Crypto Wallet Network Fallback Mechanisms: What Happens When Infura Goes Down and Your MetaMask Can’t Connect

On November 25, 2024, Infura, the most widely used Ethereum RPC provider, experienced a sustained outage that left millions of MetaMask users unable to send transactions, check balances, or interact with smart contracts. MetaMask’s dependency on Infura as its primary node provider meant that users faced complete wallet paralysis during the incident—not because their keys were unsafe or their funds were at risk, but because the application could not reach the blockchain. This scenario exposes a critical architectural weakness in many popular wallets: the reliance on a single remote procedure call (RPC) provider creates a concentration of availability risk that users rarely understand until it becomes a problem.

The distinction between wallet security and wallet availability matters more than most users realize. A crypto wallet that protects your private keys perfectly can still fail to function if it cannot communicate with the network. This article examines how different wallet designs handle network outages, which wallets maintain automatic fallback mechanisms, which remain offline during provider failures, and what practical steps users can take to protect themselves against unavoidable service interruptions. Understanding these trade-offs reveals why wallet comparison tools should evaluate not only custody and supported blockchains but also the resilience of each wallet’s connection infrastructure.

Visual comparison of single-provider versus multi-node fallback architectures in MetaMask, Phantom, Rabby, and other cryptocurrency wallets during RPC outages

How RPC dependency created a single point of failure in wallet architecture

A blockchain wallet does not store the blockchain. It stores cryptographic keys and can sign transactions, but to broadcast those transactions or query balances, it must ask a node to relay the message. An RPC provider acts as that intermediary, receiving requests from the wallet and returning information about accounts, balances, transaction history, and confirmation status. MetaMask historically routed the majority of its Ethereum requests through Infura, an Infura-owned service that became the de facto standard for web3 applications. This arrangement was economically efficient: Infura could distribute costs across millions of users, and MetaMask could avoid running its own infrastructure.

The vulnerability lay in the assumption of perpetual availability. When Infura experienced technical problems, database issues, or capacity constraints, every MetaMask user on Ethereum suddenly lost the ability to interact with the network. The wallet software itself remained functional; the private keys were never at risk; the user’s funds remained on the blockchain. But from the user’s perspective, the crypto wallet had stopped working. Balance queries returned errors, transaction submissions failed, and the interface displayed generic connection messages.

This scenario is not unique to MetaMask or Infura. Any wallet relying on a single RPC endpoint—whether that endpoint is provided by the wallet developer, rented from a service, or run by the wallet’s parent company—introduces an availability single point of failure. The architectural decision to use one provider rather than multiple sources reflects a trade-off between simplicity and resilience. A single endpoint is easier to monitor, easier to debug, and easier to align with the wallet’s user interface expectations. Multiple endpoints introduce complexity: handling divergent responses, managing fallback logic, deciding which provider to trust when they disagree, and distributing load fairly among multiple services.

Some wallet developers have chosen to absorb that complexity. According to information published on crypto wallet review platforms, wallets such as Phantom and Rabby implement multi-node fallback systems that automatically switch to alternative providers when the primary endpoint becomes unavailable. This approach shifts engineering effort from the user to the developer, reducing the operational burden on end users while increasing the developer’s responsibility to maintain multiple provider relationships and test fallback behavior regularly.

Software wallet approaches: MetaMask, Phantom, and Rabby compared

MetaMask’s architecture has evolved since its early days, but its core dependency pattern has remained relatively consistent. The wallet now offers the option to add custom RPC endpoints for different networks, but most users rely on the default providers bundled with the application. For Ethereum, this has historically meant Infura; for other chains, MetaMask may use different services or community nodes. When an outage occurs, the user’s only recourse is to manually add an alternative RPC endpoint by copying and pasting URLs into settings—a process that is not intuitive for non-technical users and is inaccessible to users in the middle of an urgent transaction.

