Solana Network Suffers 20-hour Outage, Root Cause Still Unknown

Solana outages have yet again sparked questions about the design of the premier Layer 1 blockchain.

The validators tried to restart the network twice over the weekend after the on-chain activity froze. At one time, Solana was processing only 93 TPS. Following the performance degradation, the validator community eventually had to opt for a synchronized chain restart of the network.

Solana Outage Again

According to the official blog post, the cause of the outage has not been determined yet, but Solana is still under active investigation. The team said the network was unable to recover without intervention by the validator community.

The decision to restart the network was suggested by the engineers debugging the issues. Following this, the validator community jointly decided to downgrade to the previous stable release, v1.13.6. Simultaneously, 1.14 was rolled out to reduce restart risk.

“An initial restart attempt was called off to allow time for more thorough data analysis, and ensure no user transactions would be impacted. After further analysis, the community collectively restarted the network using an older slot than the one previously selected. No confirmed user transactions were rolled back or impacted.”

Community Reaction

Last year, Solana endured 14 outages, both with small and large downtimes. Infuriated community members pointed out that the continued outages could be due to a massive design flaw in the network. A user called “DBCrypt0” said Solana’s on-chain consensus model involves network transactions consisting of consensus communication between validators and the transactions themselves, which “inflates the transaction volume as well as TPS.”

They further criticized the network, saying that out of the 4k TPS that Solana boasts, only 10% are “actual” transactions while adding that the majority of the volume is made up of validators’ messages, which ultimately bogs down the system.

In case of an outage, the communication between the validators ceases, during which they turn to Discord to chalk out a plan. However, two-thirds of validators must agree on a solution to get back up, and some of them could be offline or unaware of the outage.

Solana’s tech slowdowns aren’t new. Last year, the team behind its development even laid out ways to mitigate such incidents. More recently, the developers implemented a new feature called “Priority fees” to enable users to pay extra to avoid congestion.