Bonjour ![]()
`grafana/alloy:v1.12.1 (docker)`
`grafana/loki:3.6.3 (docker)`
Fail2ban install on VPS.
I have this error log always, always, every times:
{
"ts":"2026-01-05T10:25:30.375026104Z",
"level":"error",
"msg":"final error sending batch,
no retries left,
dropping data",
"component_path":"/",
"component_id":"loki.write.grafana_loki",
"component":"client",
"host":"loki:3100",
"status":"400",
"tenant":"",
"error":"server returned HTTP status 400 Bad Request (400): entry for stream '{component=\\\"actions\\\",
exporter=\\\"OTLP\\\",
geoip_city_name=\\\"Ashgabat\\\",
geoip_continent_code=\\\"AS\\\",
geoip_continent_name=\\\"Asia\\\",
geoip_country_code=\\\"TM\\\",
geoip_country_name=\\\"Turkmenistan\\\",
geoip_location_latitude=\\\"37.95\\\",
geoip_location_longitude=\\\"58.3833\\\",
geoip_subdivision_code=\\\"S\\\",
geoip_subdivision_name=\\\"Ashgabat\\\",
geoip_timezone=\\\"Asia/Ashgabat\\\",
instance=\\\"domaine.tld\\\",
job=\\\"fail2ban\\\",
priority=\\\"NOTICE\\\",
service_name=\\\"actions\\\"}' has timestamp too new: 2026-01-05T11:25:22Z"
}
has timestamp too new?
I see in logs:
βtsβ:β2026-01-05T10:25:30.375026104Zβ
and
2026-01-05T11:25:22Z
OK, one hour time difference but how to change?
My host is on timezone Europe/Paris, and docker is on UTC date.
I found in Grafana Loki configuration parameters | Grafana Loki documentation
`creation_grace_period`
But is it the recommanded way to avoid this time problem?
Thanks