phd_ivo@gsbacd.uchicago.edu (11/22/89)
*************************************** I am not a professional process administrator, so please forgive my ignorance if the following message reflects it. I can understand that the -console (WindowServer) process uses about 8M of Virtual Memory (VMEM)---even if it handles only one or two windows. What eludes me, though, is why (kernel-task) sits there with 14.4M of VMEM, nmserver with 6M (despite my not running NFS as of yet), and lpd with a whopping 16.5M of VMEM. Moreover, if I set "biff y" in my .login file, comsat processes just appear to multiply and in proportionate number with the number of incoming mail messages. Now, real hackers never die, but does the same thing have to apply to comsat processes? I suspect that these large memory sizes and large process numbers have something to do with the occasional failure of process upstarts It just sits there and waits. This is more a nuisance than a real problem for me. Still, does anybody know if this is expected, normal behavior, if I am doing something wrong, or if Mach is doing something wrong? Ivo Welch ivo@next.agsm.ucla.edu iwelch@agsm.ucla.edu phd_ivo@gsbacd.uchicago.edu normal behavior