Passer au contenu

Nous améliorons nos opérations pour mieux vous servir. Les commandes sont expédiées normalement depuis Laval, QC. Questions? Contactez-nous

Bitcoin accepté au paiement  |  Expédié depuis Laval, QC, Canada  |  Soutien expert depuis 2016

ICERIVER_FACTORY_RESET Info

IceRiver Factory Reset Procedure (20-Second Hold + Red Light)

ICERIVER_FACTORY_RESET — universal soft-reset operation on every IceRiver KS-series unit. 20-second reset-button hold (with red status LED flash confirmation) marks the active partition as 'needs refresh', reboots, and the bootloader copies the factory partition over the active partition. Clears web UI password, network settings (DHCP/static IP, hostname, DNS), pool URL/worker, frequency tuning and OC/UV profiles, time zone, and dashboard customizations. Does NOT change firmware version, hardware-level calibration, bootloader, kernel, or hashboard chip enumeration state. Procedure varies by model: KS0/KS3 family use a recessed reset button held while powered on; KS5/KS5L/KS5M hardware revisions may require holding reset while powering on (not after).

Informational — Monitor and address as needed

Affected Models: All IceRiver KS-series Kaspa miners — KS0, KS0 Pro, KS0 Ultra, KS1, KS2, KS3, KS3L, KS3M, KS5, KS5L, KS5M

Symptoms

  • Lost or forgotten the web UI password and the unit is otherwise running fine
  • Static IP set to a subnet no longer reachable (moved house, changed routers, switched ISPs); miner is now invisible on the network
  • Wi-Fi credentials saved against an SSID that no longer exists; unit can't reach DHCP
  • Pool URL wedged at a stratum that's been retired or migrated; web UI won't accept a new pool URL
  • OC/UV profile tuned past silicon-lottery limits; unit hashrate-degraded but otherwise alive
  • Frequency tuning stuck at values that no longer correspond to documented behaviour; manual rollback not taking
  • Web UI loads but is sluggish or flaky after months of uptime — config-state corruption from many incremental tweaks
  • Selling or relocating the unit; want to wipe all credentials before handover
  • Repeated minor fault codes (`110`, `300`-series) clear after reboot but return — config-state corruption pattern
  • Want to verify the unit boots cleanly on factory firmware before triggering a firmware update
  • First boot after purchase: validate the unit comes up clean before adding credentials
  • Suspect a previous owner's settings remain on a used unit; want a clean slate
  • Web UI accepts settings changes but the miner ignores them (config write isn't persisting)
  • After a power event: unit boots, web UI works, but settings are inconsistent or partially defaulted

Step-by-Step Fix

1

Take notes first. Write down current pool URL, worker name, static IP (if used), Wi-Fi SSID/password (if used), OC/UV profile values, hostname, and time zone. Reset wipes all of this. The 5 minutes you spend cataloguing now saves 30 minutes of reconfiguration after. Take screenshots of every web-UI tab if the unit is still reachable.

2

Power on the unit and wait `2 minutes` for full boot. All four front-panel LEDs (`D1`-`D4`) should cycle to their normal-running state; PSU fan running; chassis fans ramping. If the unit isn't fully booted, the reset daemon isn't running yet and a button hold won't be honoured.

3

Locate the reset button on your specific KS model. KS0/KS0 Pro/KS0 Ultra: recessed pinhole near the Ethernet jack on the rear panel — needs paperclip or small screwdriver tip. KS1/KS2/KS3/KS3L/KS3M: recessed button on rear or side panel. KS5/KS5L/KS5M: rear panel, accessible without a tool. Refer to the chassis sticker for hardware revision; some KS5 revisions require holding reset *during* power-on, not after.

4

Press and hold the reset button for the full `20 seconds`. Watch the red status LED on the front panel — within 5-15 seconds of holding, it should begin flashing, confirming the reset daemon has accepted the long-hold. Continue holding to the full `20 s` to ensure the request is committed; releasing earlier may not register a complete reset cycle.

5

Release the button and wait `5 full minutes` untouched. During this window the bootloader is reading the reset flag, copying the factory partition over the active partition, and rebooting. LEDs may cycle through several states; fans may stop and restart. Do NOT power-cycle, do NOT press buttons, do NOT unplug Ethernet — interrupting this window leaves the eMMC half-written and creates a brick condition that reset cannot recover from.

6

For KS5L/KS5M with the 'hold reset while powering on' hardware revision: power off the unit at the breaker or PSU AC switch. Wait `30 s`. Press and hold the reset button. While still holding, switch the PSU back on. Continue holding for `20 s` after power is restored, until the red status LED starts flashing. Some KS5 revisions disable the runtime reset GPIO during normal operation as a safety measure — chassis sticker hardware revision tells you which behaviour your unit has.

