cancel
Showing results for 
Search instead for 
Did you mean: 

Timeout Error in Get Next Item

PrajjwalRai
Level 5
Hi,
I have a process which runs on three resources parallelly. Recently, I have been getting the following error on the all the machines, sometimes together on two, sometimes after some time interval between each, but the process ends with the error on all within a maximum span of 30 min after the first failure occurs:

ERROR: Internal : Execution Timeout Expired.  The timeout period elapsed prior to completion of the operation or the server is not responding.

If someone has faced this issue, can they suggest on the best ways to tackle it, or what could be the reason that this issue is occurring.

Thanks!

------------------------------
Prajjwal Rai
Application Engineer
Schneider Electric Private Limited
------------------------------
3 REPLIES 3

Simon_AndrePede
Level 4

Hi Prajjwal, 

Hope you are well. 

I would say take a look at the database server and/or application server. It sounds like you have issues with retrieving data from the Database efficiently.

It could also be your queue item itself is very large in data. You could try disable logging and especially disable Parameter logging when you call the Action "Get Next Item". If you have enabled Parameter logging, it will take all the data in your queue item and put it in the session log. If this is a lot of data it can cause your server to timeout. 

I hope it gives a clue. 



------------------------------
Simon Andre Pedersen
Manager
PWC HK
Asia/Hong_Kong
------------------------------

Also have a look at this
http://portal.blueprism.com/customer-support/support-center#/path/1137522532
How do I fix a SQL Server "Timeout expired" error?

*Note: This KB article is also accessible in some additional languages within the Support Center. Please scroll to the bottom of the article for links to any other language versions that are currently available.
 

The following database error message originates from SQL Server:

 

"Timeout expired. The timeout period elapsed prior to completion of the operation or the server is not responding."

This may also result in the process terminating, or certain actions, such as updating or retrieving queue items, failing.

The error described indicates that the SQL Server failed to respond to a query within the timeout period with the results that Blue Prism asked for. Typically, this is caused by insufficient database maintenance, combined with too high a logging level.

See the following articles for information on appropriate logging levels within Blue Prism: 

Alternatively, this can be caused by an under-specified SQL Server database server, or other demands on the SQL server preventing it from responding in time. The following SQL Server factors may be involved:

  • a poorly designed database query 
  • a table or index scan with a lot of data behind it, waiting for a lock 
  • a server too busy servicing lots of queries and using all its CPU / memory / IO because of some inefficient queries and blocking / waiting


How to resolve this issue

Firstly ask your Database Administrator to check the nature of the activity on the SQL Server database to determine whether the issue relates to Data Volumes (cf. logging levels and database maintenance), Transaction Deadlocking or Database Server resource constraint


If the issue is caused by Data Volumes -

The 'Maintaining a Blue Prism Database Server' topics within the Blue Prism Product Online Help explain some of the recommended regular maintenance tasks that should be performed on the database.  If you require specific SQL scripts to clear out certain database tables, these can be supplied on request.  
Normally we ask that you send in a report showing a list of all the Database tables in your Blue Prism database showing the row counts and sizes of each table - this will enable us to spot which tables may be showing excessive size.

The following related articles explain some common issues and resolutions related to data volumes in a Blue Prism database:

We may also ask you to complete a Database Performance Questionnaire (attached to this article) if the clean-up recommendations do not resolve the issue. This questionnaire asks about Database Server specifications such as operating system, data file sizes, configuration and memory / CPU / disk usage, and also includes some very useful scripts that give us the longest running queries and their execution plans (e.g. how they get run).


If the issue is caused by Transaction Deadlocking -

Report the deadlocked query information by Customer Support as a new incident providing as much information as possible about:

  • The query involved in the deadlock
  • The activities being performed at the time of the deadlocking
  • The database server resource levels
  • The frequency of the issue. 

We will work with your Database Administrator to attempt to identify strategies for a resolution.


If the issue is caused by Database Server resource constraint (memory, CPU or disk activity) -

Work with your internal IT Department to determine whether the volumes of activity will require an increase in database server resources, or whether the Blue Prism activity needs to be amended to operate within existing resource constraints.


If the issue is not resolved by the above actions -

You may need to install Blue Prism v.6.3.0 or above. SQL timeout errors often got reported as a consequence of issues which were subsequently fixed in that



------------------------------
Simon Andre Pedersen
Manager
PWC HK
Asia/Hong_Kong
------------------------------

NicholasZejdlik
Level 9
I had a similar problem a couple years ago. Get Next Item was routinely timing out after 30 seconds with the same error you are getting. When I checked SQL profiler, I found that the Get Next Item procedure was taking tens of millions of reads to complete, which is very abnormal. I tracked the issue down to a bad query plan on the part of SQL's optimizer. The ultimate fix was to rebuild the index for the BPAWorkQueueItem table. After doing so, the query plan updated and things went back to normal.

------------------------------
Nicholas Zejdlik
RPA Developer
------------------------------