Phantom Wallet, developed for Solana but now supporting Ethereum and multiple other chains, takes a different approach. Rather than defaulting to a single provider, Phantom maintains relationships with several Solana validators and Ethereum node services. If one provider becomes slow or unresponsive, Phantom automatically routes requests to the next available option without requiring user intervention. This design prioritizes availability at the cost of additional complexity in the wallet’s network layer. The trade-off is justified because wallet downtime undermines user confidence regardless of the underlying security model.

Rabby, a more recent entrant to the wallet market, has adopted a multi-provider strategy from inception. Rabby queries multiple RPC endpoints in parallel for critical operations such as balance checks and uses the fastest responding source for display. This redundancy comes with a cost: Rabby’s network consumption is higher than single-endpoint wallets, it may expose the user’s queries to more service providers, and the increased complexity creates more surface area for subtle bugs. However, the architecture allows Rabby users to continue functioning even when some providers experience outages.

The practical difference emerged clearly during the November 2024 Infura outage. MetaMask users reported widespread connection failures and had to resort to manual RPC switching. Phantom and Rabby users either experienced no interruption or a brief stall followed by automatic provider switching. This difference was not a matter of wallet security—all three wallets protected keys equally well—but rather a matter of wallet availability and user experience design.

Hardware wallets, exchange wallets, and their own network dependencies

Hardware wallets such as Ledger, Trezor, and SafePal do not run full nodes either; they rely on companion software or associated services to query the blockchain. A Ledger Nano S connected to a computer running Ledger Live depends on Ledger’s own infrastructure to check balances and broadcast transactions. Trezor users connecting to the Trezor Suite similarly depend on Trezor’s backend services. These devices secure the key material through physical isolation, but they cannot access the network independently—a separation that is both a security feature and an operational dependency.

When Ledger’s infrastructure experiences degraded service, Ledger Live users face the same outage symptoms as MetaMask users facing an Infura outage. However, hardware wallet manufacturers have more control over their entire stack. Ledger can upgrade its infrastructure, add redundancy, or implement failover mechanisms without requiring users to update the firmware on their physical devices. The trade-off is that users of hardware wallets are bound to the manufacturer’s choice of infrastructure; they cannot easily switch providers the way a MetaMask user could add a custom RPC endpoint.

Exchange-integrated wallets—such as Coinbase Wallet, Binance Wallet, and OKX Wallet—present a different model. These wallets often use the exchange’s own infrastructure to relay transactions. A user might assume that an exchange’s infrastructure is more robust than a third-party RPC service, but exchange outages can still occur, and an exchange wallet failure now has the added dimension of potential custody concerns. If an exchange wallet becomes unresponsive, users may worry whether their keys are still safe or whether the exchange has frozen access. The psychological impact of unavailability is amplified when the provider also holds reserves or custody.

The Ethereum protocol itself does not specify which RPC provider a wallet must use; the protocol only defines how nodes communicate with each other. This flexibility means that wallet developers can choose their own availability strategies. Some choose simplicity and single providers; others choose resilience and multiple endpoints. Users evaluating a crypto wallet should understand not only which blockchains the wallet supports but also which infrastructure provider or providers the wallet depends on for each chain.

The hidden cost of RPC provider concentration

Beyond availability, RPC provider concentration introduces a subtle privacy concern that is rarely discussed in wallet security documentation. When you use MetaMask with Infura, Infura can observe your IP address, your wallet address, the timing and frequency of your queries, and potentially infer your balance and transaction intent. Infura claims not to persist this data, but the observation itself is unavoidable at the network level. Wallet users who value privacy should understand that their wallet’s network requests are visible to their RPC provider.

This concern extends beyond privacy to regulatory exposure. If a wallet’s primary RPC provider receives regulatory pressure, becomes subject to sanctions, or decides to implement geographic restrictions, all users of that wallet may be affected regardless of the wallet developer’s intentions. For example, if an RPC provider blacklists certain addresses or refuses to relay transactions from specific countries, the wallet becomes less useful for those users. The wallet developer can mitigate this by maintaining multiple providers with different regulatory jurisdictions, but this is not common practice.

