Created attachment 913601 [details] screenshot Description of problem: Anaconda "Installation Destination" Device Selection lists no disks. Yellow banner at the bottom says "No disks detected." There is an internal SATA SSD in this laptop, nothing unusual about the hardware. Version-Release number of selected component (if applicable): anaconda-21.45-1.fc21.x86_64 python-blivet-0.59-1.fc21.noarch How reproducible: Non-deterministic. Steps to Reproduce: 1. 2. 3. Actual results: No disks detected. Expected results: Should appear as a SATA device. Additional info: Bug 1114770 may be related. I get either this bug or bug 1114770 most of the time, maybe 20% of the time the SATA SSD appears in anaconda as sda. lsblk shows sda and all of its partitions, so seems either anaconda or python-blivet are confused.
Created attachment 913602 [details] storage.log
Created attachment 913603 [details] anaconda.log
Created attachment 913604 [details] program.log
Created attachment 913606 [details] journal journalctl -l -b -o short-monotonic
21:56:21,805 INFO blivet: type detected on 'sda' is 'multipath_member' 21:56:43,641 DEBUG blivet: ignoring partition sda1 on multipath_member OK so it probably is ultimately bug 1114770, but why does blivet sometimes ignore and conclude no disks detected, and other times it shows it as a multipath device?
OK no this is somehow different because /dev/mapper doesn't have any multipath entries with this boot, whereas there are mpathc entries in /dev/mapper for bug 1114770.
I think multpathd is just confused and it then confuses anaconda or blivet. Probably a dup of 1114770, but I'll leave it open in case the variable behavior is considered an anaconda or blivet bug.
We can only go by what the system tells us. multipath is saying that it is a valid device, although the setup seems to have failed. *** This bug has been marked as a duplicate of bug 1114770 ***