SDS API timeouts — some market categories delayed
San Diego Studio's API timed out on and off through the day. Player-card prices kept updating, but stadium/equipment/sponsorship/unlockable listings lagged — worst from ~7:30am–1:15pm ET, with a short repeat around 4:15pm ET. Fully caught up by ~4:30pm ET.
Affected: Market prices, Card sync
Read the full write-up →Hide write-up
What happened
Starting around 7:15am ET, San Diego Studio's API — the upstream we pull all card and listing data from — began timing out intermittently (30-second request limit exceeded). That's their service, not ours, and it has rough patches now and then.
Our sync jobs read from that API, so the timeouts rippled into a few places:
- Market data (worst impact): from ~7:30am to ~1:15pm ET, the market refresh couldn't pull a handful of item categories — stadiums, equipment, sponsorships, and unlockables. Listings in those categories were frozen at their late-morning values for that stretch. A shorter repeat hit ~4:15pm–4:30pm ET. The core player-card market kept refreshing normally throughout — the delay was contained to those cosmetic categories.
- Card / catalog sync: the nightly-style card, item, and supercharged syncs hit brief timeouts on and off through the day. Each one failed fast and retried on its own — no category went stale for more than a few minutes at a time.
As always, when a refresh can't complete we'd rather show you a slightly older price than a wrong one, so affected listings held their last-known values rather than writing partial data.
What we did
- Detected the upstream timeouts automatically via our SDS sync-health monitoring (each failure is recorded, and the circuit breaker pauses a sync rather than writing bad data).
- Let the retry/backoff logic do its job — most syncs self-recovered within minutes.
- For the longer market window, watched SDS recover, then let the next clean refresh backfill the delayed categories.
- Confirmed every category was current and no outage was still open before calling it resolved.
By ~4:30pm ET all categories were live and accurate again, and every sync was back to normal.
What we're doing differently
- Reviewing the 30-second timeout on the cosmetic-category market syncs — they were the slowest to recover, and a longer or smarter timeout may ride out an SDS slow patch instead of tripping the breaker.
- Using the new data-freshness badge on the market pages so a delay like this reads as "prices delayed — upstream is slow" immediately, instead of leaving you to wonder whether something's broken. That's the whole point of this status page: when the data looks off, you should be able to tell why in a few seconds.