This is a really good explanation, except that the stepping on other applications address space has not been a problem since Windows 3.1 (which is to say it was fixed in Win95, and NT).
Most of the system freezes seen these days seem to be due to storage leeks in the OS, although I traced one screen freeze to a bad driver. Oddly enough I found that I could get out if that particular freeze by hitting CTRL-ALT-DEL twice, then ESC twice. The first CTRL-ALT-DEL brought up a message on the screen, which I could not see, while the second put me in that blue screen with the "press CTRL-ALT-DEL to restart" message. Pressing ESC put me back in Windows, with the screen unfroze, and the second ESC got rid of the message in Windows. -----Original Message----- From: Daniel Casey [mailto:[EMAIL PROTECTED]] Sent: Monday, February 03, 2003 7:28 AM To: [EMAIL PROTECTED] Subject: Re: Down time I think it's partly due to hardware. S/390 hardware is very reliable/stable. I think it's partly due to software. In the "PC" arena, and by PC I mean Windows/Intel, software seems to be released before it should without the proper testing. My experience with S/390 has been that because such critical applications are running on the Mainframe, having an outage is not an option, therefore extensive testing is done. The attitude is a little different. Mainframe'ers seem to be more thorough in their work. I think the attitude among some, not all, in the PC arena is that if all else fails, just do a CTRL+ALT+DEL. There are "acceptable" risks. Incomplete testing, etc. >From what little I understand about the architecture of Windows compared to other platforms is that Windows does a poor job of memory and process management, causing a lot of the headaches. Not necessarily Blue Screens...don't really see too many of those any more, but I hear problems of applications with "memory leaks" and applications, once in memory, will walk on other applications' address space...bad, very bad. Now, my experience with Linux on Intel has been quite favorable. Once all the proper drivers for the particular hardware has been properly configured, it will run like a champ for months if not years before a reboot...er, I mean IPL. |---------+----------------------------> | | "Abruzzese, Pat" | | | <Pabruzzese@Thomc| | | omp.com> | | | Sent by: Linux on| | | 390 Port | | | <[EMAIL PROTECTED]| | | IST.EDU> | | | | | | | | | 02/03/2003 09:12 | | | Please respond to| | | Linux on 390 Port| | | | |---------+----------------------------> >----------------------------------------------------------------------------------------------------------------------------------------------| | | | To: [EMAIL PROTECTED] | | cc: | | Subject: Down time | >----------------------------------------------------------------------------------------------------------------------------------------------| I know this is off the board but I would like to know why is the "mainframe's" down time limited when the "client/servers" seem to going down whenever. In the middle of the morning, afternoon or night unscheduled. My VM/ESA 2.4.0 was IPL'ed 1/09/2002 and have been up since. I will take it down this Sunday to put Z/VM 4.3.0 in service. Why are there two sets of standards??? vr, P. Abruzzese
