View Single Post
Old 3rd February 2012
jggimi's Avatar
jggimi jggimi is offline
More noise than signal
 
Join Date: May 2008
Location: USA
Posts: 7,977
Default

I'm sure that copying the configuration file won't make any difference at all. That's strictly so you avoid confusing yourself, or others, when you (or they) look at your dmesg output.

I see raidctl failing to find the raid0 drive, and that's why I thought it might be your use of the wrong kernel, or an invalid kernel. If you are sure this kernel is tested and valid, then an examination of the device nodes in /dev/raid* may be helpful. The untarring of base50.tgz will not eliminate /dev device nodes, but it might renumber them.

One thing you might try, after restoring (again), is to copy your custom 5.0 kernel into the root partition of your production machine as some other name, and then selecting this kernel in single-user mode from the amd64 bootloader (the "boot>" prompt). For example, "boot hd0a:/50bsd -s" -- watch the dmesg and see if the autoconfiguration discovers the array and configures it. If it doesn't, the kernel will not find a root directory and will panic. If you are able to install 5.0-release onto a USB stick/disk, you can boot your custom kernel with "-a" and select the USB stick's root partition, and then while running from the stick attempt to diagnose why autoconfiguration or raidctl may be failing. In addition, the dmesg can then be collected and then posted here, or better yet, to the OpenBSD Project's misc@ mailing list, so that a wider and more knowledgeable audience can assist you.

Other choices are possible, including staying on 4.9, doing a clean install of 5.0 rather than an upgrade of 4.9, converting from RAIDframe to Softraid now with -current or with 5.1 when it is released on or about 1 May. Are any of these of interest?
Reply With Quote