**** BEGIN LOGGING AT Tue Oct 24 02:59:57 2006 Oct 24 23:25:33 is this a known issue with mdadm on openslug? Oct 24 23:25:37 mdadm: No superblock found on /dev/sdb1 (Expected magic: a92b4efc, got fc4e2ba9) Oct 24 23:25:42 (somebody didn't pay attention to the endianness) Oct 24 23:26:12 I'm trying to bring over a RAID5 array Oct 24 23:28:17 I wonder if it's in the md driver or the mdadm utility itself... I can mount filesystems from other-endianned systems just fine, so the base block drivers should be fine Oct 24 23:44:34 rwhitby: do you know if there is a known incompatibility with openslug's mdadm and "normal" mdadm? Oct 24 23:44:55 I think there's an endianness issue with either the driver or the utility, haven't dug in too much yet to see which Oct 24 23:45:43 tzanger: it's quite possible you're the first to notice it Oct 24 23:45:57 mdadm: No superblock found on /dev/sdb1 (Expected magic: a92b4efc, got fc4e2ba9) Oct 24 23:46:16 rwhitby: that's on a drive which is aprt of an array created on x86 Oct 24 23:46:18 (we're not aware of an incompatibility, but it's a fairly esoteric package compared to the rest of the feed) Oct 24 23:46:36 ok, now you're *really* likely to be the first to try that :-) Oct 24 23:46:46 :-) Oct 24 23:46:52 well, it's a powersaving move Oct 24 23:47:09 moving from a duron1300 or something with 6 IDE drives to a slug with a hub and 4 drives in bays Oct 24 23:47:21 Yep. Good plan. Oct 24 23:47:53 I screwed around with ACPI CPU throttling and hdparm drive spindowns to end up losing one of the drives in the array Oct 24 23:48:12 and after calculating the power savings (20W -> 7W/drive) it wasn't worth the constant RMAs from the spinup problems :-) Oct 24 23:53:27 I need to take a more careful look at OE; it might be worthwhiel porting my code over to its build environment if it has builds for m68knommu **** ENDING LOGGING AT Wed Oct 25 02:59:57 2006