Due to the failure of a switch (T267865) 3 elasticsearch nodes are currently unavailable (elastic2048,elastic2049,elastic2059). While most shards have moved to other nodes, some enwiki_content shards are still unallocated (see below). This has some impact on the throughput and the resilience of that index.
Failed allocation might be transient and we might be able to recover with the related cookbook. Or depending on how long we expect those nodes to be down, we might need to review the sharding of enwiki.
gehel@elastic2040:~$ curl -s localhost:9200/_cat/shards | grep -v STARTED enwiki_content_1605226908 7 r UNASSIGNED enwiki_content_1605226908 13 r UNASSIGNED enwiki_content_1605226908 6 r UNASSIGNED enwiki_content_1605226908 12 r UNASSIGNED enwiki_content_1603009218 8 r UNASSIGNED enwiki_content_1603009218 6 r UNASSIGNED enwiki_content_1603009218 2 r UNASSIGNED enwiki_content_1603009218 1 r UNASSIGNED enwiki_content_1594735844 11 r UNASSIGNED enwiki_content_1594735844 9 r UNASSIGNED enwiki_content_1594735844 15 r UNASSIGNED enwiki_content_1594735844 2 r UNASSIGNED