Adaptec ASR-7805 and 7805Q Data Recovery: Failures, Error Messages, and maxCache SSD Risks

The Adaptec ASR-7805 is the eight-port member of Adaptec’s Series 7 family: a low-profile card with two mini-SAS HD connectors, a gigabyte of cache, and the same PMC-Sierra PM8015 RAID-on-chip as its sixteen- and twenty-four-port siblings. It went into small business servers, eight-bay towers, and workstation RAID 5 and RAID 6 volumes, and — because it can be switched into HBA mode without reflashing and sells secondhand for very little — it now sits in a great many home labs and budget NAS builds. The 7805Q adds a pre-installed flash backup module and maxCache 3.0 SSD caching, which changes what “the array” physically consists of when something goes wrong.

This page covers what is specific to the 7805 and 7805Q. The complete catalog of Series 7 messages — POST screens, arcconf states, maxView events, Linux aacraid and Windows arcsas log entries, each rated by risk to the data — is on our Adaptec Series 7 recovery page.

Start a free ASR-7805 recovery evaluation →  or call 1-877-624-7206

7805 and 7805Q: What Is on the Card

ASR-7805 ASR-7805Q
Internal ports 8 (2 × SFF-8643) 8 (2 × SFF-8643)
Cache 1 GB DDR3 1 GB DDR3
Flash cache protection AFM-700 optional AFM-700 pre-installed
SSD caching No maxCache 3.0 (up to 8 SSDs, 2 TB)
RAID levels 0, 1, 1E, 5, 6, 10, 50, 60; HBA and Simple Volume modes; up to 256 devices through expanders
Form factor Low-profile MD2, 64 × 167 mm; PCIe 3.0 x8; PMC-Sierra PM8015 RAID-on-chip
Power and thermal 1.5 A at 12 V plus 0.1 A at 3.3 V (datasheet); 0–55 °C, or 0–50 °C with the AFM-700, with 200 linear feet per minute of airflow measured one inch from the RAID processor

Adaptec’s part number for the retail 7805 is 2274100-R; the box included two 0.8-metre SFF-8643 cables and a low-profile bracket. The distinction that matters for recovery is the Q: a 7805Q may be holding writes on its SSD cache pool that have not reached the hard drives, and it always has a flash module whose state is worth recording. A plain 7805 may or may not have an AFM-700 fitted, and secondhand cards frequently arrive without one.

Diagram of the Adaptec ASR-7805's two mini-SAS HD connectors, each carrying four drive lanes (Connector 0 to Devices 0-3, Connector 1 to Devices 4-7), showing how one cable or connector fault removes half of an eight-drive array

Two Connectors, Eight Drives: When Half the Array Disappears at Once

The 7805’s eight ports are grouped into two mini-SAS HD connectors, each carrying four drive lanes when the drives are direct-attached. arcconf reports members by that grouping — Connector:0, Device:0 and so on — and the grouping is the single most useful diagnostic fact about the card. If exactly four drives drop out at the same instant, and all four sit on the same connector, the cause is almost never four simultaneous drive failures. It is the cable, the backplane segment, or the connector itself.

One 7805 owner documented the pattern on a brand-new build: four SATA SSDs on one connector, only one of them (device 3) ever recognized, through cable swaps, port swaps, reseating, rescans, and a switch to HBA mode — while all four drives worked normally on the motherboard. The conclusion in that thread was a bad port on the card. On a card with only two connectors, a fault like that takes out half the array. On an eight-drive RAID 5 or RAID 6, that is an instantly failed volume with every drive in it perfectly healthy.

That array is recoverable, provided nobody “fixes” it. The three destructive moves are initializing the drives that dropped, forcing the array online after members have come and gone at different moments, and letting the operating system see raw members. Before anything else, record which connector and which bay every drive occupies. Series 7 arrays identify members by metadata rather than by port, but the physical map establishes what happened and when.

