Elasticsearch batch not indexing all items using PHP client - php

Question

I've been trying to build an index in ES and add the initial items to it (around 350k), using PHP.
I tried all kinds of batch sizes (from 10 items to 1k), check the count, check the threshold, but for some reason it doesn't index every item.
It just skips over some random items, without any errors in the batch result response. I feel like I tried everything and I have to idea what to do next
I'm using Amazon OpenSearch with the latest supported ES (7.10).
The index looks like this:
{
"wonder-search": {
"aliases": {},
"mappings": {
"properties": {
"address": {
"type": "text"
},
"city": {
"type": "text"
},
"city_id": {
"type": "integer"
},
"duration": {
"type": "integer"
},
"filename": {
"type": "text"
},
"geo_point": {
"type": "geo_point"
},
"icon": {
"type": "keyword"
},
"is_sandbox": {
"type": "integer"
},
"item_id": {
"type": "integer"
},
"item_label": {
"type": "keyword"
},
"latitude": {
"type": "float"
},
"longitude": {
"type": "float"
},
"search_text_caption_json": {
"type": "text",
"index_phrases": true
},
"search_text_city_json": {
"type": "text",
"index_phrases": true
},
"search_text_completion": {
"type": "completion",
"analyzer": "simple",
"preserve_separators": true,
"preserve_position_increments": true,
"max_input_length": 50,
"contexts": [
{
"name": "type",
"type": "CATEGORY"
}
]
},
"search_text_country_json": {
"type": "text",
"index_phrases": true
},
"search_text_cuisine_name_json": {
"type": "text",
"index_phrases": true
},
"search_text_location_name_json": {
"type": "text",
"index_phrases": true
},
"search_text_state_json": {
"type": "text",
"index_phrases": true
},
"search_text_tag_name_json": {
"type": "text",
"index_phrases": true
},
"search_text_username_json": {
"type": "text",
"index_phrases": true
},
"sort": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"sort_score": {
"type": "double"
},
"type": {
"type": "text"
},
"user_icon": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"user_id": {
"type": "integer"
},
"username": {
"type": "keyword"
},
"vanity_url": {
"type": "keyword"
},
"video_count": {
"type": "integer"
}
}
},
"settings": {
"index": {
"routing": {
"allocation": {
"include": {
"_tier_preference": "data_content"
}
}
},
"mapping": {
"ignore_malformed": "true"
},
"number_of_shards": "1",
"provided_name": "wonder-search",
"creation_date": "1671003076106",
"number_of_replicas": "1",
"uuid": "YQh1q40WTneLE4MWDWhArw",
"version": {
"created": "7100199"
}
}
}
}
}
and one item looks like this:
{
"_index": "wonder-search",
"_type": "_doc",
"_id": "wq2LD4UBUAuy7FQPhtZh",
"_version": 1,
"_seq_no": 2003,
"_primary_term": 1,
"found": true,
"_source": {
"sort": "4004",
"item_id": "4934",
"user_id": "434",
"user_icon": "/site-content/avatars/Sp8AXjTJvMbRao2oZbuiUuSVH042-1597776099045.jpeg",
"username": "chuurros",
"item_label": "Kyoto Katsugyu【京都勝牛】",
"search_text_username_json": [
"chuurros"
],
"search_text_caption_json": [
"Absolutely love their gyukatsu (beef katsu) here! Delicious and will keep you wanting more! 🥰"
],
"search_text_city_json": [
"Toronto"
],
"search_text_state_json": [
"Ontario"
],
"search_text_country_json": [
"Canada"
],
"search_text_location_name_json": [
"Kyoto Katsugyu【京都勝牛】"
],
"search_text_tag_name_json": [
"japanese",
"restaurant",
"asian",
"dining",
"topcollection-4934"
],
"search_text_cuisine_name_json": [],
"type": "video",
"vanity_url": "",
"icon": "",
"city": "Toronto",
"city_id": "439",
"latitude": "43.65682410",
"longitude": "-79.37617410",
"address": "134 Dundas St E",
"duration": "9.57",
"video_count": "0",
"sort_score": "43",
"filename": "373d75fd-4292-4e5b-a239-4b1c39ffc86c.MOV",
"is_sandbox": "0",
"geo_point": {
"lat": "43.65682410",
"lon": "-79.37617410"
},
"search_text_completion": {
"input": [
"Kyoto Katsugyu【京都勝牛】"
],
"contexts": {
"type": [
"video"
]
}
}
}
}
Any ideas why does it work like this?