Repository issues

Application Name: Advanced ETL Processor ENT
Version Number: 6.4.7.1
Repository Type: SQL
Database Type: MS SQL Server 2022
Log File: Not sure if this applies in this situation
Problem Description: Trying to add any object to a current project it taking upwards of multiple minutes, up to 5+ 6 minutes as some time, as well as 100% CPU spikes every 6 - 7 seconds
Steps to Reproduce the Problem:

• Adding any object to a project take multiple minutes
• It does not matter the object; SQL scripts, Packages etc
• Constant CPU spikes to 100%, this seems to be linked to orphaned items in our Queue table(See screen Shot) that the ETL Execution agent keeps trying to execute every 6- 7 seconds, which I am not sure those items became stuck in the Queue table

NOTE: I was able to confirm the CPU spikes were caused by the stuck queue items, by removing them from the queue via SSMS on restored machine in a test environment, but I am not sure that is the correct solution for that issue
 

The pictures quality is too bad so I cannot see anything

Have you set Language to English as described here?

www.etl-tools.com/wiki/knowledgebase/rep...l-server-repository/

1 Agent checks SCHEDULE table and if there's anything to execute it inserts record into QUEUE table than updates next execution time in SCHEDULE table
2 Than it loops over QUEUE table and executes tasks one by one.
3 When task is completed record is moved from QUEUE table to QUEUE_HISTORY

# 1 may fail because of incorrect language.

Why everything is slow

If QUEUE table has millions of records it will take a lot time to loop over = CPU spikes.

Trouble shooting

1 Stop the agent
2 Check the language
3 Delete all records from QUEUE
4 Start the agent

If you recently upgraded from a very old version you may need to patch the repository
Also Check the agent log located in the same directory
Running SQL Server monitor against repository may help to understand what is going on
Also check if if you have a lot of AETLCL.EXE processes

Keep us posted on your progress

 

Apologies I didn't realize the scaling had changed on my screen shots

Currently the language is set to English- us_english for the repository account 




 

I followed the steps and stopped the ETL agent service and directory monitor services, removed all the records stuck in the QUEUE table and restarted the services, which has resolved the looping CPU spikes

I tested adding another test object to a project and still getting a not responding screen for about 3ish minutes



 



The same result happens when deleting an object as well

Please email us your repository backup (SQL server backup)

 

Would that be the db backup itself, or just the .zip file that ETL creates ?

Is there an official support email we can send an encrypted link to share the backup with as it has sensitive information contained in it ?