Applying Search Schema fails

Since Microsoft has confirmed a service-side issue and the same SearchSettings block times out even on a new site, increasing the PnP timeout or splitting one mapping into many calls is unlikely to be a durable fix. It may simply leave a partly applied schema that is harder to audit.

For new sites, I would temporarily remove SearchSettings from the provisioning transaction so the rest of the site can be created successfully. Record the site URL, intended crawled-to-managed property mappings, template version, and provisioning time in a remediation queue. Retry the search configuration only after Microsoft confirms recovery, and make the retry idempotent so it checks an existing mapping before writing it.

For the current support case, provide one minimal site and one single failing mapping, together with the request correlation ID and UTC time from both the UI and automation attempt. Ask support to associate the case with the known engineering incident and provide service-side progress. Avoid repeatedly opening or writing the tenant schema during the degraded period, because that adds load without addressing the confirmed backend latency.