Tell HN: I salute everyone on call/working support through the holidays

Hacker News Discussion ↗

Holiday on‑call work in tech, healthcare, infrastructure, and public services draws both gratitude and criticism, as people weigh essential support against burnout, undercompensation, and exploitative expectations. Commenters describe strategies like code freezes, staggered rotations, and time‑zone coverage to reduce incidents, but also note that fragile systems, peak traffic, and security threats make some round‑the‑clock support unavoidable. Several voices stress the human side: the mental toll of being perpetually reachable, cultural and religious differences around holidays, and the value of approaching support staff with humility and kindness rather than blame.

Overall appreciation & scope

  • Many express gratitude to people working on-call or in support over holidays: IT/ops, SREs, date/time infrastructure maintainers, healthcare workers, road crews, military, zoo staff, cinema workers, etc.
  • Some distinguish between clear public-good roles (firefighters, water, power) and tech roles that keep ad/attention platforms running, which some call a “net negative.” Others counter that entertainment and ad-driven services still provide real value and help people unwind.

Holiday on-call practices & policies

  • Common practice: code freezes from late November or early December through early January, especially in e‑commerce and logistics, to avoid incidents during peak season.
  • Some companies lack such tooling/process and use freezes as a crutch, which is criticized as evidence of brittle systems.
  • Others push risky changes right before holidays (even full rewrites) to hit KPIs, causing severe outages and ruining vacations for ops/on‑call staff.

Reliability, tooling, and deployment philosophy

  • Advocated techniques: feature flags, gradual rollouts, automated rollbacks, error-rate monitoring, and partial deployments to limit blast radius.
  • Skeptics note many critical systems (POS, enterprise business logic) can’t safely rely on auto-rollback because “errors” may be silent logic bugs, not technical failures.
  • Consensus: all systems are ultimately brittle; freezes plus good tooling can coexist.

Human impact: stress, burnout, and boundaries

  • On-call is described as mentally draining; lingering anxiety about phones and alerts persists even off-shift.
  • Some now refuse on-call entirely or insist on contractual compensation; resentment is strong where it’s “off book.”
  • Offshore/low-cost engineers (e.g., in India) report expectations of 24×7 coverage under threat of job loss, framed by some as exploitation.
  • Others enjoy working holidays: quieter systems, fewer interruptions, ability to take vacations off-peak.

Cultural and holiday perspectives

  • Non-religious and non-Christian participants differ: some treat holidays as any other day, others value them as cultural/family gatherings irrespective of religion.
  • Suggestions to use diverse teams so people who don’t celebrate specific holidays can swap coverage for days that matter more to them.
  • Mentions of Israeli teams as a way to avoid downtime, with side debates touching on Sabbath observance and politics.

Support interactions & blame

  • Widely shared advice: approach support (internal or external) with humility, assume you might have erred, and be kind; this tends to elicit better help.
  • Debate over saying “I made a mistake” when the fault is unclear:
    • Pro: disarms defensiveness, skips blame games, leads to faster solutions.
    • Con: can be dishonest, abused by organizations/people to shift blame, or exploited by vendors to avoid responsibility.
  • Nuanced variants suggested: “I might have made a mistake,” “unless I’m missing something,” or focusing on problem-solving rather than fault.

Security and adversaries

  • Multiple comments warn that holidays are prime times for attackers: reduced staffing, slower responses, overloaded or distracted teams.
  • A few dark-humor posts note that poorly staffed offshore or DR environments are tempting targets.

Anecdotes & miscellany

  • Stories span quiet bunker shifts in the military, GPU cluster maintenance via IPMI at home, live video productions on New Year’s Eve, and emergency 911 date/time bugs causing prolonged cleanup.
  • Some highlight the importance of not being a “hero” who silently holds up failing systems; letting things fail can expose systemic risk and force organizational fixes.