Solana Network Outages: The Full History and What Firedancer Changes
Solana network outages have repeatedly tested the blockchain’s reliability, but the upcoming Firedancer validator client aims to change that history.
Solana’s Outage Timeline: 2021–2024
On 14 September 2021 the network went offline for 17 hours after transaction volume spiked sharply from automated bots.
Four months later, on 4 January 2022, Solana halted again for approximately six hours during an NFT mint that flooded the chain with simultaneous transactions.
A shorter interruption occurred on 1 May 2022 when the network stopped producing blocks for roughly two hours after another surge in activity tied to a popular token launch.
On 30 June 2022 validators observed a six-hour outage triggered by a configuration error that halted block production across the cluster.
February 2023 brought two separate incidents: a four-hour halt on the 6th linked to a validator software rollout and a two-hour stoppage on the 25th caused by an overload during a high-profile token airdrop.
The final major event in the period took place on 6 February 2024, when the chain paused for five hours after a large influx of transactions from a newly launched meme coin.
Across these six events the common pattern was a sudden external demand spike that exceeded the network’s capacity at the moment, producing extended periods without new blocks.
Technical Causes: A Comparative Breakdown
Solana outages have stemmed from distinct technical failure modes rather than a single recurring flaw. The most frequent triggers involve network congestion from high-frequency transaction bursts that overwhelm validator memory allocation, combined with bugs in the proof-of-history implementation that propagate fork inconsistencies across the cluster.
| Outage Date | Primary Cause | Affected Components | Recovery Time |
|---|---|---|---|
| September 2021 | Transaction spam exceeding gossip bandwidth | Validator memory, leader schedule | 17 hours |
| January 2022 | NFT mint transaction flood | Network stack, memory allocation | 6 hours |
| May 2022 | Activity surge from token launch | Consensus engine | 2 hours |
| June 2022 | Configuration error | Block production cluster | 6 hours |
| February 2023 | Software rollout and airdrop overload | Validator software, network capacity | 4–2 hours |
| February 2024 | Meme coin transaction influx | Priority fee logic, block production | 5 hours |
Each incident exposed tight coupling between the runtime and networking layers. Spam attacks exploited the lack of rate limiting at the gossip protocol level, while software bugs often originated in untested edge cases during leader rotation. Recovery consistently required coordinated validator restarts or targeted patches rolled out via the release channel. These patterns highlight how single-thread bottlenecks and insufficient isolation between components amplified small anomalies into network-wide halts.
Ecosystem and Market Consequences
Each major Solana outage produced measurable effects on TVL, token price, developer activity, and user retention. After the September 2021 incident, ecosystem TVL fell roughly 15 percent as liquidity providers moved funds to other chains. SOL price declined 25 percent in the following week. Developer commits on Solana repositories dropped 20 percent in the subsequent month while daily active addresses fell 30 percent.
The February 2023 outage repeated the pattern. TVL contracted another 12 percent within ten days. SOL lost 18 percent of its value before stabilizing. GitHub activity tied to Solana projects decreased 28 percent, and active wallet counts remained 35 percent below prior averages for six weeks. The February 2024 event triggered a further 10 percent TVL reduction and a 22 percent price drop. Developer output slowed again, and user retention metrics showed only partial rebound even after network restart.
Across the period, repeated downtime produced cumulative TVL erosion exceeding 25 percent from peak levels and sustained suppression of both developer contributions and returning users.
The Origin and Goals of Firedancer
Jump Crypto began developing Firedancer in 2022 as a ground-up rewrite of the Solana validator software. The effort responded directly to repeated outages by creating a second, fully independent client rather than patching the existing Agave codebase.
Firedancer’s core architecture deliberately avoids code overlap with Agave in consensus, networking, and runtime layers. This separation means a flaw in one client cannot trigger the same failure in the other, providing the network with genuine client diversity.
Stated targets focus on sustained operation above 99.9 percent uptime even during peak load or partial validator outages. The project also aims to raise maximum throughput while reducing memory and CPU overhead, allowing validators to maintain performance without the resource spikes that previously amplified disruptions.
How Firedancer Addresses Historical Failure Modes
Firedancer reimplements the validator in C with a modular architecture that isolates networking, consensus, and execution. This design prevents a single faulty transaction from halting the entire node, unlike Agave where processing bugs repeatedly triggered network-wide stalls.
Memory handling changes include custom allocators and fixed-size buffers that eliminate the out-of-memory crashes seen during traffic spikes. The networking layer uses a hardened QUIC stack that discards malformed packets without dropping peers, addressing the packet-flood issues that caused earlier outages.
Consensus logic runs redundant fork-choice checks in parallel, reducing the chance that a malformed block propagates during partitions. Execution adds stricter runtime bounds and anomaly throttling that Agave lacked, catching edge-case program behavior before it escalates.
Independent client diversity further limits the impact of any remaining Agave-specific flaw. Together these targeted differences directly close the failure modes documented in prior incidents.
FAQ
How many full halts has Solana mainnet experienced?
Seven full halts occurred between December 2020 and February 6, 2024, with total downtime across those events reaching roughly three days. Coordinated validator restarts were required each time after selecting a safe rollback slot with at least 80 percent online stake.
When was Solana’s most recent full outage?
The last full mainnet halt took place on February 6, 2024. Since that date the network has recorded 100 percent block-production uptime, spanning roughly thirty consecutive months as of September 2026.
What share of stake runs Firedancer?
As of mid-2026, approximately 14 percent of stake operated the full Firedancer client and another 26 percent ran the Frankendancer hybrid version. Full Firedancer reached mainnet on December 12, 2025.
Did the August 2026 TeraSwitch incident halt Solana?
No. A routing failure at TeraSwitch took nearly 29 percent of stake offline for about 33 minutes on August 12, 2026, yet blocks continued to be produced and transactions processed without any coordinated restart.
How does client diversity from Firedancer improve reliability?
Firedancer is an independent C/C++ validator client with a modular tile architecture and no shared code with the original Agave client. This separation limits the chance that a single bug will affect the entire validator set at once.
Is Solana reliable for high-volume trading today?
The network has sustained record transaction volumes since February 2024 without full halts. Incremental upgrades and client diversity have removed the primary historical failure modes that previously caused outages.
Private crypto swaps
Best rates. Secure. Wallet to wallet
5 Min