Heat on a Low-Profile Card

Adaptec’s specification for the 7805 is an operating range of 0–55 °C with 200 linear feet per minute of airflow, measured one inch from the RAID processor, and 0–50 °C when the AFM-700 is installed. The datasheet is explicit that the card must be installed only in a chassis that delivers that airflow. The card draws about 1.5 A on the 12 V rail, so the heatsink is dissipating something on the order of eighteen watts continuously in a passive design built for a rack server’s front-to-back fan wall.

Home-lab owners who put Series 7 cards in tower cases report the consequences directly. One widely read overview thread describes a Series 7 card sounding its over-temperature alarm within five minutes of boot with completely idle drives, notes that the 200 LFM requirement is printed on the card itself, and describes the stock thermal compound on secondhand cards as dried out. Adaptec also revised the Series 7 heatsink during production — from a threaded-hole mount to a clip-mounted silver heatsink identical to the Series 8 design, with the same fan kit fitting both — so two physically different coolers exist on 7805s in circulation, and neither ships with a fan.

What overheating looks like before the array fails

  • Intermittent aacraid: Host adapter abort request and Host bus reset request. SCSI hang ? lines on Linux, or Event ID 129 resets from the arcsas source on Windows, especially under sustained load.
  • Members dropping out and returning, leaving “degraded” arrays whose drives all test healthy.
  • On cards with an AFM-700, “Write-through (temporary disabled)” because the supercapacitor’s over-temperature protection has disabled the backup unit.
  • Eventually a card that hangs at “Booting the Controller Kernel” or does not complete POST.

If the array is still online, copy the important data off before improving the cooling. If the card has stopped posting, the drives are almost always intact and the configuration is on them; the risk is what happens next with a replacement card.

The Secondhand 7805: “Mode switch failed”

The 7805 is popular secondhand precisely because its mode is a setting, not a firmware flash: RAID: Expose RAW, RAID: Hide RAW, HBA, Auto Volume, or Simple Volume. The problem is that a used card arrives in whatever mode — and with whatever array metadata — its last owner left behind.

One new owner, who had never configured anything on the card, tried to switch it to HBA mode and was refused with the message mode switch failed, please delete all arrays and uninitialize disks and try again. The card was protecting arrays defined by a previous owner, on drives that were no longer attached. Adaptec’s guidance is that HBA mode should be selected only when no drives carrying Adaptec metadata are attached; the card enforces the spirit of that rule by refusing to abandon a configuration it still believes it manages.

For someone setting up a new card, the message is an annoyance. For someone trying to get a failed array back, it is a warning: the “delete all arrays and uninitialize disks” the card is asking for is exactly the operation that destroys recoverable data. The metadata the card will not let go of is what a recovery engineer needs.

The reverse trap is the one documented on a sibling 71605: a used card believed to be in HBA mode was actually in Simple Volume mode, and existing data drives connected to it were no longer mountable after the first boot. Check and set the controller mode with no data drives attached — and on any used Series 7 card, expect leftover metadata.

Installing a used 7805 without risking data

  1. Install the card with no drives attached. Boot into the Ctrl-A utility or UEFI setup and confirm the controller mode.
  2. If the card reports arrays or initialized drives you did not create, that is a previous owner’s configuration. It is harmless while no drives are attached.
  3. Check the firmware build (ARCCONF GETVERSION). See firmware below.
  4. Confirm cooling before loading it with drives.
  5. Connect one expendable drive first, confirm how the card presents it, and only then connect drives carrying data.

The 7805Q: maxCache Makes the SSDs Part of the Array

The 7805Q adds maxCache 3.0, which uses up to eight SSDs (2 TB aggregate) as a read cache and, optionally, a write cache in front of the hard-drive arrays. Adaptec’s rules for that write cache are strict: it can be enabled only when the maxCache container itself is redundant (RAID 1, 1E, or 5), and Adaptec’s own configuration guide warns that a failed cache SSD can cause data loss for every array that uses SSD write caching. The container cannot be deleted until maxCache read and write caching have been disabled on every logical drive that uses it.

