Best of all, on 5.2 you are unlikely to have the problem. We have not had even one hung user in nearly 6 months of running 5.2. The only time we had problems of any kind was at implementation time. We had to fall the system back 3 times because of show-stopper problems encountered while we were checking the system out prior to letting the users on. One was with IBM and 2 with other vendor code. When we finally were able to stay on 5.2, it became smooth sailing.
Reporting a problem on back-level software is not always a bad experience. Sometimes, someone more learned than us customers in the magic incantations needed to search the database might run out of things to do and have a look-see to determine if there is already a PTF available. Sometimes, they even do it as a break from looking at the dump that they have been staring at for hours. -----Original Message----- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Berry van Sleeuwen Sent: Thursday, January 04, 2007 11:23 PM To: [email protected] Subject: Re: Logoff force pending Alan, Unfortunatly there is no dump. First of all because we can't risk the VM = to be unresponsive (or even IPL) because our zlinux machines must be up = 24x7. One of them has almost daily problems with performance (or so they = claim) and an other is not allowed to log off at all. (Some wiseguy sold = the system without downtime and also without failover so we can't even do= maintainance on this VM without creating issues on that machine) Also this is an issue on a zVM 4.4 system. Last week I had a hard time = convincing IBM we do have a supportcontract on supported software so you = can imagine the problems I would have in reporting an issue on unsupporte= d software. And since the user eventually logged off I can't get a dump at a convenient time in the future should there be one. Even so, next time I have a user like this I will be on zVM 5.2 so then I= can safely open a PMR. Also our VM configuration will be probably support= full failover for the linuxguests by then so that we can failover to a = different VM if we need to. Thanks, Berry.
