Inputs.mqtt_consumer stoped to work after update to Telegraf 1.33.0

Telegraf
1

I updated my system (new installation, updated influxdb, telegraf, grafana etc, restored backups). I got everything back to work again except one telegraf processor script.

Telegraf 1.33.0

[[inputs.mqtt_consumer]]
  interval="5m"
  flush_interval="5m"
  username = "myusername"
  password = "mypassword"
  servers = ["tcp://127.0.0.1:1883"]
  topics = ["collectd/+/+/+/+/#"]
  data_format = "csv"
  csv_header_row_count = 0
  csv_timestamp_column = "timestamp"
  csv_timestamp_format = "unix"
  csv_column_names = ["timestamp", "value"]
  csv_column_types = ["float", "string"]
  csv_delimiter = ":"
  [[inputs.mqtt_consumer.topic_parsing]]
    topic = "collectd/+/+/+/+"
    measurement = "measurement/_/_/_/_"
    tags = "_/site/device/plugin/field"
  [[processors.strings]]
    [[processors.strings.trim_right]]
      field = "value"
      cutset = "\u0000"
  [[processors.pivot]]
    tag_key = "field"
    value_key = "value"
  [[processors.converter]]
    [processors.converter.fields]
      unsigned = ["uptime", "stations", "bitrate", "signal_quality", "ping_droprate-*"]
      integer = ["signal_*"]
      float = ["if_*", "ping-*", "ping_stddev-*"]
  [inputs.mqtt_consumer.tags]
    destination = "Telegraf_collectd"

If I debug telegraf,

telegraf --config /etc/telegraf/telegraf.conf --config-directory /etc/telegraf/telegraf.d/ --test --test-wait 610

I can see these kind of logs entries:

> collectd,destination=Telegraf_collectd,device=mydevice,host=orangepizeroplus,plugin=iwinfo-phy0-sta0,site=mysite,topic=collectd/mydevice/mysite/iwinfo-phy0-sta0/stations stations=1u 1735340078333000000

This used to work, all other consumer scripts are working, so why is this one not working?

Did something changed in meanwhile on the Telegraf update?

I can’t see any other error messages (is there any verbose mode?)
and I stopped to see this specific data on influxdb dataexplorer (I can see other consumers updating data)