I have uploaded a csv file through the Splunk Web which has 157 columns and 4000 events.
I noticed that while directing to the index only 105 columns are visible with 4000 events.
Is there any limitation in splunk? And this is not the first time too I noticed this. Kindly help.
I seem to be experiencing the same issue as reported here with a very small/simple csv file that suggests to me that only csv columns which have *at least* one non-empty value in them are being imported. Could somebody please confirm whether this is expected behavior (hopefully not) or some sort of bug?.
CSV files should be uploaded as
lookup files, not
event data. I am not sure what happened when you indexed this file, but try using it as
lookup file instead and they try
|inputlookup YourLookupFileName.csv and see if it works better (it probably will).
Thanks @woodcock. But this do not give a clarity why the columns are truncated.
I ingest csv file from the AWS S3 into splunk always.
These feeds do not qualify to be a Lookup.
Please help to understand why the truncation of columns.
|inputlookup your_AWS_S3.csv |head 1 |transpose
index=your_index |head 1 |transpose
When compared, there may be characters that cannot be used in field names.