7

Verify the network reset took effect. After the 5-minute reboot window, check your router's DHCP lease table for the unit's MAC. The unit should have a fresh DHCP lease at a default hostname (typically `ICERIVER-<suffix>` or `KS3M-<suffix>`). If the old static-IP lease is still being assigned by your router because of a sticky DHCP reservation, remove the reservation and let the router assign fresh.

8

Verify the web UI password is back to factory. Default password on KS-series is documented in the IceRiver FAQ — typically a model-specific default printed on the chassis or in the user manual. If you don't have the manual, `iceriver.io/faq/` lists current defaults. Log in with the factory password as your first verification step.

9

Reconfigure pool URL, worker, and any static IP / Wi-Fi credentials. Use the notes from step 1. Take fresh screenshots after each setting is saved — useful for the next reset, useful as evidence for warranty support, useful if a future firmware update wipes settings unexpectedly.

10

Check firmware version after reset. Reset does not change firmware version, but confirm the unit is on the version you expect — sometimes operators discover after reset that they were on an older firmware than they thought because OTA updates had been failing silently for months. If older than the latest stable on `iceriver.io/firmware-download/`, plan a controlled OTA from the now-clean state.

11

If the reset hold doesn't produce a red LED flash, confirm the unit is fully booted (step 2). If LEDs are stuck or fans aren't running, the reset daemon isn't alive and reset won't trigger — that's a brick scenario. If LEDs cycle normally but the long-hold produces no flash response, the reset button itself may be faulty. Multimeter on continuity, probe the button leads with the button un-pressed — should read open. If it reads closed (electrically pressed when mechanically released), the button is stuck and needs replacement.

12

Verify reset took effect via UART (advanced). USB-to-TTL adapter (`CH340`/`CP2102`/`FTDI`, `3.3 V` logic) connected to the model-specific debug header at `115200 8N1`. Power-cycle and observe U-Boot output. After a successful reset, U-Boot logs the partition copy operation in its boot messages — typically a line referencing `mmc write` or partition refresh. If you don't see that log line, reset didn't take and the unit may be running on cached settings.

13

Edge case: reset hangs mid-cycle. If the unit goes through the LED flash + reboot but then gets stuck on a single LED state and never finishes booting, the partition copy was interrupted mid-write and the active partition is now half-corrupt. Recovery: SD-card recovery boot (Tier 2 of the IceRiver brick recovery page) or UART + eMMC reflash (Tier 3 of brick recovery). Reset CAN brick a unit if interrupted; the 5-minute 'don't touch it' rule exists for exactly this.

14

Edge case: third-party firmware reset behaviour is undefined. If your unit is running `xyys`, `tswift`, or any `rdugan/iceriver-oc` fork: some preserve the IceRiver factory partition intact and reset works as documented; others overwrite the factory partition with their own defaults, in which case reset returns to the third-party firmware's defaults rather than IceRiver stock; a few forks disable the reset GPIO entirely. Test reset behaviour on a unit you're willing to brick before relying on it on a production unit.

15

Edge case: persistent settings survive reset. Rare but reported — a unit appears to reset (LED flash, reboot, default IP) but a specific setting (typically pool URL) reappears at the broken value. Two causes: (a) the factory partition itself has been modified by a previous firmware flash and now contains the broken setting as its 'default', or (b) a hardware-revision-specific configuration cache lives outside the active partition (some KS5 revisions stash pool config in NVRAM that reset can't touch). Recovery: full firmware reflash from `iceriver.io/firmware-download/` to overwrite the factory partition, then reset.

16

Stop DIY and ship to D-Central when (a) reset hold produces no red LED flash with the button verified electrically sound, (b) reset triggers a reboot but the unit doesn't come back up after 30 minutes (interrupted-reset brick), (c) settings persist with broken values across multiple reset cycles (factory partition corruption), or (d) reset is needed on undocumented third-party firmware and one attempt has already gone wrong. Tier 4 destination has UART tooling, eMMC programmer access, and stock IceRiver firmware images for every KS model + hardware revision currently in the field.

When to Seek Professional Repair

If the steps above do not resolve the issue, or if you are not comfortable performing these repairs yourself, professional service is recommended. Attempting advanced repairs without proper equipment can cause further damage.

Related Error Codes

Still Having Issues?

Our team of Bitcoin Mining Hackers has been repairing ASIC miners since 2016. We have seen it all and fixed it all. Get a professional diagnosis.