Differences

This shows you the differences between two versions of the page.

--- cluster:225 [2024/05/06 19:03] – hmeij07
+++ cluster:225 [2024/05/21 14:06] (current) – hmeij07
@@ Line 63: / Line 63: @@
 <code>
-# n78 first ... (no problem, tests success)
+# n78 first ... can reimage cuda-11.6 from n101 (no problem, tests success)
 # make sure /usr/src/kernels/$(uname -r) exists else
 # scp into place from n100 (centos8, possibly caused by warewulf...)
+# however old nvidia packages still in OS (driver 510 toolkit 11.6)...
+#  rpm -qa | grep ^nvidia | wc -l results in 16 packages...
+# what happens on dfn check-update ???
+# n[100-101] skipping for now
+# this is a package install and there is no nvidia_uninstall (runfile)
+# an upgrade would require internet 'dnf check-update; dnf update')
+# switching between rpm install and runfile is NOT recommended
+# and 'dnf erase nvidia*' may leave a hung system behind
 # n79 next (no problem)
@@ Line 82: / Line 91: @@
 # n80 (same error upon reboot after driver install)
 # n81 (same error upon reboot after driver install)
+# n90 (same error upon reboot after toolkit install, not driver. weird)
+# n88 (failed toolkit install, ran /usr/bin/ndia-uninstall, reboot
+#      re-installed driver, reboot, re-installed tookit, reboot,
+#      no error occurs! )
+# n87 (ran nvidia-uninstall first, driver then toolkit, errors shows up)
+# n86 & n85 same as n87
+# n84 (no error shows up)
+# n82 (as n87 but error shows up after driver before toolkit install)
+# n83 (error shows up fater toolkit install, not driver install reboot)
 sh ./NVIDIA-Linux-x86_64-550.67.run