Graylog discards messages when it detects that there's no Elasticsearch master elected

[asachs01]

What?

This is from HS-373864503

if the cluster is in cluster block for no master Graylog discards all messages with should be indexed.

Following gets logged to Indexer Failures Page:
{"type":"cluster_block_exception","reason":"blocked by: [SERVICE_UNAVAILABLE/2/no master];"}

In my opinion this is a retryable state, which should be retried, same as a Cluster Block for Full Filesystem (which gets retried).

---

Edit

Please note that the customer indicated that they'd also like us to consider this message as a retryable state;

' {"type":"unavailable_shards_exception","reason":"[log_30d_4656][0] primary shard is not active Timeout: [1m], request: [BulkShardRequest [[log_30d_4656][0]] containing [458] requests]"} '