SAPI Lodging Workspace
Ops Review
Viewing Sep 22 → Sep 29 · Back to latest

SLDG Shield · Sep 22 → Sep 29

Andrei GeorgescuRenê Cardozo
2issues · 0 customer-facing72production pings · ~22 pages1488 → 713Alert transitions2open follow-ups · 0 without ticket

Week overview

Monitor rework: error rates instead of error counts

Search monitors moved from error counts to error rates ([SLDG-3108] Search monitors on error rates) mid-week, so low-traffic suppliers stopped paging on single failures.

Effect
Alert transitions per week1488 → 713−52%

Issues & resolutions

FBO error monitor flapping on low-volume suppliers

OperationalSep 22 → Sep 24
Problem
The findBestOffers error monitor counted errors, so a handful of WebBeds timeouts at night paged the Shield several times.
Action
Moved the monitor to an error rate with a request floor ([SLDG-3108] Search monitors on error rates).
Effect
FBO monitor transitions212 → 9

CCA reseed dead letters in sandbox

OperationalSep 25
Problem
Sandbox CCA reseeds sent blank stored totals, which the CCA job treated as invalid and dead-lettered.
Action
Treat blank stored totals as missing ([SLDG-3130] Blank stored amounts on the CCA path).
Effect
Sandbox CCA dead letters4 → 0
Follow-ups (1)
[SLDG-3130] CCA reseed dead letters
Minor items (1)
Pagingincident.io is now the only pager ([SLDG-3109] incident.io-only paging)

Needs attention

Item
SLDG follow-ups2
incident.io-only pagingSLDGSLDG-3109
AVH child occupancy at BookSLDGSLDG-2689
Waiting on other teams1
WebBeds timeouts at night: is a longer timeout on findBestOffers safe?SPIwaiting

Shield load

Production · #sapi-lodging-alerts72 pings38 non-actionable · ~22 pagesDev/staging · #sapi-lodging-alerts-secondary41 pings33 non-actionableAlert transitions1488 → 713
Follow-ups · 2 open of 3
Search monitors on error ratesSLDGSLDG-3108
incident.io-only pagingSLDGSLDG-3109
AVH child occupancy at BookSLDGSLDG-2689
Done this week · 1 follow-ups closed
Search monitors on error ratesSLDGSLDG-3108