Here is a simple recipe to reproduce the massive test lossage in -current:
cd /usr/tests/dev/raidframe && atf-run
In the output you will find:
tp-start: 1575709106.330783, t_raid, 7
tc-start: 1575709106.330864, old_numrows_config
tc-so:Executing command [ rump_server -lrumpvfs -lrumpdev -lrumpdev_disk
-lrumpdev_raidframe -d key=/disk0,hostpath=disk0.img,size=1m -d
key=/disk1,hostpath=disk1.img,size=1m unix://sock ]
tc-so:Executing command [ rump.raidctl -C raid.conf raid0 ]
tc-se:rump.halt: reboot: Socket is not connected
tc-se:t_raid: ERROR: The test case cleanup returned a non-ok exit code, but
this is not allowed
and then all tries to use the same socket will fail.
On macppc I get this kernel log when the rump_server process dies:
[ 51469.7829284] trap: pid 9120.28 (rump_server): user read DSI trap @ 0x20 by
0xfdca89d4 (DSISR 0x40000000, err=14)
Martin
