Oracle Analytics Cloud and Server

Home› Oracle Analytics› Oracle Analytics Forums› Oracle Analytics Cloud and Server› Oracle Analytics Cloud and Server

Categories

Welcome to the Oracle Analytics Community: Please complete your User Profile and upload your Profile Picture

S_NQ_JOB_PARAM

Received Response
101
Views
4
Comments
Nayas
Nayas Rank 2 - Community Beginner
edited Aug 13, 2024 7:24PM in Oracle Analytics Cloud and Server

Hi Everyone,

We have upgrade one of our environment, from 11.1.1.6 to 11.1.1.9.

That environment had lots of try and erorr, catalog upgrade (from various source).

Right now, the s_nq_job_param has duplicate records for the same Agent.

We were thinking about disabling every agents, cleaning the s_nq_job_param table and reactivating the ones that we need.

We cannot find any information / procedure to follow to clean / rebuild the s_nq_job_param table.

We can easily take a backup of the table, truncate it and enable the agents, but we cannot find if that's a good / supported way of doing it.

Does anyone would know?

Thanks!

Answers

  • Gianni Ceresa
    Gianni Ceresa Mod 
    Nayas wrote:Right now, the s_nq_job_param has duplicate records for the same Agent.

    So all the rows are in double for the same exact JOB_ID ? Or you have 2 different JOB_ID which reference the same agent?

  • Nayas
    Nayas Rank 2 - Community Beginner

    Hi Gianni,

    We have 2 different job_id referencing the same Agent.

    So when a scheduled Agent run, it errors out because the job_id in the report in the Catalog as the latest job_id.

    in other words, say we have job_id 128 and 300 referencing the same report.

    When job_id 128 runs, it errors out, because the job_id in the report in the catalog has 300 :

    Scheduler job id 128 points to ibot /users/ibot_creator/Agents/10xxxx/100xxx/1008xx/Report_CSV with job id 300

    Also, I have a doubt about s_nq_job table.

    I can also see duplicate entries, but do I understand correctly, that this is normal, to keep track of historic scheduling?

    Only the lastest (greater job_id in s_nq_job) is the current scheduling definition?

    Thanks!

  • Gianni Ceresa
    Gianni Ceresa Mod

    It's a bit more weird than that

    S_NQ_JOB is the one having JOB_ID as PK, while in S_NQ_JOB_PARAM it's a FK.

    For 1 given agent you are supposed to have a single JOB_ID, both in S_NQ_JOB and S_NQ_JOB_PARAM.

    If you really want to restart "clean" you can truncate both tables (and maybe the ERROR one too as errors without the corresponding IDs will make no sense) and then if you open and save the agents again they will create new entries in the tables.

    Or you can just delete the rows pointing at JOB_ID which aren't the ones of the agent (saved inside of the XML of the agent itself, in the first XML tag, the <saw:ibot, as attribute).

    AFAIK there isn't a doc covering how to managed by hand those tables as you are never supposed to put your hands there.

    If you just truncate all the tables related to jobs, open the agents and save them again you will start with a clean DB again.

  • Nayas
    Nayas Rank 2 - Community Beginner

    Hi Gianni,

    Ya, I opened an SR with Oracle for this, but no answer so far.

    I though asking here might be faster

    Thank you for the information.