Hi all,

I've been trying to get my HighPoint 1820A raidcontroller to work in Slackware (12.0, kernel 2.6).

Although it seems to be supported by default, I want to use the drivers provided by HighPoint. Using those, I can use their RAID-management utility to configure/monitor my arrays.

However, trying to get these things to work has been proven difficult.

I compiled the 'opensource' drivers from HighPoint and installed them:
> make
> make install

I then probe for the module:
> modprobe hptmv

(Is it normal that I don't return to the CLI after this command? It just 'hangs' and I need to use another usersession (ALT+Fx))

After a few moments, the controller starts to beep. Logging in onto the RaidManagement utility, I checked the 'events' -> seems 1 drive is causing troubles.
I reboot, and go to the HighPoint BIOS: although I created a new RAID5 array yesterday, it seems to be broken already. Strange this is: all drives are mentioned in the list, all with the same array_name. Checking the arrays themselves, it turns out that I now have 2 arrays: 1 of 3 drives and another of 1 drive.

Shuffling around with cables and drives, I can't seem to pinpoint a certain faulty drive: if I don't boot to linux, the 'faulty' drive remains the same. However, when I boot to linux and create a new array, another drive will be marked as 'failed'.

In short: I believe it is a driver problem, causing 1 drive to be kicked from an array. Is it possible that some other SATA/SCSI driver/module causes these symptoms?

Has anyone else encountered these issues?
Could it be that the controller is broken? (it hasn't been used that much though).

PS: using smartmontools, I'm performing a (long) selftest on the 4 connected drives. Note that hptmv is not loaded at this moment, and that I can access all drives on the controller without problems. (they are currently not in an array).

I'm pretty sure it's the drivers now.

Once again I created a new array (4x400GB, RAID5). Yet again, immediately after creation, the controller starts beeping. Checking the events in de RaidManager (RM), it stated that the device on channel 1 had failed. I shut down the computer, removed the drive (and now attached it to my windows machine to check it once again), booted the computer again and created a new array with the 3 remaining disks, using the RM.
Again, right after completion (actually: this time the RM returned a "create array error", outcome is the same), controllers starts beeping. CHecked the events (in the RM) and noticed that this time 2 other drives 'had failed'.


Grmbl..

Maybe I should try another distro (debian?) using the distro-specific drivers on the HighPoint website?

So, for testing purposes I installed the controller and the drives on my Windows machine. It seems that this is running as it should, confirming that it's probably a linux driver problem.

However, in the mean time, I've installed Debian 4.0 on the server. First time I actually work with Debian so it demands some adjusting, but I suppose I'll be ok. (was a bit disappointed by the Debian Installer though, Slackware allows more detailed adjusting of the installation).

Next thing to do is to install the controller in de linux box again and check the debian drivers. Fingers crossed!

Debian seems to be a no go..

Apparently, the Debian drivers are compiled for another kernel. In that case, I'm referred to the 'opensource' drivers (I used with Slackware).

1. Suppose the outcome will be the same
2. make troubles me - I guess I'm missing some files to actually compile the opensource drivers.

So, intermediate update:

installed SlackWare 12.0 again.
Compiled the drivers and the management tools.



Interesting: after loading the hptmv module, it takes a few moments (1-2 minutes) and then the controller starts to beep anyway. (so it seems I don't even need to try to create an array).

Why does this module think those drives are failing??


PS: everything worked flawlessly under windows.

Moved: This thread is more suitable in Slackware and has been moved accordingly to help your thread/question get the exposure it deserves.

Mod Note: moved per OP's request

Did anyone else just use the built-in drivers (Slack 12.0)? Did you encounter any problems?

I've been trying Software RAID the last few days, but I just keep running into problems there too..

I wonder if it may be the HighPoint drivers in the kernel that cause the system to crash.. (btw: kernel drivers mention they are for 3xxx controllers??)

FYI: I was able to get everything to work without problems under Ubuntu Hardy Heron Server Edition (8.04). Note that I don't use the HighPoint drivers (the 1820A works without problems with the default kernel) as I don't have time to experiment (+ I can't backup my 'old' software raid5 arrays ).

Maybe I'll try Slack again when newer versions are released.

The symptoms you describe by what I've seen are usually an indication that the sata_mv module is active when you try to load the hptmv module. If so, this will cause big problems. You need to remove the sata_mv module and blacklist it before you try loading the hptmv module.

If I recall correctly, I actually tried that (read about it somewhere else on the internet).


Next time I make an attempt (maybe somewhere in july/august, won't have time before then) I'll take this into consideration.

Thanks