According to FCA guidance on cryptocurrency custody, wallet developers and RPC providers operate in overlapping regulatory space. The FCA’s expectations for safeguarding of assets apply to custodians, but RPC providers themselves may not be subject to the same licensing requirements depending on their specific function. This regulatory gap means that users cannot assume that an RPC provider operates under the same oversight as a regulated wallet or exchange. The layer that connects your wallet to the blockchain may be subject to less stringent requirements than you would expect.

Fallback strategies for users: custom RPC, bridge solutions, and alternative wallets

For users willing to engage in more technical configuration, adding custom RPC endpoints provides immediate redundancy. Public endpoints are available for Ethereum, Bitcoin, Solana, and most other major blockchains through services such as Alchemy, QuickNode, and various decentralized networks. A user can add multiple endpoints to MetaMask and manually select between them if the primary endpoint fails. This approach gives the user control but requires knowledge and effort to set up and maintain.

An alternative approach is to use a wallet that handles fallback automatically. Switching to Phantom or Rabby eliminates the need for manual endpoint management, assuming the user is willing to adopt a different wallet interface and security model. This transition is not costless: moving funds between wallets requires transactions, understanding a new user interface takes time, and some users may have security concerns about switching software. However, for users who experience frequent outages or who have already suffered through a provider failure, the cost of switching may be justified by improved availability.

A third approach is to maintain multiple wallets with different primary providers. A user might keep ETH on MetaMask for everyday use and maintain an additional wallet such as Phantom or Trust Wallet for the same funds. This approach is not recommended for securing large amounts because it multiplies the attack surface and the number of recovery phrases that must be protected. However, for small amounts or funds that are regularly accessed, maintaining a redundant crypto wallet provides insurance against any single wallet’s availability issues.

Bridge solutions such as running a local node provide the strongest form of resilience but require substantial technical knowledge and resources. A user running a full Ethereum node can point MetaMask or any other wallet to their local node, eliminating dependence on external RPC providers entirely. However, running a node requires storage, bandwidth, computing power, and time to synchronize. This solution is practical for users with significant holdings or advanced technical expertise, but not for casual users.

What wallet developers must improve: redundancy, transparency, and user communication

The ideal resolution to RPC provider outages is not better user documentation or manual workarounds. It is wallet architecture that automatically handles provider failures without user intervention. This requires wallet developers to maintain multiple RPC provider relationships, implement intelligent fallback logic, and test these systems continuously. The November 2024 Infura outage should have been a wake-up call for any wallet developer still relying on a single provider.

Transparency about network dependencies is equally important. A wallet that clearly discloses which RPC providers it uses, how it handles provider failures, and what options users have during outages builds trust through honesty rather than obscuring technical details. Some wallets display generic “network error” messages without explaining whether the problem is the user’s internet connection, the blockchain network itself, or the wallet’s RPC provider. Better error messages could help users understand what is happening and what actions they can take.

User communication during outages matters significantly. When Infura experienced downtime, MetaMask eventually posted status updates, but many users had no way to know whether their wallet would recover or whether they had lost funds. Wallets should provide clear, immediate notification when known provider issues are affecting their services. This could be a banner in the application, a notification to the user’s email address, or an in-app status page.

The blockchain industry has not yet converged on best practices for RPC infrastructure resilience. CFTC oversight of cryptocurrency markets focuses primarily on trading and custody, not on wallet availability. However, as crypto adoption grows and more users depend on wallets for significant funds, availability will become a regulatory concern. Wallet developers should not wait for regulation to force redundancy; they should implement it now as a competitive and operational necessity.

