TDLOAD failing with 2589 Request block is not found error

Tools & Utilities
Enthusiast

TDLOAD failing with 2589 Request block is not found error

Hi Team, this issue is really urgent.

I have loaded a table with 4 columns and one identity column. Almost, every row is duplicated 4 times and table size is around 33 gb. There is another copy of this table having 19GB data in which only duplicate data is available.

 

Now I am running both export with same settings such as SourceInstances=4, SourceMaxSessions- 24 and sometimes no mention of MaxSessions. 

 

tdload -j variable.txt job_name.

 

Smaller table export was successful while bigger table failing again and again on same data size. If it is a restart, it hangs for some time and then failed. Really need urgent help.

 

I am using AWS teradata developer instance having 6 AMPs.

21 REPLIES
Teradata Employee

Re: TDLOAD failing with 2589 Request block is not found error

Can you please provide the contents of the job variable file?

(You can leave out sensitive information like passwords)

 

-- SteveF
Teradata Employee

Re: TDLOAD failing with 2589 Request block is not found error

The definition of a DBS 2589:

2589 Request block is not found.

Explanation: The file system does not find the request block in the spool file. Either we have a software bug or the spool

file has been deleted by the host (i.e., an End Request has been issued by the host).

Generated By: LCTExpSt

For Whom: End user

Remedy: If the spool file was not explicitly deleted by the host, then contact your Support Representative.

 

You may want to open an incident for this and have the database people look at it.

 

-- SteveF
Enthusiast

Re: TDLOAD failing with 2589 Request block is not found error

Hi,

 

Below solution I have looked at but there is no such file deletion. Also , I am running it on AWS Free version, hence opening an incident is not an option here.

 

 

Variable file content:

 

SourceTdpId='XX.XX.**bleep**.**bleep**',
SourceUserName='abc',
SourceUserPassword='XXXXXXX',
SourceTable = 'db_name.table_name',
SourceInstances = 4,
SourceMaxSessions =24,
TargetFormat = 'delimited',
TargetTextDelimiter = ',',
TargetQuotedData = 'YES',
TargetFilename = 'db_name.table_name.2018-03-15.csv'
Enthusiast

Re: TDLOAD failing with 2589 Request block is not found error

I tried to run with lower instances and lower maxsessions also. Then job taking much time and getting failed on lesser file size.

Teradata Employee

Re: TDLOAD failing with 2589 Request block is not found error

I spoke with the DBS developer. First, they would like you to open an incident and provide all of the details.

Next, if you could get this error to occur with as little data as possible, that would be great.

They have asked for this information:

  

Error 2589 is returned if the block number requested by FastExport client does not exist in the spool file. I suggest getting the following information:

  1. Number of blocks returned by DBS at the end of the SELECT statement.
  2. The block number of the request which got the error.
  3. The export block size.

 

In order to get #2, I would need trace information from the TPT job.

#1 and #3 would be available in the normal log.

The trace information on long running jobs will be huge, thus the smaller amount of data that can still result in this error, the better.

 

I need you to set the TraceLevel attribute for Export operator (in the job variable file, set ExportTraceLevel='all'), run thie job, and then send me the binary log (the .out file). The name and location of that file will be output on the console.

 

 

 

 

-- SteveF
Enthusiast

Re: TDLOAD failing with 2589 Request block is not found error

>>>>>>>>>>.Block>>>>>>>>>>>>>

Using memory mapped file for IPC

TPT_INFRA: TPT04101: Warning: Teradata PT cannot connect to Unity EcoSystem Manager.
The job will continue without event messages being sent to Unity EcoSystem Manager.
TPT_INFRA: TPT04197: Warning: OMD API failed to initialize
Teradata Parallel Transporter PlanExecutor Version 16.20.00.00

Found CheckPoint file: /opt/teradata/client/16.20/tbuild/checkpoint/load_inven_jobLVCP
This is a restart job; it restarts at step MAIN_STEP.
Total Available Memory: 19926880
DSMemoryPercentage: 70
Using Message Size: 1048576

