buffer.type field being rejected

[Eubo0]

Question

I am using vector 0.48.0. I am trying to enable disk buffering for the elasticsearch sink, but when I specify a type for the buffer it gets rejected. Curiously it also rejects me specifying the buffer type as "memory". The docs website shows it as an option. Is this option supported for the elasticsearch sink or am I doing something wrong?

Vector Config

[sources.http_logs]
type = "http_server"
address = ""
encoding = "ndjson"

[transforms.set_index]
type = "remap"
inputs = ["http_logs"]
source = '''
app = "orphan"

ts = parse_timestamp(.timestamp, "%+") ?? now()

year = format_timestamp(ts, "%Y") ?? "year"
month = format_timestamp(ts, "%m") ?? "month"

."@timestamp" = ts
.index = "app-" + app + "-" + year + "-" + month
'''

============== Sinks ==============

[sinks.to_elasticsearch]
type = "elasticsearch"
inputs = ["set_index"]
endpoints = [
""
]
buffer.type = "disk"
buffer.max_size = "200MB"
buffer.when_full = "drop_newest"

[sinks.to_elasticsearch.auth]
strategy = "basic"
user = ""
password = ""

[sinks.to_elasticsearch.tls]
verify_certificate = false

[sinks.to_elasticsearch.bulk]
action = "create"
index = "{{ index }}"

Vector Logs

2025-07-30T15:35:17.523108Z ERROR vector::cli: Configuration error. error=data did not match any variant of untagged enum BufferConfig in sinks.to_elasticsearch.buffer

[Eubo0]

I was being a dummy, I simply needed to provide the "max_size" field as a number and not a string. All is well