i am having a possibly similar issue after upgrading to 1.5.9.
During the install it prompted me that it was changing the fogstorage password. i copied the note which was:
The current fogstorage database password does not meet high
security standards. We will generate a new password and update
all the settings on this FOG server for you. Please take note
of the following credentials that you need to manually update
on all your storage nodes’ /opt/fog/.fogsettings configuration
files and re-run (!) the FOG installer:
snmysqluser=‘fogstorage’
snmysqlpass=‘redacted’

i changed the referenced settings, changing the user to fogstorage from fogmaster. then reran the install as noted. then ran the database update from fog web ui and got error:
Failed to query: Error: SQLSTATE[42000]: Syntax error or access violation: 1142 ALTER command denied to user ‘fogstorage’@‘localhost’ for table ‘nfsGroupMembers’ Error Message: Error Code: “42000”, Error Message: [“42000”,1142,“ALTER command denied to user ‘fogstorage’@‘localhost’ for table ‘nfsGroupMembers’”], Debug: SQL: [92] ALTER TABLE nfsGroupMembers ADD COLUMN ngmHelloInterval VARCHAR(8) AFTER `ngmMaxBitrate

in cli (after hitting enter):
Update fogstorage database password…Failed!

Can i roll back to a previous version?

I dont know which version specifically but when the ui was changed i am now having numerous issues.

when a machine is inventoried i now no longer see a list of macs, just primary and if its a usb nic because no laptops have ethernet ports it doesnt allow me to remove it as its a required field. I try putting in a dummy mac address so i can register a new machine to image with that usb nic and it still picks up the old machines name even after i cleared the mac from its record.

on the machine side getting client errors: object reference not set to an instance of an object.

saw another post saying this was a virtual nic issue. and i had created my first image in esxi. even though i saw no virtual nics or hidden devices in devman i made a new image from a physical device, like a caveman. same error.

device also is not rebooting now but want to resolve these fog/client issues before trying to identify that.

@george1421 sorry, i was still researching what might be happening.

For a little bit more background, i can’t get my vm to boot to the win10 1803 iso using efi but I didnt have an issue with the ltsb version last year.

I setup a second esxi server for testing, running 6.5 to see if it was issue with 5.5 and i am able to boot win10 1803 using efi on that test server but not able to pxe boot.

efi network comes up then says downloading nbp file then continues booting. I dont see any other messages but i will do a slow mo video see if i can see if anything flashes up briefly.

On the production server that I can’t get the iso to boot, I was able to get it to pxe boot after i deleted the nic and re-added it using vmxnet3.

recap:
prod esxi server 5.5u3 – cant boot 1803 iso using EFI
Test server esxi 6.5, able to boot iso to EFI but cant PXE boot EFI.

I just did quick google search that brought me to this thread and was just making a general commenting that i was having issues. i can make a new thread if that would be better.

i did not have any issues last year getting efi vm’s to pxe boot but cant get them to work for the life of me now. I dont recall exactly what verion of esxi i had last summer but i think it was 5.5 u1? on u3 now and only other change is win10 1803 vs 1609 ltsb.

hmmm, maybe this was an issue with a previous svn. i just made a test iamge and deleted it and it was removed from the disk.

problem solved

@Tom-Elliott
I did check the “delete file data” check box

@Wayne-workman
“deleting enmass” option isnt really needed or necessary. But if i have an old iteration of an image that is no longer needed and i remove it from the webui i should not also need to login to the cli to delete it from the disk.

i was clearing out some old images to free up some space but after deleting the old images free space stayed the same. when i login to cli i can still see all the folders with images still there. i will delete these manually but would like to not have to delete them manually from cli. not sure if this is a bug or if there is some config that would be causing this.

snv 4250

Sorry for the delay. It was logging an error in the event log that the fog service hung I set it to delayed startup and that seems to be a reasonable workaround for the issue.

i recently updated my image to include some update Adobe software as well as some new drivers. During boot now the fog service hangs which causes it to sit at “please wait” until the service times out then the login screen comes up. I am not sure what is causing it. the service is currently set to automatic. should ichange this to delayed startup? any other ideas?

Thanks,
Caleb

for clarification, if i edit the host in fog, selecting option to join domain it populates from the default ad setting i have configured and the machine promptly reboots and joins the domain (very slick).

hello,

I have a group in fog configured to join members to the domain. It seems like this was working before but it does not appear to be remembering the settings now. If i do a full host registration, populate host name, choose image, choose group then default answers for the rest until prompted to image now which i say yes to. It images fine and renames properly but doesnt join to domain. If i go in and select join domain on the host it will. When i look at the group and i check the join domain option it populates the fields from the default AD settings, i hit update then reload the page and the ad settings are blanked out again.

I was able to work around the issue. I had to migrate the trouble vm to the same host fog was running on and disable the second nic on the dhcp server (while pxe booting it was grabbing the ip of the second nic, not sure why).

thanks for the help guys.

@Wayne-Workman what ip is supposed to be there, in the tftpd-hpa file? the article says 0.0.0.0, is that correct?

@Tom-Elliott the service sops and then starts.

@Tom-Elliott

here is the contents of the tftpd-hpa file from my fog server:

/etc/default/tftpd-hpa

FOG Modified version

TFTP_USERNAME=“root”
TFTP_DIRECTORY=“/tftpboot”
TFTP_ADDRESS=“:69”
TFTP_OPTIONS=“-s”

@Wayne
I have two windows vm’s running on a vmware cluster, same subnet, that i use to create images. both are win7. from the one vm i run the tftp test referenced in that article and it succeeds.

tftp is running on the fog server, and other systems are able to pxe boot. i also disabled the firewall.

getting tftp timeout. looking at the wiki article: https://wiki.fogproject.org/wiki/index.php/Troubleshoot_TFTP
under the tftp settings file section it says that on ubuntu the file should look something like this:
'# /etc/default/tftpd-hpa
'# FOG Modified version
TFTP_USERNAME=“root”
TFTP_DIRECTORY=“/tfptboot”
TFTP_ADDRESS=“0.0.0.0:69”
TFTP_OPTIONS=“-s”

When i look at mine the tftp address is just “:69”
would this cause issues?

not sure if its related but during a reboot i saw an error flash by briefly, it said:

Could not reliably determine the server’s fully qualified domain name, using 127.0.1.1. Set the ‘ServerName’ directive globally to suppress this message

a little more info:

i re-ran the install and when i got to the part when it wants you to browse to the management page to update the database; when i loaded the page it did not prompt, just had the login screen (did not prompt first time through either).

tried different browser, able to browse through webui until i click “create new image”, gets http 500 error.

Earlier i was trying to upload an image and it kept hanging at “checking in”. I let it sit for a few hours to be sure. i rebooted, no change so i updated to latest trunk (4240). Tried to upload again and failed with nfs mount failed. check .mnt files were in /images and /images/dev, /etc/export looks correct as well. I tried to create a new image and am getting white screen with message along top left corner of window says “Add new image definition”. Not sure what to try next, any recommendations?

@Wolfbane8653 thanks for the help. I thought there might be issues running both svn and git but it went smoothly. Thanks again.