Teradata Parallel Transporter Executor Version 16.20.00.00
Teradata Parallel Transporter Executor Version 16.20.00.00
Teradata Parallel Transporter Coordinator Version 16.20.00.00
Teradata Parallel Transporter Executor Version 16.20.00.00
Teradata Parallel Transporter Executor Version 16.20.00.00
Teradata Parallel Transporter Executor Version 16.20.00.00
Teradata Parallel Transporter DataConnector Operator Version 16.20.00.00
$FILE_WRITER[1]: Instance 1 directing private log report to 'FileWriterLog-1'.
$FILE_WRITER[1]: TPT19011 Instance 1 restarting.
$FILE_WRITER[1]: DataConnector Consumer operator Instances: 1
$FILE_WRITER[1]: TPT19206 Attribute 'OpenMode' value reset to 'WRITEAPPEND'.
$FILE_WRITER[1]: TPT19203 Required attribute 'OpenQuoteMark' not found. Defaulting to '"'.
$FILE_WRITER[1]: TPT19203 Required attribute 'CloseQuoteMark' not found. Defaulting to '"'.
Teradata Parallel Transporter Export Operator Version 16.20.00.00
$EXPORT: private log specified: ExportLog-1
$FILE_WRITER[1]: ECI operator ID: '$FILE_WRITER-21514'
$EXPORT: connecting sessions
$FILE_WRITER supports the max row size 64KB.
$EXPORT supports the max row size 1024KB.
Job step 'MAIN_STEP' uses the max row size 64KB.
Job is running in Buffer Mode
Task(SELECT_2[0002]): restart completed, status = Success
Task(SELECT_2[0001]): restart completed, status = Success
Task(SELECT_2[0003]): restart completed, status = Success
Task(SELECT_2[0004]): restart completed, status = Success
$FILE_WRITER[1]: Operator instance 1 processing file '/perfTest/inventory.18-03-17.csv'.
Task(INSERT_1[0001]): restart completed, status = Success
Task(SELECT_2[0002]): checkpoint completed, status = Success
Task(SELECT_2[0003]): checkpoint completed, status = Success
Task(SELECT_2[0001]): checkpoint completed, status = Success
Task(SELECT_2[0004]): checkpoint completed, status = Success
Task(INSERT_1[0001]): checkpoint completed, status = Success
$EXPORT: sending SELECT request
Data Block size: 1048584 Buffers/Block: 2 Data Buffer size: 524288

>>>>>>>>>>>>>>>>>Export Log>>>>>>>>>>>>>>>>>>>>>>>>..
Teradata Load Utility Version 16.20.00.00 64-Bit
Teradata Parallel Transporter Version 16.20.00.00 64-Bit
Job log: /opt/teradata/client/16.20/tbuild/logs/load_inven_job-1885.out
Job id is load_inven_job-1885, running on gcp-linux
Teradata Parallel Transporter DataConnector Operator Version 16.20.00.00
$FILE_WRITER[1]: Instance 1 directing private log report to 'FileWriterLog-1'.
$FILE_WRITER[1]: DataConnector Consumer operator Instances: 1
$FILE_WRITER[1]: ECI operator ID: '$FILE_WRITER-5333'
Teradata Parallel Transporter Export Operator Version 16.20.00.00
$EXPORT: private log specified: ExportLog-1
$FILE_WRITER[1]: Operator instance 1 processing file '/perfTest/

@feinholzwrote:

I spoke with the DBS developer. First, they would like you to open an incident and provide all of the details.

Next, if you could get this error to occur with as little data as possible, that would be great.

They have asked for this information:

  

Error 2589 is returned if the block number requested by FastExport client does not exist in the spool file. I suggest getting the following information:

  1. Number of blocks returned by DBS at the end of the SELECT statement.
  2. The block number of the request which got the error.
  3. The export block size.

 

In order to get #2, I would need trace information from the TPT job.

#1 and #3 would be available in the normal log.

The trace information on long running jobs will be huge, thus the smaller amount of data that can still result in this error, the better.

 

I need you to set the TraceLevel attribute for Export operator (in the job variable file, set ExportTraceLevel='all'), run thie job, and then send me the binary log (the .out file). The name and location of that file will be output on the console.

 

 

 

 



inventory.18-03-17.csv'.
$EXPORT: connecting sessions
$EXPORT: sending SELECT request
$EXPORT: TPT10508: RDBMS error 2595: The FastExport select request has been aborted.
$EXPORT: disconnecting sessions
$EXPORT: Total processor time used = '2.26005 Second(s)'
$EXPORT: Start : Sun Mar 18 11:46:27 2018
$EXPORT: End : Sun Mar 18 12:43:24 2018
Job step MAIN_STEP terminated (status 12)
Job load_inven_job terminated (status 12)
$FILE_WRITER[1]: Total files processed: 0.
Job start: Sun Mar 18 11:46:21 2018
Job end: Sun Mar 18 12:43:28 2018
You have new mail in /var/spool/mail/root

Teradata Employee

Re: TDLOAD failing with 2589 Request block is not found error

In order to assist you, I would need you to email me the binary log file. The .out file.

As I indicated before, the console output shows the location and name of the file.

 

-- SteveF
Enthusiast

Re: TDLOAD failing with 2589 Request block is not found error

Hi,

 

Since file was very big and I was unable to download the file, hence got delayed.

 

please find file at below location. 

 

https://drive.google.com/drive/folders/1oy4zvQcTUF_vYClvtwLEuG3WSv4ebWGx

 

Enthusiast

Re: TDLOAD failing with 2589 Request block is not found error

Can you please let me know if file has enough info from file