Any current version of the HAL daemon crashes in hald/linux/blockdev.c when started on a system with (soft)RAID volumes. The current stable and unstable amd64 versions of HAL were tested and exhibited the exactly same behaviour. The problem is discussed on the Debian bug lists, but it is present on Gentoo as well. My assumption is that the problem is inherent ot the current HAL sources and does not depend on distribution och platform. (ref. https://bugs.launchpad.net/hal/+bug/361689 ) ^^^^ Sorry for the link, but the information in the conversation is revealing and the site provides the diff too. The Debian patch applies cleanly, once Gentoofied, and it seems to resolve the issue on one particular system (mine, well...) Reproducible: Always Steps to Reproduce: 1. Create a RAID volume (mdadm --create md1 --raid-devices=2 --level=1 /dev/sda5 /dev/sdb5 ) 2. Start hal ( hald --daemon=no --verbose ) 3. See that hald segfaults. Actual Results: The system indicated that hald had failed to start. There was no usable information in the log. Multiple cleanups ( hald -zap ) gave the same result. Manual run of hald revealed the segmentation fault. Expected Results: Hald should be running fine even if it sees RAID volumes :) The content of the patch applies cleanly to blockdev.c. I am *not* the author of the patch, full credits to the actual author.
According to upstream report, this is in 0.5.13 release. Closing fixed. Thanks for reporting.
