My organization has come to really depend on FOG. They wanted me to build a hot standby server incase the primary one goes bad because we are running on physical hosts, and sometimes we can be imaging 100’s of laptops within a week it could be super critical to deadlines if this box went down.
Since we spent the money on new hardware to mirror the existing physical FOG server, I was curious if there was a way to utilize both at the same time. Is anyone else doing something similiar. How would I config FOG to do this?

Right now FOG is not the dhcp server, but our dhcp server is sending the laptops a dhcp request which tells all the new laptops where the fog server is to get PXE booted. knowing what my environment looks like how should i build this second server to maximize the functionality of having 2 beefy fog servers. I’d hate to leave it as a hot standby.

.

@llightfoot

When I inspect the page and go tot he network tab hit refresh @
https://10.33.250.5/fog/management/

It gets a server 500 response from /fog/management/
and the second thing it attempts to load is favicon.ico which is a 404 error.

L
llightfoot
about 5 hours ago

I should give some more specifics.

This is a first server build on physical hardware.
Rhel 7
I utilized the rhel 7 install instructions
I did not use any specialized command line options when kicking off the bash script to install fog.
I did disable selinux
I did disable firewall
I can access the login page after passing through the ssl error but there is nothing to log into.

The problem after a lot of investigation was a function that was only available in php 5.6. and later wasn’t available because the redhat repo that was available to the host only allowed 5.4.
the installer tried using 3 remi repo’s remi-php72, remi, and remi-safe to get access to the newer version of php.

There was an error in the logs that sent me down the right path in a round about way.
cat /root/fogproject/bin/error_logs/fog_error_1.5.9.log

the errors read this:.
remi-release-7.9-2.el7.remi.noarch
No such command: config-manager. Please use /bin/dnf --help
It could be a DNF plugin command, try: “dnf install ‘dnf-command(config-manager)’”
No such command: config-manager. Please use /bin/dnf --help
It could be a DNF plugin command, try: “dnf install ‘dnf-command(config-manager)’”
No such command: config-manager. Please use /bin/dnf --help
It could be a DNF plugin command, try: “dnf install ‘dnf-command(config-manager)’”

since i was doing this install on 2 servers simultaneously i was able to test it both ways
run that command - dnf install ‘dnf-command(config-manager)’

which was able to execute and enable those 3rd party repos and get the packages installed upon re-running the .installfog.sh

Ultimately i added those packages to my satellite server and made it available.

I believe a true fix would be to create a switch option for the install script that turns on dnf config-manager. Ultimately we only want it to be pulled down from our repo on our satellite server, but it would have been nice to be able to control it during the install. / upgrade process in the future.

When i run the installer (i have enabled https)
it tells me to validate that the database install / update your database schema by going to the url.

When i go there and click through the ssl untrusted errors (which are all valid)
the page is blank as if nothing has loaded

I go back to the cli and hit enter to continue to get past the install issue

cat /root/fogproject/bin/error_logs/fog_error_1.5.9.log

it lists httpd.service starting loaded / active
php-fpm.servive loaded / active

Sep 08 14:55:56 ssvpfog01l1l.idm.lhi.com systemd[1]: Starting The PHP FastCGI Process Manager…
Sep 08 14:55:56 ssvpfog01l1l.idm.lhi.com systemd[1]: Started The PHP FastCGI Process Manager.
ERROR 1146 (42S02) at line 1: Table ‘fog.globalSettings’ doesn’t exist

not sure what else to look at?

i have validated that the services have started as expected. but the page isn’t loading.