Attempted to create too many ROS containers for projection VDWTCUDB.pwr_st_evt_b1

HINT:  Likely due to partitioning.


Hi All,


Hope you are doing well we are getting the below error whenever we load data one after the other , and the table is partition on the basis of month and has close to 20 years of data . We had several discussion with Vertica DBA and they want us to remove the old data which is not required , we can do that but I had a doubt is there any way where we can support huge data of several years


Note : When we do MO or Purging then our jobs succeed but after few days we again face the issue once the load on that table is done in a bulk.



Amit Terse


  • Options

    Hi, 20 years with montly partition is 240 partition what it should be just fine for Vertica. the conf setting is at 1024.


    What it may be happening here is the way that you load data. Things that you can consider are :


    1- Are you doing Bulk load ( COPY  or Inserts)

    2- Are your loads DIRECT ( bypassing wos ) if so stop doing it. Do not use direct in your statements.

    3- How many active partitions do you have ? select get_config_parameter('ActivePartitionCount'); it should be 1 if it is not find the reason why it is grather.

    4- Are you loading to aggresively in different tables and mergeout is not catching up with this table?


    So, first we need to find the reason why you reach the 1024 so often.


    Hope this helps.


Leave a Comment

BoldItalicStrikethroughOrdered listUnordered list
Align leftAlign centerAlign rightToggle HTML viewToggle full pageToggle lights
Drop image/file