Can we relax the validation in ClusterRerouteService (not a bug)?

[jsjtzyy]

Please use the following questions as a guideline to help me answer
your issue/question without further inquiry. Thank you.

Which version of Elastic are you using?

[ *] elastic.v7 (for Elasticsearch 7.x)
[ ] elastic.v6 (for Elasticsearch 6.x)
[ ] elastic.v5 (for Elasticsearch 5.x)
[ ] elastic.v3 (for Elasticsearch 2.x)
[ ] elastic.v2 (for Elasticsearch 1.x)

Please describe the expected behavior

I was trying to call RetryFailed in ClusterRerouteService to retry on failed allocation in ES cluster. The code is quite simple (see below)

_, err := m.Client.ClusterReroute().RetryFailed(true).MasterTimeout(m.MasterTimeout).Do(ctx)

I expected it to succeed even though there was no request body or commands.

Please describe the actual behavior

Got missing allocate commands or raw body coming from here:

elastic/cluster_reroute.go

Line 195 in b7de00d

if err := s.Validate(); err != nil {

Any steps to reproduce the behavior?

I did this on Kibana and succeeded, so I wonder if we can relax the validation to accommodate simply retry_failed case?

POST _cluster/reroute?retry_failed

[olivere]

Hi! That does work, but I suspect this to be a problem with ES. It probably is a nop, i.e. it doesn't do anything in ES.

The documentation clearly states that the request body is required (see here). So I think it's good to warn users if they forgot to add a request body. Otherwise they think everything is OK while it really isn't (because ES just swallowed the error).