Owners discover how deeply maxCache is wired into the card when they try to repurpose one. One 7805Q owner deleted all three volumes on his eight-drive server and un-initialized six of the eight drives, but two remained greyed out in the Ctrl-A utility and could not be un-initialized, which blocked the switch to HBA mode. The two were the maxCache SSDs. maxCache is not configurable from the card’s BIOS at all; he had to boot Adaptec’s support environment and disable maxCache from maxView before the card would let go of them. Four years later another owner hit the identical wall.

For recovery, the same fact cuts the other way. If a 7805Q array fails with write caching enabled, the SSDs may hold the most recent writes to the file system. Ship the maxCache SSDs with the hard drives, labelled by connector and bay. Do not delete the maxCache container, remove SSDs from the pool, or un-initialize them to get the hard-drive array back online — each of those actions can discard cached writes that have not reached the platters.

Firmware on the 7805

The 7805 shares the Series 7 firmware line (7.5.0) with the rest of the family. The builds that matter:

Build What 7805 owners should know
32033 (August 2014) Introduced Simple Volume mode. A storage integrator advised customers off this build, citing several drives reported degraded at once when none were defective, and volumes disappearing from the operating system.
32084 The build integrators recommended at the time.
32106 (2016) Reported on Series 7 cards to leave write cache stuck at “Write-through (temporary disabled)” regardless of the flash module’s condition; the reported workaround was a rollback to 32084.
32114 (2017) Fixed RAID 5 and RAID 6 logical drives that would not rebuild, a firmware stack overflow triggered by a faulty drive, write cache that could not be enabled without a flash module, and “Error: Controller Kernel Stopped Running” at POST.
32118 (2018) The last public Series 7 build.
7.6-0 (32136) Found on secondhand Series 7 cards and not a public Adaptec release; owners report abort errors on it and report rolling back to 32118 without bricking the card.

None of this should be applied to a 7805 that currently holds a degraded or failed array. The fixes in 32114 are real, but a firmware flash is the wrong moment to find out the card is unstable, and an HDA-mode recovery flash with members attached is itself a risk to the data.

Replacing a Dead 7805

When a 7805 fails, the obvious replacement is a Series 8 card — an 8805 has the same port count and a faster interface. It will not read the arrays. Microchip’s support position, quoted by a user who asked, is that arrays cannot be imported from previous-generation Adaptec SAS RAID adapters to the newer SmartRAID adapters. An administrator who made the equivalent move from a 71605 to an 81605ZQ documented the result in detail: every drive shown as Ready rather than Foreign in arcconf getconfig 1 pd, nothing under Manage Arrays, no logical drives in Windows or maxView, and no import or “scan for foreign configuration” option anywhere. arcconf has no command to mark a drive foreign, and maxView cannot see an array the firmware does not detect.

The drives in that situation are intact. Three things destroy them: creating a new array on the new card to “recreate” the layout, initializing the Ready drives, and letting Windows or a Linux installer initialize or partition the raw members that Expose RAW passes straight through.

A working Series 7 card with compatible firmware will normally read the arrays: power down, move the drives, confirm the card is in RAID mode, and do not accept any configuration-change prompt at POST until every member is present. If no Series 7 card is available, or if the 7805 failed mid-rebuild or after members had already dropped, the arrays should be reconstructed offline from drive images. The AFM-700 does not migrate between Series 7 and Series 8 either, so a Series 8 replacement also means a new flash module.

Operating-System Notes for 7805 Hosts

  • Linux SCSI timeout. Adaptec’s Series 7 readme recommends raising the SCSI command timeout from 30 to 90 seconds; otherwise the system may hang with an EXT4 error when a good drive is removed from a rebuilding array.
  • Linux boot record. The boot record is always written to Device 0. Deleting or reordering arrays so that Device 0 changes makes Linux fail to boot; restoring the original array sequence brings it back. A “missing” boot volume after array housekeeping is often this, not data loss.
  • Windows sleep and hibernation. The Windows inbox driver for Series 7 cards was reported in 2022 to leave RAID volumes hanging at 100 % load, timing out every access, and occasionally blue-screening after wake from standby or hibernation on Windows 11 22H2 and Server 2022; the reports named the 71605 and 72405 explicitly, and a 2023 comment showed it still present. On any Series 7 host, disable sleep and hibernation.
  • Shucked drives. Owners of Series 7 cards report drives pulled from external enclosures showing up as two of twelve until the SATA 3.3 V power-disable pin is blocked. That is a power issue, not a controller failure — but it looks like mass drive failure at POST.

