Engine deployment fails with engine booting to grub rescue
After working through some DNS problems I was able to install oVirt 4.4.1.1 and progress the engine deployment stops within the ansible scripts. The system deploys the gluster domain, creates the engine, starts the engine, however it will never fully boot and complete the deployment. Bringing up the console of the engine sets it into rescue mode. I have attempted to set the appropriate variables in grub rescue set root=(hd0,msdos1) set prefix=(hd0,msdos1)/boot/grub insmod normal And this will not boot the engine. Is there a procedure to mount a CENTOS 8 rescue CD/ISO to troubleshoot this or is there a specific fix published for this problem? The specific error I am seeing in the Engine Console is "attempt to read outside of HD0. I upgraded the ovirt engine appliance package to ovirt-engine-appliance-4.4-20200716090128.1.el8.x86_64 which produces the same result. I have gotten this error deploying on both my Intel and AMD based servers. any help is appreciated. thanks Andy
After doing a bit more research, I was able to get the engine to boot by issuing CTRL-ALT-DEL through remote viewer. I am uncertain as to why that appears to fix the problem with the engine, however after such I have added additional hosts, and reinstalled the first host that failed on the original deployment. During the three previous failed deployments, using the OVIRT rpms from the resources site, I manually specified the engine OVA path, which produced the same results. I dont know if this install problem has to do with imageio or how all the OVA's were sealed on creation but the logs from the engine doesnt show anything out of the ordinary and the engine has all the environment variables inputted during the deployment. Will attach the setup and imageIO logs for reference. thanks
participants (2)
-
Andy -
farkey_2000＠yahoo.com