Pjetro de 8.Jul.10 02:17 AM a Web browser Domino Server 6.5.5 FP2 Windows 2003
We had some application LN servers up for nearly 2 years, which is astonishing stability (cca one thousand users). After 2 years one of the member of our important cluster crashed and the 2nd member of the cluster took all the work (thatīs the purpose of the cluster). But the two servers in the cluster were not really equal, the firt server was colaborating with other configured systems (out of Domino) and the second one was not. It was necessary to bring up the first member of the cluster asap. But the restart was not possible, always occured "error 1067" after cca 10 seconds of the server uptime, when we tried to restart Domino service. Restart of the operation system did not help. We are using transactions logs, so we created tens of gigabajts of the logs again - not helped.
Finally we find out the server crashed just after the router starts. So we tried the following: we started the server without all the tasks and the server was stable. Just after we started the router, server crashed. Other tasks had no impact on this, only router. Router was the cause. As we checked the server DBs, we noticed the mail.box is abnormally big, over 21 GB. Totally corrupted, no compact and no fixup helped. So eysy to rename and restatr the server (it created new mail.box if it has a template). And voala, everything worked again! Just a few missed undelivered mails.
So checking the system DBs of the server and run an offline (not server up) compact -i -B -c and fixup -L to prevent system BDs damage - is the good idea. But thatīs other thing called "maintanance".