7805 Messages: Quick Index

The messages 7805 owners most often search for, each linked to its full explanation and risk rating on our Series 7 page:

If Your 7805 Array Is Down Right Now

  1. If the volume is still accessible, copy the most important data off it before any rebuild, cooling work, or firmware change.
  2. Capture ARCCONF GETCONFIG 1 AL and ARCCONF GETLOGS output, or photograph the POST and Ctrl-A screens, before changing anything.
  3. Label every drive with its bay and its connector (0 or 1), and note serial numbers.
  4. Do not change the controller mode, un-initialize drives, flash firmware, or move the drives to a Series 8 or SmartRAID card.
  5. Do not connect member drives to a Windows or Linux computer.
  6. On a 7805Q, keep the maxCache SSDs with the hard drives and do not delete the maxCache container.

How We Recover 7805 Arrays

We image every member — hard drives and any maxCache SSDs — on write-blocked hardware in our ISO 5 Class 100 cleanroom, replacing failed components in physically damaged drives with donor parts first. Our HOMBRE reconstruction software reads the Series 7 metadata from each image, establishes member order, stripe size, and parity layout, and determines which members dropped out and when. The array is assembled virtually — no 7805, no replacement controller, no firmware involved — and the file systems are recovered from the assembled volume. The full Series 7 approach is described on our Adaptec Series 7 page; the sixteen-port sibling has its own ASR-71605 page, and other Adaptec generations are covered on our Adaptec RAID controller recovery page and RAID data recovery hub.

ASR-7805 Documentation

ASR-7805 Frequently Asked Questions

Four drives vanished from my eight-drive array at the same moment. Are they dead?

Almost certainly not. Four drives share each of the 7805’s two connectors, so four simultaneous drops point to a cable, backplane, or connector fault — one owner traced exactly this to a bad port on the card. Do not initialize the drives or force the array online; record the layout and power down.

My used 7805 says “mode switch failed, please delete all arrays and uninitialize disks.” I never made an array.

A previous owner did, and the card still holds that configuration. With no data drives attached, clearing it is harmless. If you are seeing this while trying to recover your own array, stop: the operation the card is asking for destroys the metadata a recovery depends on.

Two drives on my 7805Q are greyed out and cannot be un-initialized.

They are almost certainly the maxCache SSDs. maxCache is managed from maxView or arcconf, not from the Ctrl-A BIOS. If the hard-drive array holding your data has failed, leave the SSDs exactly as they are and send them with the drives.

Will an Adaptec 8805 or a SmartRAID card read my 7805 arrays?

No. The drives appear as Ready with no import option, and Microchip’s support position is that arrays from earlier Adaptec generations cannot be imported to SmartRAID adapters. Use another Series 7 card, or have the arrays reconstructed from images.

Does the 7805 need a fan?

In a rack server with the specified 200 LFM of airflow, no. In a tower or a small NAS case, owners consistently report that it does — alarms within minutes of boot are described in community threads. Heat-related resets and dropped drives are an early warning; back up first, then fix the cooling.

What should I send you?

Every drive that was attached to the card, labelled by bay and connector — including the maxCache SSDs on a 7805Q. We do not need the server, and in most cases we do not need the controller.

Start Your ASR-7805 Recovery

Power the system down, label the drives, and do not initialize, un-initialize, force, recreate, or migrate anything. Our engineers will image every member, reconstruct the array without the controller, and give you a firm quote before any recovery work begins. Gillware has been recovering RAID arrays for more than two decades.

Start a free ASR-7805 recovery evaluation →

Or call 1-877-624-7206 during business hours (M–F 8 am–7 pm, Sat 10 am–3 pm Central), or schedule a 15-minute consultation.

877-624-7206 Free evaluation