mountall will keep trying to reconnect anyway and with the main loop fixes, that initial connect goes away
The remaining bug is that sometimes mountall might not prompt for boredom (but only if plymouth takes >2s to open its socket), or might not prompt for fsck failure (but only if plymouth takes longer to open its socket than fsck takes to run)
Both of those cares are fairly unlikely
And it's far too close to release to muck around with Upstart's ptrace code
mountall will keep trying to reconnect anyway and with the main loop fixes, that initial connect goes away
The remaining bug is that sometimes mountall might not prompt for boredom (but only if plymouth takes >2s to open its socket), or might not prompt for fsck failure (but only if plymouth takes longer to open its socket than fsck takes to run)
Both of those cares are fairly unlikely
And it's far too close to release to muck around with Upstart's ptrace code