How ECB dodged a payment disaster in 10 hours of tech meltdown \ stacker news ~tech

pull down to refresh

How ECB dodged a payment disaster in 10 hours of tech meltdown www.reuters.com/markets/europe/how-ecb-dodged-payment-disaster-10-hours-tech-meltdown-2025-03-06/

454 sats \ 4 comments \ @StillStackinAfterAllTheseYears 8 Mar tech

I keep hearing that an advantage of fiat is that there are support systems and things in place to prevent disasters, and yet, seems like they only barely worked here (and even then, after the fact).

view all related items

121 sats \ 0 replies \ @optimism 8 Mar

Here's the good news:

This is where decentralization truly shines. Even with current pool centralization concerns.

Scenario: Foundry (the currently biggest pool at ~ 32%) goes down for a day across the board. (This fwiw would be about 100x as incompetent as the ECB as there's Foundry's money and reputation at stake, the ECB is a monopoly.)

The block time will expand with a few minutes and the backlog may grow.

However, most mining operations also don't like losing revenue, so assuming that Foundry would immediately shut down their stratum servers upon not being able to actually produce blocks, miners will switch to another pool either instantly (configured fallback pool) or within the hour (manually) and block spacing will no longer be affected.

In any case, the deviation will be short lived and totally within normal block spacing distribution: no transactor will likely even notice, except when looking at pool stats because a discrepancy would show: "Foundry did not produce any blocks that day, wth?" and of course the endless FUD on the bird app / half of IBIT being sold because the Investoors Sans Technology shall shit their pants.

21 sats \ 1 reply \ @itsrealfake 8 Mar

they say it was faulty hardware..

The ECB's latest annual report shows T2 was available at all times in 2021, 2022 and 2023. In 2020 it was up and running 99.46% of the time - below the targeted 99.7%. "A payment system that never fails may not be buildable," said Aaron Klein, a senior fellow at U.S. think-tank the Brooking Institution who specialises in financial technology. "And, if it is, it may be more costly than tolerating a few hours of delay."

56 sats \ 0 replies \ @optimism 8 Mar

Which is an inconceivable screwup if true.

A decade ago I've replaced tx processing systems 2 orders of magnitude smaller than this in terms of money flowing through them that were 20 years old (i.e. older than the ECB was at that time), most often clustered across many servers, everything triple redundant per site and have at least a fully sync'd hot site for fallback in case an entire datacenter gets EMP'd.

So "hardware failure" is either a lie (most likely), or the competence is nowhere to be found (still kinda likely.)

0 sats \ 0 replies \ @mf 8 Mar