BlueSCSI in Quadra annihilating system folder, Daynaport

[Arbystpossum]

I'm not sure where to post this, I'll start here and it can be shoved off elsewhere if needed.

I have a Quadra 605 with an internal BLueSCSI v2. Every time I try to install the DaynaPORT software, it'll get 60% through the install then error out. The error says that the changed will be reverted, but it instead does something to the system folder that kills it. By kills it I mean, when I exit the DaynaPORT installer, there ar enough desktop icons, the top menu bar is blank and the system can no longer boot if reset. I saw a video from Mac84, he used a newer version of the software that said you need extensions turned off. I tried again with extensions off, and it still destroys the system. I tried installing the drivers in a Basilisk2 instance but it doesn't detect the hardware and refuses to install. The only thing I can think to do is try flashing the firmware again, I flashed it once already with the latest before attempting anything. Drive containers work just fine too.

 

[eric]

it'll get 60% through the install then error out

What error?

the system can no longer boot

What system?

Also please make sure you go through https://github.com/BlueSCSI/BlueSCSI-v2/wiki/Troubleshooting#what-to-try-first first when trouble shooting - rules out easy issues first.

 

[Arbystpossum]

I'm not in, so I'll have look again. I'm digging through the troubleshooting tips now, though my issue seems unique.

What error?

It's a generic "Can not complete installation" error. I forget the exact words, but it doesn't imply the hardware isn't present, or imply much other than it failed.

What system?

I've tried this on MacOS System 7.1.2, 7.5.3 and 8.1, it does the same thing every time. Hard drive images are set to SCSI ID 1,2 and 3, DaynaPORT is NE4.hda

 

[eric]

Are you using the HDA provided with the drivers? Do you have open transport or MacTCP installed? I haven't tried MacOS 8 but I believe any 7.x will work.

 

[Arbystpossum]

Are you using the HDA provided with the drivers? Do you have open transport or MacTCP installed? I haven't tried MacOS 8 but I believe any 7.x will work.

I have used the .hda with the driver bundle and a newer driver from Macintosh Garden, the .hda one says it's from 1993 and the later is from 1997, both do the same to the system. 7.1 had MacTCP, 7.5 and 8.1 had Open Transport.

 

[Arbystpossum]

Update to this, it appears as if the Bluescsi 2 is corrupting data. I setup the Bluescsi 2 with only NE4. Hda and the .ini file telling it how to connect to my network. On the backside I have a bluescsi 1 externally, I did the OS and software installs to the bluescsi 1 without any issue. It didn't stop me from installing the Daynaport software, so I know the system saw it as Scsi ID4. Following the steps, I setup wifi accordingly, but I am unable to connect to the internet.

As a test, with a fresh 7.5, I tried installing other programs, which all also failed on the Bluescsi 2. Any folder or file the installer touches, it corrupts. Trying to open the damaged system folder in Basilisk2 game gives me a -199 error.

I tried flash_nuke.uf2, then loaded the latest release and it's still corrupting files.

 

[Arbystpossum]

I rolled the firmware back to v2023.10.12, which was the first to mention/support the Daynaport SCSI ethernet. The data corruptions seems to have stopped, but the ethernet portion doesn't seem to be working still. I turned debug mode on, recommendation of Mac84. I have a few things to try yet, full format of the SD card, hopefully the debug log can tell me something I understand. I know I'm kinda talking to myself, but hopefully this will be documentation of a strange issue and the way to solve it.

 

[Arbystpossum]

I hate when you fix something and you don't know what you did. Drives me bonkers.

So, the only thing I know of that made a difference with the BlueSCSI 2 wifi working finally, is I changed the network .hda file from NE4.hda to NE5.hda, no other changes. Upon rolling back the update and changing what SCSI ID the BlueSCSI was using, the file corruptions have stopped and I can connect out.

tl;dr, If your BlueSCSI is corrupting files and the wifi doesn't work, roll back the firmware and change the SCSI ID for the Daynaport file.

Case closed?

 

[Arbystpossum]

Had more data corruption again. This rollercoaster is killing me. I rolled the frimware back to before Wifi was supported, the system seems to be running faster, successfully installed 4 programs and rebooted 2 times without any data corruption.

 

[eric]

Do you see any errors in the log? Can you turn on debugging and recreate the issue (and upload any log.txt or err.log)? We've seen 0 perf difference with wifi or not (since the wifi stuff doesn't touch any of the normal scsi communication anyways, but we did test) and being on a different ID seems like a red herring as it would make no difference.

 

[Arbystpossum]

I'll flash the latest firmware and setup the conditions to hopefully re-create these issues. As soon as something goes sour, I'll post the log here. I see things in it, but I do not understand what they mean. In a hurry, I took 2 pictures of the log, to see if I could stare at them long enough to give me a clue as to what's going on. The first is the beginning where it finds images, identifies them and looks fine, the other is a small part of a huge chain, whenever I try to unsuccessfully try to use the DaynaPORT/wifi function. It's probably not helpful, more/better to come.

 

[telcoserf]

I'm seeing similar behaviour with mine, on a similar system -- an LC 475, which is essentially the same as a Quadra 605. Will go down the troubleshooting rabbit hole now, but I thought I'd add another data point to this, in case it's affecting others. It seems like the moment I install the Daynaport driver, the entire SD card is corrupted and the only way to get back to a working state is to format the card and copy the hda image back over to it and try again.

 

[Arbystpossum]

I have no idea what I did right, but I kept messing with it, blasted the firmware out several times and it just stopped being buggy. I think I tried the latest firmware, then tried the first firmware that supported the wifi feature. I'm not sure what firmware it's on now. The network feature even works, though, any network access seems to make the computer freeze for half a second every 3 seconds. I'm not sure what that's about.

Re-flash the firmware, and then again. I don't know what I did, but I hope you shock-start the thing, because it's awesome when it finally behaves.

 

[hanov3r]

I've run into this issue on a Powerbook 540c. Using the OpenRetroSCSI-7.5.3 image from the mega.nz link in the BlueSCSI docs along with the .hda from the DaynPORT documentation page, I can consistently corrupt the boot image and generate a discordant chime and sad Mac `0000 0007` error screen on reboot. I have not yet tried reflashing the firmware or renaming the NE4.hda file, so I'll give those a try.

 

[Mk.558]

This is also similar to what I've gotten on a borrowed Quadra 650. The BlueSCSI will work for about 5-10 minutes, then nuke itself. The only way out is to reformat the whole SD card all over again. Haven't had an issue with my IIci, the SE/30 is down pretty hard so I can't test it. Is it a 68040 problem?

 

[davewongillies]

I've had a similar issue with BlueSCSI on my Quadra 700. If I copy lots of files into disk while booted into System 7 it'll eventually completely trash the entire SD card. The same thing happens if I do an os install too.