Hadoop as target

[Pranita123]

Hello Community,

We're having trouble connecting to HDFS(Hadoop manage endpoint) in Linux.

The error message says: "Failed to connect to HDFS path '/'" and "Failed to curl_easy_perform: curl status = 56, Failure when receiving data from the peer (CONNECT tunnel failed, response 503)."

In addition, initially, we were able to successfully establish a connection with the target and successfully load data into it.

Your assistance in this situation is greatly valued.

Regards,

Pranita

[john_wang]

Hello @Pranita123 ,

Thanks for reaching out to Qlik Community!

There are many reason may lead the same error, includes:

• Hive ODBC version
  - You might disable the Hive ODBC Access to know if the error caused by it
• Timeout issue
  - Set internal parameter executetimeout in target endpoint and increasing it to higher value eg 6000
• if the account have enough permission to access HDFS folder "/"
  - in general the data will be written to a sub-folder under "/"
• We ever experienced such error caused by a bug on the Cloudera side
  - Did you ever upgrade Hadoop and then got the error
• Set below components logging level to Trace or Verbose to get great detailed information from task log file
  - COMMUNICATION/FILE_FACTORY/TARGET_LOAD/TARGET_APPLY

Hope this helps.

John.