04:57:40 Is there a way to directly access the EFI partition from a running OmniOS instance (no zones)? I tried the suggested command in pcfs(4) but was told there wasn't a boot partition on the disk. I see it when I boot into EFI shell, so I know it exists. 04:58:30 "mount: /dev/dsk/c1t4d0p0:boot is not a DOS filesystem." 10:18:29 Yes, just add `s0` to the device and mount it as PCFS 10:18:38 pfexec mount -F pcfs /dev/dsk/c6t589CFC20095F0001d0s0 /a 10:18:40 for example 15:25:42 I tried that last night. 15:25:45 ; please mount -o ro /dev/dsk/c1t5d0s0 /mnt 15:25:46 mount: /dev/dsk/c1t5d0s0 is not this fstype 15:25:59 you need the `-F pcfs` too 15:26:17 yeah, I tried that but did it wrong. This time it worked. 15:26:28 I tried so many combinations and permutations I lost track. 15:26:55 sadly, the log file that's in /EFI isn't showing up :( 15:27:06 (I know its there, I can see it when I boot into EFI shell.) 15:27:37 actually, it might be in the directory above EFI. I'm going to go back to EFI shell and see if I can move it. 15:27:38 That's strange. There s a log storage area in the EFI variable store, I wonder if that's what shows up in the shell? 15:28:03 Give me a few minutes and I'll know :) 15:29:28 IPMI is useful at times. 15:49:51 yeah, EFI shell shows the log file as being at the same level as the EFI directory. 15:51:57 :/ 15:52:15 moving the files into the EFI directory didn't make them visible in the mounted filesystem. 15:56:34 even moving them into the same directory as bootia32.efi and bootx64.efi doesn't make them visible. 15:56:40 I guess I'm driving in with a USB key. 17:02:17 * nomad slaps forehead 17:02:26 fs0 might not have mapped to t4, it might have been t5. 17:02:39 ah well. I'm going to say "it's too early to think" and move on. 19:03:25 :/ Looks like there might be a driver problem with this HBA. I presume that's an illumos bug report, not an OmniOS one. 19:03:36 (as in, newer driver is needed) 20:03:32 What's the HBA? 20:08:41 Boradcom 3108 20:09:03 specifically the 'invader' 20:09:48 The host booted and 'worked' with the driver in the LTS release but it's logging a lot of errors and the device says a driver update is needed "to enable extended LD support" 20:10:32 I've just upgraded the host in question to 151044. I suspect I should clear the device logs and see what shows up. 20:14:39 Hmm. Less certain on that one then. 20:32:58 well, I found it 20:33:30 the mr_sas version reports 6.503.00.00 20:33:50 The version I need is 7.723.02-00-1. 20:34:00 so I guess I need to file a bug report with illumos. 20:34:10 This is going to cause me some grief :( 20:40:28 Well, https://illumos.org/issues/15302 has been created. 20:40:36 hopefully I provided sufficient information. 20:40:58 andyf, danmcd_, please poke me with any updates that need to be made to that ticket. 20:58:38 OK, so the firmware is reporting that. 20:58:50 The version number we have is kind of meaningless. 20:59:09 So it really is about figuring out what that support is and implementing it. 21:43:19 that's disapointing. 21:45:59 Sorry, I'd rather be up front and honest with you. 21:46:17 of course 21:46:31 I'm just sitting here with a file server that I can't put into production until this is resolved. 21:47:22 I'm not a kernel dev (I'm not even a dev) so I've no idea how much work is involved in this. 21:47:30 I just think "it's non-trivial". 21:47:54 It is a bit hard because the mr_sas series historically hasn't been well documented.