• As a result of the recent outage of some nodes in the network, the SDF has released version 16.0.0.0 for Stellar Core to prevent future downtimes.
• The community still needs to vote on the implementation.

Following the crash of some Stellar network nodes on April 6, the Stellar Development Foundation (SDF) has released a major upgrade with v16.0.0.0. In an official statement, the SDF said that the update fixes the issues that were recently encountered. Focusing on preventing nodes from losing synchronization with the network, the organization has asked for the support of the entire community.

As CNF reported, the outage led to an investigation by the SDF engineering team. They were able to determine that the majority of the network remained online, with most of the affected nodes being operated by the SDF and a wallet called Lobstr. However, “many organizations” are still in the process of “regaining access to the network”.

The Stellar community gets to decide

The fixes for the bugs will be available for Stellar Core v16.0.0 and Stellar Core v15.5.0. Operators of a Horizon node will need to upgrade to version 2.1.0. Then, on April 10, the community gets to vote to decide on the implementation of Protocol 16 which “should solve all the problems and bring everyone back online.”

We are planning to schedule a validator vote on that upgrade for Saturday 04/10/21 at 1500 UTC. In advance of that vote, all node operators need to install up-to-date software. If you have not, your node will lose sync with the network.

Furthermore, operators of a Horizon node will need to consider additional factors. Within the upgrade process, there will be a database migration. Therefore, they will need to enable an additional feature or do an extra upgrade. The SDF clarified:

The migration will be performed by the ingestion system (through a state rebuild) and, thus, if some of your Horizon nodes are not ingestors (i.e. no –ingestion flag enabled) you may experience 500s in the GET /claimable_balances/ requests until an ingestion node is upgraded.

The update for Horizon will include new functions that will allow the operators to determine if the node is “healthy” to receive traffic, add asset statistics summaries, give users more control over the storage destination for Captive Core information, among others.

As usual in these processes, users who fail to upgrade their nodes will lose synchronization and network access until their nodes have the new version. Remarkably, the SDF has postponed the integration of features, such as asset clawback, that would have been originally be part of Protocol 16. The new features will be deployed with Protocol 17, the organization added:

We appreciate everyone’s support, cooperation, and patience as we work to resolve these issues, and we are excited for this upgrade to restore service to anyone who is still having trouble.