Ran into an issue today with a client that is using an instance of OSRM to compute routes on a mix of data from OSM (and thus valid IDs) and data from a local source (which was given arbitrary IDs > 1quadrillion to avoid conflicts, we couldn't renumber everything right now). When performing a route match request against the higher node IDs the reported route coordinates were correct, but the node IDs were not. Upon further investigation it appeared that the reported IDs were overflowing beyond 2^34.
Steps to reproduce
Start an OSRM match server with some node IDs higher than 2^34. Make a route match request (eg. v1/driving). Check the reported node IDs.
osrm-backend version: 5.26.0
Specifications
Unknown server the client operates. It's linux, some VM from a cloud provider.
The good news
We have like 1600 days to resolve this before actual OSM data causes this issue.
Issue
Ran into an issue today with a client that is using an instance of OSRM to compute routes on a mix of data from OSM (and thus valid IDs) and data from a local source (which was given arbitrary IDs > 1quadrillion to avoid conflicts, we couldn't renumber everything right now). When performing a route match request against the higher node IDs the reported route coordinates were correct, but the node IDs were not. Upon further investigation it appeared that the reported IDs were overflowing beyond 2^34.
Steps to reproduce
Start an OSRM match server with some node IDs higher than 2^34. Make a route match request (eg. v1/driving). Check the reported node IDs.
osrm-backend version: 5.26.0
Specifications
The good news
We have like 1600 days to resolve this before actual OSM data causes this issue.