Broad Log Data Retrieval Issue | Community
Skip to main content
adithyacs86
adithyacs86
Level 5
December 21, 2020
Solved

Broad Log Data Retrieval Issue

  • December 21, 2020
  • 4 replies
  • 3019 views

Hi Team, 

 

When we try to run any query against broadlog RCP table, we always see that the data retrieval is very slow, for example :- 

 

above is the only query to read the data from the broadlogRCP table, which takes long hours to get the data and sometimes even times out. 

 

We also have checked if there are any parallel jobs running on the same table or any other through the query monitoring, but did not see any except this workflow. 

Since it is Adobe Hosted environment we are not able to see the logs properly. 

 

I was just wondering should this table be optimized or analysed before we start the run the query activity. 

 

Thanks,
Adithya

This post is no longer active and is closed to new replies. Need help? Start a new post to ask your question.
Best answer by DavidKangni

Hi Aditya,

 

I will just query on channel equal to Email and status equal Sent.

Reason:

  • primary key can not be null
  • if email address contains , it will return a bounce with status failed

If your query is a frequent query and is taking long add db index on channel and status fields in the broadlog rcp schema.

<dbindex name="status"> <keyfield xpath="@status"/> </dbindex>

Thanks,

David

 

4 replies

_Manoj_Kumar_
Community Advisor
_Manoj_Kumar_Community Advisor
Community Advisor
December 21, 2020

Hello @adithyacs86 ,

 

There are two conditions which I think are not required.

 

channel fo delivery equal to Email

and status equal to sent

 

Use these two conditions only and you will have the same result and will surely take less time to execute.

     Manoj     Find me on LinkedIn
adithyacs86
adithyacs86Author
Level 5
December 21, 2020

@_manoj_kumar_ , 

 

I tried the way you mentioned but the result is still the same, the workflow is still running now. 

 

_Manoj_Kumar_
Community Advisor
_Manoj_Kumar_Community Advisor
Community Advisor
December 21, 2020
Yes, it will take time because you are querying a lot of data. but it shouldn't time out.
     Manoj     Find me on LinkedIn
DavidKangni
Community Advisor
DavidKangniCommunity AdvisorAccepted solution
Community Advisor
December 21, 2020

Hi Aditya,

 

I will just query on channel equal to Email and status equal Sent.

Reason:

  • primary key can not be null
  • if email address contains , it will return a bounce with status failed

If your query is a frequent query and is taking long add db index on channel and status fields in the broadlog rcp schema.

<dbindex name="status"> <keyfield xpath="@status"/> </dbindex>

Thanks,

David

 

David Kangni
L
Llamour
February 5, 2021

Hi Adita,
Could you use text to describe your query instead of a screenshot?
I see 2 kind of queries and one is not relevant:
1- Primary key is not empty and Delivery Channel is Email and Address does not contain ',' and status is sent
2- Primary key is not empty and Delivery Channel is Email and Address does not contain '.' and status is sent

Only second is correct but has poor performances:
1- you'll never find a sent email with a ',' in its address as this is a forbidden caracter in email address... So you don't need to use a "does not contain" operator that is really consuming DB ressources on a such table
2- if the character to search is a ".", then the result is 0 as you'll always have this caracter in any email address

Anyway, broadlogRcp table is usually the table with the greater amounts of records in your database. Query it carefully and if performance are not relevant according to your needs, you should perhaps reduce its Purge retention delay in the deployment wizzard

Terms & ConditionsAccessibility statement

Loading footer...