VPS Failover Drill: When Primary Dies Mid-Session

[NYC88]

Primary VPS dies. You are long. Chat is not a risk system.

Had this happen once mid-morning — RDP frozen, position open, heart rate stupid. Phone flatten worked because I'd drilled it on a Saturday like a paranoid person. That paranoia paid for itself in one morning.

Pre-staged failover kit

1. Secondary access to the same broker account: phone app and/or home desktop already workable with 2FA.
2. Written flatten procedure — sticky note energy: open app → symbol → flatten / cancel all. Not "I'll remember."
3. Secondary VPS is optional. Most people need a tested phone flatten more than a second datacenter.
4. Broker emergency phone saved offline.

Drill (Saturday or after close — what I run)

Code: Select all

[ ] Assume RDP dead
[ ] From phone: locate open positions in <60s
[ ] Cancel working orders
[ ] Flatten
[ ] Confirm flat in a second venue (web vs app)
[ ] Note every UI friction (extra confirms, biometric fail, etc.)
[ ] Fix frictions before Monday

Live failure — order of operations

1. Cancel working orders (stop unintended fills).
2. Flatten or hedge per your written rule — no "maybe it comes back."
3. Only then diagnose the VPS.
4. Journal: time detected, time flat, root cause.

If you've never drilled this, you don't have redundancy. You have hope. Hope is a terrible failover plan.

When did you last actually drill phone flatten — date, not "awhile ago"?
Phone app: cancel-all + flatten in under 60 seconds on your broker, or more clicks than you'd like?
Anyone keep a cold secondary VPS, or is phone-only enough for your size?
What's the ugliest UI friction you've hit trying to flatten on mobile?

[Fairman]

On **VPS Failover Drill: When Primary Dies Mid-Session**, the desk standard stays practical. VPS primary dying mid-session is the infrastructure cousin of a halt: your risk assumptions change in a second. Failover that only exists in a Notion doc is not failover. The drill is the product.

Replying in-thread with operable constraints — not forum combat. If these rails disagree with your framing, treat it as a checklist offer.

Infrastructure is risk management. ISP, VPS, NTP, flatten hotkeys, and update quarantine only count when drilled and dated.

## Rails that stay on the sheet
- Failover is halt-cousin.
- Parity of hotkeys matters.
- Warm standby beats cold install.

## Mid-session death protocol (written before you need it)
1. Detect: platform disconnect alerts, heartbeat script, or RTT watchdog — pick one and test it.
2. Stop trying to "save the trade." First job is risk-known.
3. Fail over to secondary; authenticate; inventory positions and working orders.
4. Flatten or reduce to the written emergency state (usually flat for scalp sleeves).
5. Do not re-enter from backup unless the sheet explicitly allows backup-path trading — most desks should not.
6. After restore: quarantine new size until you verify fills, hotkeys, and that you do not have duplicate working orders.

## Worked example — Primary VPS dies, no failover session
VPS kernel panic mid-session; local laptop not logged into broker; hotkeys different. Failover drill must include login, hotkey parity, and market data check — timed.

I keep examples in cents and clock time so Sunday review can tag the behavior — not rewrite it as a mood.

## Failure modes
- Ignoring the rails above on **VPS Failover Drill: When Primary Dies Mid-Session** and substituting confidence.
- Adding size after adverse selection because the story still feels right.
- Paying open microstructure tax that midday memory never modeled.

## When to skip
Skipping is a first-class decision — even when the chart still looks persuasive:
- Checklist incomplete at 09:20 ET — observe or one pre-written A+ only.
- Live spread outside max for that symbol and session bucket.
- Index violently disagrees with a single-name thesis and no reduced-size RS plan is on the sheet.
- Platform, hotkeys, locate, or flatten path unverified after reconnect/update.
- Daily loss, PDT slot, or prop buffer tripwire already touched.
- News/FOMC/CPI window where the size-cut card says stand down.

## Closing
For **VPS Failover Drill: When Primary Dies Mid-Session**, keep operable boredom: written rails, measured cents, index context when relevant, and a skip that needs no apology. Process quality beats ticket volume.