@tom-elliott

DELETE FROM `fog`.`snapinAssoc` WHERE `saHostID` NOT IN (SELECT `hostID` FROM `fog`.`hosts`) OR `sSnapinID` NOT IN (SELECT `saSnapinID` FROM `fog`.`sID`);

Returns #1146 - Table ‘fog.sID’ doesn’t exist

@hregis said in Login incorrect when an image deletion request:

/var/www/fog/lib/fog/fogftp.class.php

Don’t know why this ext3 thing is coming to my mind all the time when people have FTP errors on deleting stuff. Never mind. See the code on why you get those errors. It first tries normal deletes before it does a recursive delete on the files. As the directory is not empty it fails at first. So it’s all fine!

Just ask, one of the moderators or myself can do so.

0_1501451714838_635a607c-f488-4dee-b051-a94f26459b76-image.png

@tom-elliott
I installed 1.4.4. and am back up and running again. Everything seems to still be in tact. Thanks again for the help.

@fractal13 Hey, long time no hear, hope all is well.

Wondering if you (being the original writer and all) wouldn’t mind reviewing my latest procsfdisk.awk script? Had a few issues in 1.3.x days that kind of required a rewrite of it all, which went fairly well. Last night (early this morning) I hope to have solved the mystery of extended partitions though I’m sure it’s not perfect. I can’t see the “flow” nicely. So a good review could be very very useful. (Others are welcome to it as well.)

Thank for your answers. Due to the nature of our work and why we use this FOG server it’s unfortunately air-gapped from the internet so updates have been hard to come by regularly. The only thing that makes sense on root-cause here was one severely borked imaging operation. Even then I’m grasping at straws for a good answer on what triggered the error.

However, I pushed through our ITsec group and got the server updated to current stable release 1.4.4 this morning which did resolve this error and it’s deploying images normally again.

Thanks for the info guys feel free to revoke my nerd card for it being resolved with a standard update.

Something JLE said is the first thing I thought about (ip helper-addresses). If you can get access to login to the switch, find the relevant documentation online for your switch and verify if the VLAN has an ip helper-address configured.

On our ProCurves, we could check like this:

show ip helper-address

or we could configure like this:

config vlan <vlan_number> ip helper-address <IP_of_your_DHCP_server> write mem end

@SmartyParts I am marking this solved but I’d highly recommend looking into UEFI boot (ipxe.efi, snp.efi, snponly.efi, …) as more and more devices and operating systems are heading for this.

@tom-elliott You rock Tom! Thanks for the help man! : )

@sebastian-roth about the live system, i need to find the time today, i have the book until next friday.

Thank you for the quick response. I normally create my images using a virtual pc. This time I did not do that and forgot to check the bios settings. I did not even think about checking that. Thanks again.

@hancocza All the more reason version information is needed everytime. Thanks for letting us know it’s back and running.

@Tom-Elliott Hi Tom i figured it out a college had changed settings in the DHCP all sorted now.

thanks again

Hi @Tom-Elliott,
That might just be the solution. Not sure why, but I had the perception that having a db with no password was the recomanded way. Again, thanks for helping me out and for developing and giving support to this fantastic tool!

@ITCC Please update to RC’s if you can?

I don’t know what’s causing it to create “false” entries in the database, though I cannot get it to replicate the problem in the current code I’m working with.

@plegrand I just stumbled upon the FOG_ENFORCE_HOST_CHANGES setting in the FOG web UI and remembered your post:

This setting only operates with the new client. Default value is 1 which allows the new client to enforce name changing on every cycle it checks in, so any change on FOG will take place on the next cycle. If unset (value 0) it will only perform hostname change and/or AD Joining on host restart.

You might want to give this a try to get rid of the messages.

@Sebastian-Roth
@Tom-Elliott

I just updated to RC4… I’ll keep you all posted! Thanks!

@Sebastian-Roth fixed.

@Sebastian-Roth I would say yes, The network wasn’t down at all. I was able to ping and checked the Linux firewall to be sure it was disabled.

It started doing it again this morning. But everything works. So I will chalk it up to have a system that has been updated more times than I would like to think about unless you guys want to look into it further.