RDD / Whatever it is saying now / Punctured stripe on a Dell PERC

PERC · puncture · double fault · bad blocks in virtual disk · rebuild with errors

A punctured stripe on a Dell PERC. The array says Optimal. It has a hole in it, and it knows.

A puncture is what a Dell PERC does when a rebuild meets a sector it cannot read on a surviving member. Rather than fail the virtual disk, the controller marks that stripe as bad on every member, finishes the rebuild around it, and reports the array Optimal with a note about bad blocks. Dell's own description is a feature designed to allow the controller to restore the redundancy of the array despite the loss of data caused by a double fault; Dell's own note is that all punctures are double faults but not all double faults are punctures. The data in the punctured stripes is gone, and the only cure Dell offers is to delete the virtual disk, recreate it and restore from backup. Before that, the set should be imaged, because what the puncture took is usually small and what the recreate takes is everything. One member imaged is £300 + VAT after the free look; a punctured set imaged and read is £500 + VAT upwards.

Free first lookOne fixed figure in writingNo data, no bill on most jobsReturn postage paid

Rather talk it through? An engineer answers the bench line
0800 6890668

Power down. Label every slot before a drive comes out. Do not rebuild while a second drive is flagged. Do not clear or import a foreign configuration. Do not initialise, reformat or PSID-revert anything. A rebuild reads every sector of every surviving member, and on a set whose drives were bought together it is where the second failure is found. Nothing on the drives gets worse while the server is off.

How a puncture happens, and what the bench can still find.

A rebuild computes the missing member's contents from every survivor, stripe by stripe. When a survivor answers a read with an unrecoverable error, the controller has two choices: fail the virtual disk, or skip. A PERC skips. It marks the stripe as bad in its own table on every member, writes nothing meaningful for that stripe on the new drive, and carries on. The rebuild completes, the virtual disk goes Optimal, and the log records a rebuild with errors and a count of bad blocks.

From then on, any read of a punctured stripe returns an error to the operating system, which sees it as a bad sector on the volume. Depending on where the stripe fell, that is an unreadable file, a corrupt database page, or, if it fell on file-system metadata, a volume that will not mount. The controller will not clear the puncture by itself, and Dell's guidance is to back up what can be read, delete the virtual disk, recreate it, and restore. The recreate initialises the members.

The bench's interest is in two things. The first is what is still readable on the set as it stands, which is usually nearly all of it, copied out from images of the members before anyone recreates anything. The second is the drive that was replaced: the failed member the rebuild was replacing still holds the stripes the survivor could not give, and if it was kept, its image can fill the holes. That is why the failed drive should not leave with the engineer until the rebuild has been checked for errors.

What it says, and what it means.

Describe yours to us →

What you see	The usual reason	Where that leaves you
Rebuild completed with errors; bad blocks: n	A puncture	Image the set before recreating anything
Files unreadable on an Optimal virtual disk	Punctured stripes under those files	Copied out from images; holes filled from the failed drive if kept
Volume will not mount after a rebuild	A puncture on file-system metadata	Metadata repaired on the virtual volume from images
Bad blocks count rising on Optimal array	Punctures being hit during use	Stop; image now
Virtual disk recreated; data gone	The members were initialised	Assessed honestly; usually little

From the parcel arriving to your files going back.

Work we have closed →

Logged the day it lands, and the first look costs nothing Free

A case number goes on the parcel and a number on every drive the day it is opened, matched to the slot you wrote on it. An engineer settles what has actually happened before anything spins: the interface, the sector format, the firmware, the security state, and what the controller was saying when it stopped. Back to you come two things together: a straight note of what is liftable and what is not, plus one figure, fixed and written down. Accept it, or decline and owe us nothing.

Nothing to pay for lookingA single figure, put in writingNo rebuilds, no imports, no resets

The drive on its own bench

A punctured set is imaged member by member, and the controller's bad-block table read from the metadata so that the punctured stripes are known before the volume is opened. Where the drive that failed before the rebuild was kept, it is imaged too, and its stripes used to fill the holes.

