Zabbix Expert Needed to Fix Incorrect Data Across All Historical Graphs

We are looking for an experienced Zabbix specialist to investigate and permanently resolve an issue affecting all graphs in our production Zabbix environment. Environment: Zabbix 7.0 LTS PostgreSQL 16 with TimescaleDB Approximately 15,000 monitored hosts History retention: 45 days Issue: All Zabbix graphs are displaying incorrect historical downtime or 0 values, even though the corresponding item history shows continuous successful values with no recorded downtime. For example, an ICMP availability graph shows downtime within the 30-day period, but the item history for the same period contains only successful values of 1 and no 0 values. We have already reviewed and adjusted the housekeeping, history retention and trend retention settings, but the issue remains. Scope of Work: Identify the root cause of the incorrect data across all Zabbix graphs. Review the Zabbix history, trends, graph calculations, housekeeping and TimescaleDB configuration. Correct the existing historical graphs across all monitored hosts and clients. Preserve all valid historical monitoring data. Implement a permanent solution to prevent the issue from recurring. Verify that the 7-day, 30-day and other historical graph periods display accurately. Provide documentation on the root cause and all changes performed. This is a large production environment, so all work must be carried out carefully through a supervised remote session. Please apply only if you have strong hands-on experience with: Zabbix 7.0 PostgreSQL TimescaleDB Zabbix history and trend tables Graph and data-retention troubleshooting When applying, please explain whether you have previously resolved a similar issue affecting all Zabbix historical graphs.