Hi. I just tried to update the Debian package of insserver from 1.14.0 to 1.16.0, and discovered that the script ordering produced is completely bogus. The bug is in the unpatch package too, but not detected by the test suite. Here is the last output from 'make check':
info: test if loop involving virtual facility is detected. $local_fs : mountall $local_fs : mountall $local_fs : mountall $local_fs : mountall $local_fs : mountall $local_fs : mountall insserv: There is a loop between service hibernate-cleanup and mdadm if started insserv: loop involving service mdadm at depth 3 insserv: loop involving service mountall at depth 2 insserv: There is a loop between service hibernate-cleanup and mdadm if started insserv: loop involving service hibernate-cleanup at depth 1 insserv: There is a loop between service hibernate-cleanup and mountall if stopped insserv: loop involving service mountall at depth 3 insserv: loop involving service mdadm at depth 2 insserv: There is a loop between service mdadm and mountall if started insserv: There is a loop between service mountall and mdadm if stopped insserv: loop involving service hibernate-cleanup at depth 4 insserv: exiting now without changing boot order! insserv.conf: $local_fs mountall init.d: hibernate mdadm mountall.sh boot.d: rc0.d: rc1.d: rc2.d: rc3.d: K01mountall.sh K02hibernate S01hibernate S02mountall.sh rc4.d: rc5.d: rc6.d: rcS.d: success: 7 test executed, 0 nonfatal tests failed. Note how rc3.d contain both K01mountall.sh and S02mountall.sh. This is completely wrong. It should only have S02mountall.sh, given that the header look like this (from tests/common): insertscript mountall.sh <<EOF ### BEGIN INIT INFO # Provides: mountall # Required-Start: # Required-Stop: # Default-Start: 3 # Default-Stop: # Short-Description: Mount all filesystems. ### END INIT INFO EOF Anyone got a clue what is going on? -- Happy hacking Petter Reinholdtsen
