Adaptec 71605 Corrupt data on Raw drives, UEFI, GPT

[ulicky]

Hi,

I found out, that that UEFI firmware (BIOS firmware is ok) is duplicating GPT table to middle of drive, where regular data is and thus making corruption.
It happens during boot process. Problem is in firmware, it is OS (Windows/Linux tested) independent.
On 10TB drive affected drive sector, where GPT is written is 2 353 004 511, on 8 TB affected drive sector is 2 743 151 247.
Look at this it is GPT table. This should not be in the middle of drive (only beginning and end).

You can read this thread, where I discovered that this bug is real and reproducible SnapRAID / Discussion / Help: Strange file corruption

Can someone help me replicate problem on same card?

Thanks

 

[TubaMT]

This is super interesting. So from reading that forum post you linked, the Adaptec 71605 cards have issues and causes data corruption with 8 and 10TB HDDs when booting with UEFI? But if you use BIOS then it seems to work fine. I was about to purchase a bunch of the cheap Adapter 71605 cards available on ebay, but now I don't know.

I'm wondering if anyone else has these issues. I haven't heard anything regarding this except for that linked discussion and this post.

 

[kiteboarder]

Hi all,

I have qty 4 of 10 TB individual NTFS drives on this controller. They're nearly full with backup data from other machines. I do not yet know if there is lurking corruption, as I have not yet figured out how to locate which specific file would be at the specific corruption address. And I REALLY don't want to run a "file compare" on 40TB of data. Honestly just haven't gotten around to dealing with this yet.

If anyone knows how to figure out which specific file would be at that address, I'm all ears.

Thanks!

 

[Firebug24k]

I just got a few of these and will do some testing with a ZFS array of 8, 10, 12TB drives over the next week. Plan is to copy over 50TB or so of data and scrub and see what happens. Anyone got any other smarter testing methodology they want me to check, let me know. I've got 300TB of data on ZFS arrays running with LSI cards at the moment that operate perfectly so I'm starting from a 'known good' condition.

 

[thy.flesh.consumed]

I just got a few of these and will do some testing with a ZFS array of 8, 10, 12TB drives over the next week. Plan is to copy over 50TB or so of data and scrub and see what happens. Anyone got any other smarter testing methodology they want me to check, let me know. I've got 300TB of data on ZFS arrays running with LSI cards at the moment that operate perfectly so I'm starting from a 'known good' condition.

Please do report back. I did my own testing here https://forums.servethehome.com/ind...0-x8-16-port-sas-controller.28836/post-276001 and I couldn't find anything wrong. I'm planning on putting this card into service sometime next month and it would be nice to get another confirmation that the card is solid and can be trusted with the data.