Initiator / USB MSC mode mode crashes Pico

[danosuke]

Though my BlueScsi functions great in target mode when set to USB MSC mode target drive is detected on bus with ID#. As soon as windows tries to pass commands to the drive Pico crashes. I have tried 3 different SCSI drives and this behavior is the same with all of them.

SEAGATE ST318418N capacity 4340 MB
SEAGATE ST318417N capacity 4340 MB
IBM DDRS-34560 capacity 4357 MB (this one sounds like a table saw but works )

SEAGATE _ST318418N_BlueSCSI.log

Pico is receiving power from USB and PwR_ON jumper is off. Drives are set to receive Termination Power from themselves and are terminating the bus. Back Feed is off. Two Initiator Jumpers are on. Bluescsi.ini is set as recommended in Docs for MSC mode.

After the crash it will reboot itself. If I power cycle the drive Pico recognizes it again and crashes again. Tried two different Windows Machines. I am stumped.

[erichelgeson]

The log looks odd, there are just some characters/lines missing. Is there an err.log?

Lets do a few easy things - try a different USB cable, look for an err.log, format the SD Card with a full overwrite format with sd card formatter in the docs (maybe weird log corruption due to sd card?), and make sure nothing is in bluescsi.ini besides the few lines you need for this mode.

[danosuke]

Sorry for the delay. I attached the console log in the original message request. Here is the error log file.
err.txt

[erichelgeson]

Thanks for the log, this is a very odd error - the drive presumably(according to the log) is not sending back the correct amount of data. Can you enable debug logging too so we can see the exact command it's sending and the phase it's in? I suspect it may be the drive is changing phases quickly at the end before we're able to read the last 4 bytes (but this is just a guess).

Also does the BlueSCSI work fine otherwise? Do these drives work fine otherwise? Want to rule out a hardware issue.

[tomfeather]

I hope I'm not confusing things but I think I'm seeing the same issue - the Pico resets at the point where the BlueSCSI tries to communicate with the drive. I have the same issue with two different Conner drives, a CP 30100 and a CP 3040A.

I was using the latest firmware - 2025-10-27. I then tried downgrading the the previous version, 2025-08-02 and the BlueSCSI took an image of both drives with no problems. I think there may be something in the latest firmware that is causing this issue?

I have attached one of my error log files with debug mode on - err.txt

Thanks.

[erichelgeson]

Thats a great hint for debugging - @danosuke can you try the 08.02 release? If you get the same result I can bisect to see what the issue is.

[danosuke]

I will test ASAP.

…

On Mon, Dec 1, 2025 at 11:36 AM Eric Helgeson ***@***.***> wrote: Thats a great hint for debugging - @danosuke <https://github.com/danosuke> can you try the 08.02 release? If you get the same result I can bisect to see what the issue is. — Reply to this email directly, view it on GitHub <#306 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/B2NYJUG4OWFB6YHABWHLMHT37RU7LAVCNFSM6AAAAACNMSORRWVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTKMJSGYZTGOA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[erichelgeson]

We've identified the issue, will be fixed in next release! For now please use the prior release to do initiator.

[danosuke]

Finally extracted my BlueScsi from the Mac Classic it was entombed in. I tested with 08.02 release and behavior in USB Bridge mode/MSC is the same. Still see's the drive on the bus and crashes immediately. I also tested Initiator mode to automatically create a disk image and it works perfectly. I made images of all four of the SCSI drives I had tested MSC mode on. Is there a way to write HDA images back to a disk? that would be amazing.