC

Message from ReloadProvider: Task failed due to timeout getting engine connection

2017-02-21 11:00:39 UTC

Changing task state from Queued to FinishedFail

2017-02-21 10:30:39 UTC

Changing task state from Started to Queued

2017-02-21 10:30:38 UTC

Changing task state from Triggered to Started

2017-02-21 10:30:36 UTC

Trying to start task. Sending task to slave-node xxxxxx.xxxx

Thanks.

[igorgois_]

thanks for the answer, craig

we did what you said: improved the tasks start time to decrease the queue. Thus, we didnt get this error anymore

but yesterday the error came back... This is not a stable situation, if something goes different, the queue error will come back

we are also using sap connector here. I will delete the logs folder, but im not confident about this as root cause.

Does anyone has another clue?

Thanks in advance

[craigbence]

igor,

I agree the underlying reason could be "masked" by these changes. However, the two recommendations from Qlik were to 1) Clean out the logs and 2) stagger the tasks. So far - for us - this has worked, but again we agree this seems odd.

Alternatively, if the current situation does not completely solve the problem, are next step is to convert our long-running/high utilization tasks which use the Qlik SAP Connector to the standard SQL connector. Our understanding of the SAP Connector is that it uses SAP BAPI's to make the calls to the DB, so using a SQL connector instead would remove the additional, non-value-added, layer.

Again, these are just suggestions to try and help!!! GOOD LUCK!