Read at its native interfaceFailing heads to the clean bench

Imaged once, at its native sector size

Every drive is imaged sector by sector, head by head where it needs it, with the bad areas last. Drives with 520- or 528-byte sectors are imaged as they are and the extra bytes stripped from the image, never the drive. Self-encrypting drives are unlocked on the imager with the key you supply. From then on the originals are not touched again.

Head by head, weak areas lastNative sector size preserved

The image, and the set

Where the set is healthy and one member was the job, the image goes back to you on a fresh drive, sector-identical, ready to rebuild from. Where the set is the job, every member is imaged and the array reassembled in software from the images: order, chunk size, parity rotation and the reshape point read from the drives' own metadata, the file system repaired on the virtual volume, never on the originals. The array work is covered in full on our RAID array site, and it happens on the same bench.

One member: a sector-identical image on fresh mediaA set: reassembled in software from the images

You see the file list before you pay

What was recovered is listed for you first, and only then does a bill exist. Approve the list and it is invoiced; turn it down and it is not — and where nothing has come back, most jobs carry no charge at all. Recovered data travels home on fresh media bought in for your job, with the postage at our end. Your case is not closed until you have opened the files on a machine of your own.

No charge until you accept the figureFresh media, supplied with the job3–4 days at the bench

From the bench

Do not recreate the virtual disk. The recreate initialises the members. Image first; recreate afterwards, on a set you have already copied.
Keep the failed drive. It holds the stripes the rebuild could not compute, and it is the only place they still exist.
Punctures are usually small. A few stripes on a multi-terabyte set, and the files under them; the rest copies out whole.
Punctures on metadata are the bad case, and even those are repaired on the virtual volume more often than not.

One job, followed all the way through.

UK · RDD-2026-0563JOB LOGGED ✓

An R740 in RAID 5 that rebuilt with 47 bad blocks after a drive was replaced, and a SQL database that would not open afterwards

The engineer had offered to recreate the virtual disk; the owner sent the set here first, with the failed drive he had kept. The five current members were imaged and the PERC's bad-block table read; the failed drive was imaged with its weak areas last and its stripes used to fill 44 of the 47 holes. The database was repaired on the virtual volume with three pages rebuilt from its log, and copied out whole.

99.99% of the volume recovered8 days here, and back by post

Illustrative example — replace with a genuine case

What helps, and what harms.

Do this much first

Keep the failed drive the rebuild replaced
Export the log with the bad-block count
Power down and label the slots
Send every member, including the failed one

What sets us back

Deleting and recreating the virtual disk
Running a consistency check on the punctured set
Letting the engineer take the failed drive
Copying the volume off through the punctures repeatedly

Questions answered before you commit.

What is a punctured stripe?

A stripe the PERC marked bad on every member after a read error during a rebuild, so that the rebuild could finish. The data in that stripe is lost; the array is otherwise intact.

Can punctured data be recovered?

Sometimes, from the drive the rebuild replaced, if it was kept. Otherwise the punctured stripes are gone and everything else is copied out from images.

Dell says delete and recreate the virtual disk. Should I?

After the set has been imaged and copied, yes. Before, no: the recreate initialises the members.

What does it cost?

£300 + VAT for one drive after the free look; a punctured set of two to four members from £500 + VAT; larger sets from £1,250 + VAT.

How long does it take?

5–10 days at the bench for a set.

Where this usually leads.

PERC virtual disk degraded or failed →Dell PowerEdge and PERC →Predictive failure, SMART and grown defects →

Nothing gets worse while it is powered down.

Looking at it is free. Back comes a list of what opened and what did not, together with a single price to finish, set down in writing while you are still free to say no. Until that list reaches you, leave the server off and the drives in their slots.

0800 6890668

Price my job Packing and posting

Every kind of member drive

Manufacturers

Whatever it is saying now

What it costs, and what follows