Custody and BIP39 standards: why decentralization matters at the wallet level

The emphasis on RPC redundancy should not overshadow the more fundamental security distinction between custodial and non-custodial wallets. A non-custodial crypto wallet—whether it uses single or multiple RPC providers—never gives up control of the user’s private keys. If the wallet becomes unavailable, the user’s funds remain on the blockchain and can be recovered using the recovery phrase on any other wallet that supports the same blockchains. According to BIP39 standards, a recovery phrase generated by one wallet should be restorable on any other wallet that implements the standard, creating a form of portability that reduces lock-in risk.

This decentralized recovery mechanism is the critical insurance policy against wallet unavailability. If MetaMask becomes permanently unavailable—not just experiencing an outage but actually shutting down—a user can import their MetaMask recovery phrase into Phantom, Rabby, or any other non-custodial wallet and regain full access to their funds. This is true for Bitcoin wallets, Ethereum wallets, and most chains that use BIP39 or similar standards. The RPC provider outage is temporary; the recovery phrase is permanent.

Custodial wallets, by contrast, do not provide this recovery insurance. If Coinbase Wallet becomes unavailable, a user cannot simply import their keys elsewhere because Coinbase Wallet may not issue a recovery phrase at all. Some exchange-integrated wallets use keyless designs where the exchange controls the recovery process. This trade-off exchanges some security convenience for reduced user portability. If the exchange service becomes unavailable, the user depends entirely on the exchange to restore access.

Practical steps for wallet users to reduce outage risk

The most straightforward protection is to keep your recovery phrase stored safely and to know the seed phrase format your wallet uses. A MetaMask user with a secure recovery phrase backup can always recover funds on Phantom, Rabby, or another compatible wallet if MetaMask experiences prolonged failure. This is not the same as protecting against key theft or malware, but it is essential insurance against service discontinuity.

Users should also test their wallet recovery process before they need it. Create a new test wallet using your recovery phrase on a different wallet application, verify that it shows the correct addresses and balances, and then securely delete the test wallet. This test ensures that your recovery phrase is actually usable and that you understand the recovery process. Many users have discovered during emergencies that their recovery phrase is corrupted, written down incorrectly, or incompatible with the wallet they are trying to recover to.

For significant holdings, maintaining redundant access is reasonable. Keeping the same seed phrase imported on two different wallet applications—for example, MetaMask and Phantom—does not improve security, but it does provide availability redundancy. If one wallet experiences an outage or is discontinued, the other remains functional. This approach is only appropriate for recovery insurance and should never involve typing the recovery phrase into online websites or services.

Finally, users should monitor wallet infrastructure news and follow wallet developers on social media or through official blogs. The next major RPC provider outage is inevitable, and being informed in advance about which wallets use redundant infrastructure can help you avoid being caught by surprise. A wallet that has publicly committed to multi-provider fallback architecture is more likely to survive the next outage gracefully than one that still relies on a single provider.

Frequently asked questions

What should I do if my crypto wallet stops connecting to the network?

First, verify that your internet connection is working by testing another application. If your internet is fine, check the wallet’s status page or official social media for known outages. If the outage is affecting the wallet’s RPC provider, you can wait for service to restore or manually add an alternative RPC endpoint (if your wallet supports custom endpoints). As a last resort, you can switch to a different wallet application using your recovery phrase, though this requires knowledge of your backup seed phrase.

Is my crypto wallet more secure if it uses multiple RPC providers?

Multiple RPC providers improve availability and resilience, but they do not directly improve security of your private keys. A wallet’s key security depends on whether it is custodial or non-custodial, how it protects seed phrases, and whether it resists malware. However, a wallet using multiple RPC providers is less likely to leave you stranded during an outage, which is an important practical benefit. Security and availability are separate concerns that both matter for a useful wallet.

Can I lose my crypto wallet funds if my RPC provider goes down?

No. Your funds exist on the blockchain itself, not in your wallet application. An RPC outage prevents you from interacting with the blockchain, but it does not affect your actual coins or tokens. As long as you have your recovery phrase backed up safely, you can always regain access using a different wallet application. The outage is inconvenient but not a threat to your funds’ security.