(Illustration by Gaich Muramatsu)
First I go try to make an 880M RVM data thingie (experimenting as if I were going to make a 20G server thing). Doing it manually (since the script wouldn't figure it out), [root_at_atlas coda-5.0.0]# rdsinit -f /vdat/cuc-log /vdat/cuc-data 922746880 0x20000000 964689920 0x100000 80 32 rvm_initialize succeeded. Going to initialize data file to zero, could take awhile. done. ? ERROR: rds_zap_heap RVM_EINTERNAL. didn't work, so I selected the 315M size in vice-setup, so far so good, but vice-setup-rvm should allow for a wider range of sizes... I start the server, the thing eats up RAM out the wazoo: 983 root 11 0 252M 231M 696 R 0 13.8 92.1 0:04 codasrv And it crashes (see attached log file and stack trace). I have 256M real + 128M swap space on the system - is that enough for a 315M RVM data file? So, how can I help fix these problems? :) Also, I remember way back there used to be a kernel module that would allow open-by-inode or something, to speed codasrv up with the /vicepa stuff. Is this still existent relevant? -- Elliot "In film you will find four basic story lines. Man versus man, man versus nature, nature versus nature, and dog versus vampire." - Steven Spielberg