Sunday 28 July 2013

Database Error: Failed to connect to database using user [ ]and connection string [Postgres Liferay Source].].

Database Error: Failed to connect to database using user and connection string [Postgres Liferay Source].].

solution:

Below parameters were changed in odbc.ini file and the jobs succeeded subsequently.
 
EnableDescribeParam=0(was pointing to 1 earlier)

EncryptionMethod=1(was pointing to 0 earlier)

Primary checks before doing the above thing.


1. check are you able to connect to that data base,if you are able to connect to that data base but still its giveing error then you need to check the above parameter settings.

2.if you are not able to connect to that data base then there was a problem at data base side.

3.There may be a case that we need to check the ConnectionRetryCount in ODBC driver,this will occur because of the network latency so that the client is not able to sent the credentials with in the 5 seconds.

Try changing the ODBC parameter“ConnectionRetryCount” from 0 to 3.

postgres[20199]: [4-1] ERROR: could not send data to client: Broken pipe error in informatica/ERROR: could not send data to client: Broken pipe

ERROR:postgres[20199]: [4-1] ERROR: could not send data to client: Broken pipe


Solution:

we found that There was trace enable @ the ODBC level which was eating up resource causing process to choke and fail. All we did is disabling the trace for particular DSN.

if the Above one is not work check for the following:


0.Try to run the batch load(load the source data into batches) ,if batch load successful then most of the cases there is no data issue otherwise it is a data issue.


1.There may be a case that in source metadata changed i.e columns data types and sizes,so we need to check that source table once.

if it is the case we need to re import the source by deleting the already existing source and refresh the mapping and session.
2.if there is no issue with the source metadata ,then we need to check data discrepancy

i.e we need to check our source data for unsupported data mean that if there is any special characters are coming from source data in any column .

if it is the case we need to find that and according to that character we need to change the code page compatibility in informatica and try to load with the changes.

3.if the source/target table contains the CLOB data type try to change that and load once.


4.There may be case that there is a issue with DTM buffer size/commit interval then we need
to increase/decrease this according to the data.

5. if all the above is not work try to create a new mapping with the same business logic if it will not work contact to the informatica admin people they will do the informatica server level changes.