Unraid Parity Check Duration + Scheduling Calculator

This calculator estimates how long an Unraid parity check takes from your parity disk size and the speed your array checks at, shows how many quiet windows a long check needs if you split it, and explains what correcting and non-correcting checks do.

A parity check reads every disk in the array and compares the parity it calculates with the parity stored on the parity disk. It is routine maintenance: it finds unreadable sectors and mismatches while parity can still fix them, rather than in the middle of a rebuild. It also keeps every disk busy for hours, so it is worth knowing how long yours runs and when it fits.

Your array

Parity disk size (your largest disk)

Slowest disks in the array (assumed speed)

Your observed check speed, MB/s (optional)

The average speed of a past check on this server. When you enter it, it replaces the assumed speed and the activity setting.

Other activity during the check

Check mode and quiet time

Parity check mode

Preset to the mode Unraid uses for scheduled checks by default (source below).

Length of your quiet window (hours)

How often that window comes round

Estimated active check time

…

Fitting it into your quiet time

What this mode does

Reading the result

Zero errors: parity matched the data and every disk could be read during that check. That is strong evidence parity was consistent at the time. It does not guarantee a future rebuild, because a disk can fail or develop unreadable sectors later, and it does not check file contents against checksums.

Errors found: a count of places where calculated and stored parity disagreed, not a count of damaged files. A mismatch does not show which side is wrong. Look at every disk's SMART report, check cables and power, and read the system log before you decide to correct anything.

After an unclean shutdown: After an unsafe shutdown Unraid automatically starts a parity check using the mode set under Settings > Scheduler > Parity Check (non-correcting by default); it is not automatically a correcting check. (Unraid source, applies to Current docs (no version stated), checked 2026-10-06) Errors after a power cut or crash can come from the shutdown, but also from a failing disk, a loose cable or the power supply, so do not assume a correcting check is the right first step.

How we calculate

Active check time = parity disk size ÷ average check speed. Units are decimal: 1 TB = 1,000,000 MB, so 8 TB at 160 MB/s is 8,000,000 ÷ 160 ÷ 3,600 = 13.9 hours. A check runs to the end of the parity disk, which is at least as large as your largest data disk, so this tool asks for the parity disk size.
Average check speed is set by the slowest disk, the disk controller and anything else using the array. Without your own figure we use an assumed range for the slowest kind of disk you have (table below) and scale it for other activity. Both are assumptions, not measurements, which is why the result is a range.
Your observed speed, if you enter it, replaces both: it already reflects your disks, controller and usual load. Use the average speed of a past check that ran under similar conditions.
Dual parity adds no multiplier here: the check still reads every disk once.
Quiet windows = active time ÷ window length, rounded up. Splitting a check changes the calendar time, never the active time. The range runs from the fast end to the slow end of the estimate.

Assumed average speeds (MB/s)
Slowest disks	Parity check (reads)	Rebuild or parity build (writes one disk)
7200 rpm class hard drives	120 to 160	120 to 160
5400 rpm class or older hard drives	90 to 120	90 to 120
An SMR hard drive in the array	90 to 120	40 to 110
SATA SSDs	400 to 550	150 to 300

Assumed effect of other activity (share of the idle speed left)
Activity	Speed left
Idle: nothing else running	100% to 100%
Light: some streaming or reads	85% to 95%
Moderate: active use, some writes	65% to 85%
Heavy: big copies, VMs or containers busy	40% to 70%

An SMR disk is treated as the slower hard drive range for a check, because a check only reads. Its write penalty matters for rebuilds and parity builds, which the rebuild estimator covers.

Worked example

An 8 TB parity disk, 7200 rpm class drives, moderate activity during the check, a 7 hour quiet window every night.

Speed: 120 × 0.65 = 78 MB/s at the slow end, 160 × 0.85 = 136 MB/s at the fast end.
Active time: 8 TB ÷ 136 MB/s = 16 h to 8 TB ÷ 78 MB/s = 28 h.
Split into 7 hour nightly windows: 3 to 5 nights.
The same array with an observed speed of 145 MB/s from a past check: about 15 h, or 3 nights.

Correcting and non-correcting checks

Mode	Reads	When it finds a mismatch
Non-correcting	Every data and parity disk	Records it and changes nothing
Correcting	Every data and parity disk	Records it and rewrites parity to match the data

Both modes read the same data, so they take about the same time; a correcting check also writes parity at each mismatch it finds, which adds time only there. The difference is what happens at a mismatch. A mismatch only shows that data and parity disagree, not which side is wrong: if a disk returned bad data, a correcting check writes parity to match it. That is why it pays to find the cause of errors before correcting them.

What Unraid does by default, and what its documentation advises

Default mode for scheduled checks: Scheduled parity checks (Settings > Scheduler > Parity Check) default to non-correcting mode. (Unraid source, applies to Current docs (no version stated); default changed to non-correcting in the 6.11.2 release notes, checked 2026-10-06)
Whether a new install schedules checks, and how often: we could not confirm this from Unraid's documentation, so look at the scheduler settings on your own server.
What the documentation advises: The docs ADVISE automated, correcting checks monthly or quarterly. This is advice, not a stated install default. (Unraid source, applies to Current docs (no version stated), checked 2026-10-06)
Splitting a long check: A parity check can be split into smaller pieces run over multiple days or weeks in set time windows (cumulative parity check). The source does not say whether progress survives a reboot. (Unraid source, applies to Unraid 6.10.0+ (release notes 6.10.0), checked 2026-10-06)

No parity disk? A parity check needs parity to compare against. A read check is needed when the array has no parity devices assigned, or more disabled drives than parity drives; it examines every sector of every disk in the array. (Unraid source, applies to Current docs (no version stated), checked 2026-10-14)

The default and the advice are different things and can point different ways: a default is what the software does until you change it, advice is what the documentation suggests you choose. This calculator does not add a rule of its own.

