Story goes,

I have a set of servers i built then i have a set of servers that were here before me.

Same release level of Centos 7
Same bios
same Config for the most part.

When i updated the kernel on a server i built, i have no issue. it boots up fine.

When i update the kernel on a server that was built before mine, i get into the Dracut shell with the below errors. if i switch back to the old kernel, it boots up fine as well.

- ive tried rebuilding the initramfs for the new kernel, same result.
-tried running lvm vgscan and i get nothing returned besides the message stating it may take awhile to scan then it instantly moves down to the next line of "dracut:/#" while returning no results.
-ive compared the two lvm.conf files to see if there was any difference at all and i could not locate any.

anyone have any other ideas of what to try or look into?

something else to note, when i ran the dracut -f to rebuild the initramfs i got the below error.

but when i navigate to that directory i see files in there all named modules.xxx.

bump, any suggestions or tips?

bumpp. anything?

check in the root folder you should have two hidden files
.diskinfo
.treeinfo

Once you have these two files yous issue must be resolved

i have neither of those files.

My first thought on looking at your OP was that for whatever reasons, the system cannot find your partitions. You mention LVM, it could be that LVM has not activated your partitions or similar. You do need to rebuild the initramfs when making major system changes, like installing drivers etc.

Incorrect UUID's in /etc/fstab ?
Problem with your GRUB configuration ?
Corrupted partitions ?
Filesystem problems ?

While this isn't about CentOS installed on "bare-metal", it may give you some ideas.

yea, lvm scan brings nothing back in the dracut shell. blkid only bring up the 20T raid that is setup.

if fstab was incorrect, wouldnt it crash in the older kernel as well? same with corrupted partition? ive checked the grub config and nothing jumped out at me.

After one kernel update, I had exact same issue. LVM did not see local disk.
I am running on HP hardware and at the time was at CentOS7.4
Unfortunately, system I had was critical and did not have a lot of time to troubleshoot.
So, I ended up doing a clean install and with the new install I do not have LVM for /boot and /
And, since then I ended up doing same on other nodes (clean instal with no LVM for /boot and /).
These days /boot probably does not have to be different partition, but I guess I am old school.
Sorry, this perhaps does not help you with finding solution to your problem...

ultimately i believe this will be the resolution. But I'm adamant about wiping it and re-imaging since I created 40 servers that are almost identical and update without issue.