<div dir="ltr">Hmm. That bug was supposed to have been fixed a while ago... like a few months. But if not, there's more to it and we have to track it down. A double 'da0' is typically due to a unit allocation race. i.e. two ahci channels allocate their da* units at the same time without the correct lock.<div><br></div><div>I will look at the code paths again and try to figure out why it might still be happening.</div><div><br></div><div>-Matt</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Fri, Mar 23, 2018 at 1:15 AM, Daniel Bilik <span dir="ltr"><<a href="mailto:ddb@neosystem.org" target="_blank">ddb@neosystem.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class="">On Fri, 23 Mar 2018 09:52:50 +0200<br>
"M. L. Wilson" <<a href="mailto:ipc@peercorpstrust.org">ipc@peercorpstrust.org</a>> wrote:<br>
<br>
> I had a similar issue some weeks ago. I could not really track down the<br>
> real culprit but your description and dmesg output were similar to mine.<br>
> Somehow, a full world/kernel rebuild solved the issue as I have not seen<br>
> it again.<br>
<br>
</span>Thank you for responding... As I've written, I am on master, and I always<br>
do full buildworld+buildkernel when upgrading. Currently I run...<br>
<br>
5.1-DEVELOPMENT DragonFly v5.1.0.1073.g51a52-DEVELOPMENT #0: Mon Mar 19 11:11:20 CET 2018<br>
<br>
... and with this kernel I've hit the issue this week. Thus my previous<br>
question, whether there was some commit between 10th March and today that<br>
could have fixed it.<br>
<br>
--<br>
Daniel<br>
</blockquote></div><br></div>