Dear all, Summary:
It has come to our attention that specific aspects of an integration between PeeringDB and IX-F (in its current form) is causing considerable issues for some PeeringDB users. We have temporarily disabled importing data from IX-F into PeeringDB to alleviate some headache. Background: Every day PeeringDB ingests JSON formatted information from IX-F ("IX-F data"). The intent of this ingestion process is to help align information submitted to PeeringDB by network operators and information submitted to IX-F by IXP operators. The expectation was that this would help streamline scenarios such as where an ASN disconnects from an IXP but didn't update their PeeringDB record reflecting disconnection, and if the IXP subsequently would re-assign the IP address to a newly participating ASN, often times the new ASN wasn't able to add the new IXP connection to their PeeringDB record without intervention from the PeeringDB Admin Committee. However there also are scenarios where an ASN is in the process of connecting a new IXP circuit, and wants to update their PeeringDB record and add this new connection ahead of the completion of the full provisioning process to streamline their peering discussions. In such cases the lack of information about this new circuit in the IX-F data could cause a deletion of this user submitted information. A significant challenge here is that sometimes records in PeeringDB are out of date (which hinders new IXP participants), and sometimes IX-F data doesn't (yet) contain information or this information is out-of-date (either case hinders a different group). I think the path forward is to express the relations and automated actions between the information sources as a state machine which (unlike the current version) does account for the full lifecycle from provisioning to decommissioning of circuits to IXPs. I hope you as community will continue chime in on what we should do next. On github there are various issues open related to this problem space: https://github.com/peeringdb/peeringdb/issues/505 https://github.com/peeringdb/peeringdb/issues/513 https://github.com/peeringdb/peeringdb/issues/518 https://github.com/peeringdb/peeringdb/issues/540 We want PeeringDB to be a useful, reliable, and consistent tool for everyone. However the current implementation of the process to augment PeeringDB's data with IX-F data appears to be unable to account for all possible scenarios in a way that is without exception helpful to all PeeringDB's users; hence the decision to disable the importer for now. Kind regards, Job _______________________________________________ Pdb-announce mailing list Pdb-announce@lists.peeringdb.com https://lists.peeringdb.com/cgi-bin/mailman/listinfo/pdb-announce