cancel
Showing results for 
Search instead for 
Did you mean: 

Data Classification suddenly ceases

Washout
Level 0

Hi, just wondering if anyone has seen something similar or could point us in the right direction.

We've recently installed a classification server (win 2016). When we issue classification tasks from the workspace the server goes off and fetches the files and starts to process them from our NetApp filers.

However for larger shares eventually the classification will simply stop. CPU usage will drop off to nothing. The server will continue to fetch files until the safeguard we have in place kicks in and that stops too.

It doesn't seem to start up again, or if it does it's unuseably slow. We've got the latest version (6.1.2.9905) installed across the management server, collectors and the classification server.

Any pointers would be appreciated!

3 REPLIES 3

Rod_p1
Level 6
Employee Accredited Certified

It would appear that you have been disconnected between the management server and classification server(s). You do have the newest version which was released recently. It is new enough in the field we have not seen the issue you describe and would need to review logs, system events, configuration, policies enabled, etc.

If you have a current support contract I would suggest that you open a support case to have a dedicated technician review your environment and assist you with any performance tuning to avoid bottlenecks. If you are on your own I suggest you check the \vic0.0.log for the failure to communicate such as:

SEVERE:  #{2370} [ClassifyJob$ClassifierThread.classify] Failed to classify the following file: *********

Caused by: javax.ws.rs.InternalServerErrorException: HTTP 500 Internal Server Error

The vic.log can also hold some clues for interpretation:

 ERROR: #{dw-33203 - POST /vic/api/v1/classification/classify} Internal Server Error (500): Internal Server Error
org.apache.commons.lang3.exception.ContextedRuntimeException: Classify failed
Exception Context:
    [1:tenantId=_Default]
    [2:documentName=D:\DataInsight\data\classification\content\1\475\*********]
    [3:policyIds=[]]
    [4:policy=null]

Caused by: com.veritas.vic.extractor.ExtractorException: null
Caused by: java.io.IOException:
Caused by: org.apache.tika.exception.TikaException: TIKA-198: Illegal IOException from org.apache.tika.parser.microsoft.OfficeParser@54e6ab22
Caused by: org.apache.poi.poifs.filesystem.NotOLE2FileException: Invalid header signature; read 0x00670073004D003C, expected 0xE11AB1A1E011CFD0 - Your file appears not to be a valid OLE2 document

or

Caused by: org.apache.tika.exception.TikaException: Table overflow

Under full debug we would see the connection on the classification server:

Host: localhost:8989
Connection: Keep-Alive

you should be able to confirm that with the netstat command.

You can check on the ClassifyReqState in the commd0.0.log to verify the operation at the time of the slowdown.

Note: these are examples and may not be related to your issue

There are other files and databases an experienced support technician will request you remit for review to guide you towards increasing performance in your particular environment but I would not suggest attaching them here in a public forum.

Hopefully this is of some use in guiding you to your next steps.

Rod

Thanks Rod , I have been assigned to this case now.

Will looking for resolution.

 

Regards

Prakhar

Hi, 

I am getting the same issue, any hint should be really appreciate, what should be causing the issue?