Increasing cluster event (put-mapping) timeout from 30s


(Bobby Impollonia) #1

While bulk-loading data into a new ES index, I am hitting the "failed to process cluster event (put-mapping) within 30s" error on the first chunk of data. I have read the previous threads on this topic and understand what the error means and possible workarounds: create the mapping explicitly before loading any data, use less shards, etc.

My question (which I could not find answered in the previous threads) is whether it is possible to increase the amount of time allowed for put-mapping from 30s to a larger number. I tried passing a timeout value for my request and while this does increase the amount of time allowed for the actual inserting of the documents, the implicit put-mapping operation is still failed after 30s.


(Christian Dahlqvist) #2

I would recommend looking into why cluster updates are taking so long. Increasing the timeout (not sure how this can be done) may only give you temporary relief and potentially cause a lot more problems down the line.