Parity check time by disk size

Active time for an idle array at each assumed speed range, from the same formula as the calculator. Add time for other activity, or use your own observed speed.

Parity disk	7200 rpm class hard drives	5400 rpm class or older hard drives	An SMR hard drive in the array	SATA SSDs
4 TB	6.9 h to 9.3 h	9.3 h to 12 h	9.3 h to 12 h	2 h to 2.8 h
8 TB	14 h to 19 h	19 h to 25 h	19 h to 25 h	4 h to 5.6 h
12 TB	21 h to 28 h	28 h to 37 h	28 h to 37 h	6.1 h to 8.3 h
16 TB	28 h to 37 h	37 h to 2.1 days	37 h to 2.1 days	8.1 h to 11 h
20 TB	35 h to 46 h	46 h to 2.6 days	46 h to 2.6 days	10 h to 14 h
24 TB	42 h to 2.3 days	2.3 days to 3.1 days	2.3 days to 3.1 days	12 h to 17 h

Unraid facts this page uses

Each fact below is taken from Unraid's own documentation or release notes, re-found at the source on the date shown. Latest stable Unraid release when checked: 7.3.3.

Fact	Applies to	Source
What Unraid does by default
Latest stable Unraid OS release is 7.3.3 (docs release notes: 7.3.3 dated 2026-10-08).	Unraid OS stable, as of 2026-10-12 (7.4.0-rc.1 dated 2026-10-08 is a pre-release)	Unraid source, checked 2026-10-12
Scheduled parity checks (Settings > Scheduler > Parity Check) default to non-correcting mode.	Current docs (no version stated); default changed to non-correcting in the 6.11.2 release notes	Unraid source, checked 2026-10-06
After an unsafe shutdown Unraid automatically starts a parity check using the mode set under Settings > Scheduler > Parity Check (non-correcting by default); it is not automatically a correcting check.	Current docs (no version stated)	Unraid source, checked 2026-10-06
What the Unraid documentation advises
The docs ADVISE automated, correcting checks monthly or quarterly. This is advice, not a stated install default.	Current docs (no version stated)	Unraid source, checked 2026-10-06
How the array behaves
A parity check can be split into smaller pieces run over multiple days or weeks in set time windows (cumulative parity check). The source does not say whether progress survives a reboot.	Unraid 6.10.0+ (release notes 6.10.0)	Unraid source, checked 2026-10-06
A read check is needed when the array has no parity devices assigned, or more disabled drives than parity drives; it examines every sector of every disk in the array.	Current docs (no version stated)	Unraid source, checked 2026-10-14

A parity check is not a backup

Parity lets the array rebuild a missing disk. It is not a copy of your data, and it does not protect against deletion, ransomware, theft, fire or more disk failures than your parity covers. See our 3-2-1 backup strategy guide, and our Unraid boot drive failure guide for a real recovery where existing parity is what made recovery possible.

Frequently Asked Questions

How long does an Unraid parity check take?

About the size of your parity disk divided by the average speed of the check. A check reads every disk at the same time, so the slowest disk, the disk controller and anything else using the array set the pace. At an average of 120 to 160 MB/s an 8 TB parity disk takes about 14 h to 19 h. If you know the average speed of a past check on your own server, enter it: it beats any assumed range.

Should a scheduled parity check be correcting or non-correcting?

A non-correcting check reports mismatches and changes nothing. A correcting check also rewrites parity wherever it finds one. A mismatch shows that data and parity disagree, not which side is wrong, so correcting blindly can write a bad value over a good one. Scheduled parity checks (Settings > Scheduler > Parity Check) default to non-correcting mode. The docs ADVISE automated, correcting checks monthly or quarterly. This is advice, not a stated install default. Whichever you schedule, find the cause of any errors before you correct them.

How often should I run a parity check?

The docs ADVISE automated, correcting checks monthly or quarterly. This is advice, not a stated install default. We could not confirm from Unraid's documentation whether a new install schedules checks on its own, so look at the scheduler settings on your own server. A regular check finds unreadable sectors and mismatches while parity can still fix them, rather than during a rebuild.

What happens after an unclean shutdown?

After an unsafe shutdown Unraid automatically starts a parity check using the mode set under Settings > Scheduler > Parity Check (non-correcting by default); it is not automatically a correcting check. If a check after a power cut or crash finds errors, find the cause before correcting anything: errors can come from the shutdown itself, but also from a failing disk, a loose cable, the power supply or a crash. Look at the disks' SMART reports and the system log first.

What does a parity check with zero errors prove?

That parity matched the data and every disk could be read during that check. It is strong evidence that parity was consistent at the time. It does not guarantee a future rebuild: a disk can fail or develop unreadable sectors later, and a parity check does not verify file contents against checksums.

Does a parity check detect bit rot?

No. A parity check compares parity with the data. It cannot tell which side is right and does not check files against checksums. File checksums, for example on a Btrfs or ZFS pool or from a file integrity tool, are a separate check, and a pool scrub is not the same thing as an array parity check.

Can I split a long parity check across several nights?

A parity check can be split into smaller pieces run over multiple days or weeks in set time windows (cumulative parity check). The source does not say whether progress survives a reboot. The calculator shows how many quiet windows of your length the check needs. Splitting it changes the calendar time, not the active time: the check still has to read every disk once.

Does dual parity make a parity check take twice as long?

Not by any fixed amount we could source. The check still reads every disk once; a second parity disk adds calculation and one more disk to read. This calculator applies no multiplier for dual parity, so use the speed of your own past checks if you have it.

Method last reviewed: 2026-10-14. Unraid facts last checked at source: 2026-10-14.