Virtual TapeServer 6.04.03 for NonStop Servers Release Notes

Known Issues | 31
Documentation
 Workarounds should be documented for sudden shutdowns. Following a hard
(unclean) shutdown of VTS (such as a system crash or loss of power), the system may left
in an inconsistent state. Many services running on the server use PID files to tell the
system that a service is running. When a service is started, the PID file is created. When a
service is shutdown, the PID file is deleted. If VTS is shutdown uncleanly, the PID files
will not be deleted. When the system comes back up, it assumes that the service is
running and does not try to restart the service. This will cause VTS to function incorrectly
or not at all. Here is an example of a service failing to start due to a stale PID file that
resulted from a crashed server, in /var/log/messages. (TL-3444)
Nov 6 12:29:18 ProdVTS rhdb: Starting PostgreSQL - Red Hat Edition
service: failed
WORKAROUND: Reboot the server twice. This will allow the server to clean up the PID
files as the server shuts down and allows all of the necessary services to start properly
when the server boots up.