Indexer queue sizes

[jpillai]

Hi all,

We have been facing some errors with Splunk indexers, where it says something like below.

```

Failed processing http input, token name=<HECtoken>, channel=n/a, source_IP=, reply=9, events_processed=62, http_input_body_size=47326, parsing_err="Server is busy"

```

And I found in some discussions that increasing queue sizes may help sometimes. We are indexing ~400GB per day and it makes sense to increase the queue sizes as default values might not be good enough in this case.

However, the splunk docs doesnt have a detailed explanation of which queues can be set in server.conf and what are the proportions that we need consider. Can someone help with understanding this?

[richgalloway]

AFAIK, indexer queues are not configurable.  You can, however, use maxQueueSize in outputs.conf on the forwarders to set the size of the output queue.  That's the queue where packets are stored if the destination becomes unavailable.

In the case of HEC inputs, it's the responsibility of the client to retry any request that gets a non-200 (OK) response code ("server is busy" in this case).