@Tom-Elliott said in Fog 1.5.10 Install error libcurl4 Failed during install:

@jaumeramos Would you be willing to test:

diff --git a/lib/common/functions.sh b/lib/common/functions.sh
index b20b0e482..05f40d08c 100755
--- a/lib/common/functions.sh
+++ b/lib/common/functions.sh
@@ -725,6 +725,9 @@ installPackages() {
                     if [[ $OSVersion -gt 17 ]]; then
                         packages="${packages// libcurl3 / libcurl4 }">>$error_log 2>&1
                     fi
+                    if [[ $OSVersion -ge 22 ]]; then
+                        packages="${packages// libcurl4 / libcurl4t64 }">>$error_log 2>&1
+                    fi
                     if [[ $linuxReleaseName_lower == +(*ubuntu*) && $OSVersion -ge 18 ]]; then
                         # Fix missing universe section for Ubuntu 18.04 LIVE
                         LANG='en_US.UTF-8' LC_ALL='en_US.UTF-8' add-apt-repository -y universe >>$error_log 2>&1

Changing the functions to this code?

I’m hoping to catch as 17-21 worked fine with lubcurl4, but 22 needs libcurl4t64

It’s ugly, I have pushed if you just want to pull from dev-branch

Hi,
The dev-branch worked well for me, thanks!

Ubuntu MATE 24.04
6.8.0-31-generic
All packages fully updated before running the FOG installer (for better or worse, haha)

Thanks again for all the work on this project, it has been great to see it continue on over the years,
Dave

Great, thanks again. I saw an older thread about automatically associating the pending MACs with a host, so you don’t have to open each host and click + to add the wireless MAC.

Is that actually possible and/or could you point me in the right direction? Because it seems like it would be a time-saver…

Thanks,
D.L.

Thanks for the information Tom, I appreciate it. So are you saying that the wireless MAC (pending MACs) are a function of the FOG service? For some reason I thought that these were picked up via PXE during the host registration … but it sounds like these are actually added once Windows boots, via the FOG service over wired ethernet … ?

If it is happening at the FOG service level, then will it pick up the wireless MAC address if I just leave it plugged in to ethernet until it has joined the domain, etc.??

Thank you,
D.L.

Hi all,
If I recall correctly in FOG 0.32 when you registered a laptop via wired ethernet, it would pick up the wireless MAC address - then you would just open that host and it would have that MAC address waiting. Then you would just click the + sign next to it and it would be associated with that host…

In FOG 1.2 I see that you can still hit the + sign and manually add your wireless MAC, but why doesn’t it discover it automatically anymore?

Opening each host and just clicking + on the pending MAC address wasn’t too bad, but manually typing in all of the wireless MAC addresses will take a long time…

If anyone has any advice or information on this it would be greatly appreciated. Is there a way to restore this functionality? Or better yet, automatically associate the pending MAC addresses so you don’t even have to open each host?

Thank you for and advice or help,
D.L.

Hi,
Although it is a hassle, one simple way around this is to re-build a 0.32 box on Ubuntu 10.04, deploy the old images to a host or VM, then re-capture with a new installation of FOG 1.2 with default settings. That will transfer the image to the new FOG server with partclone and that image will work from there on out…

Best regards,
D.L.

If I recall correctly, there were some machines that failed with GRUB, and there were others that failed with EXIT, and I definitely cannot risk that right now. I guess that’s is more of a feature request, but I agree with Gordon that per-workstation or maybe per-group adjustment of this setting would be great!

Thank you for all of your work,
D.L.

I would also like to know about this on a per-workstation basis. I have a few HP dc7900’s that show this behavior, stuck on SANBOOT 0x80 after imaging. I’ve tried updating the BIOS to the latest, etc, and cannot get past this…

The strange part is how most dc7900’s are fine, but I have at least 4 where this happens - perhaps it has to do with the make/model/firmware of the HDD or NIC?

If my memory serves me correctly, in our environment EXIT boot type caused problems on a greater number of systems, and that is a risk that I’d rather not take right now, as it would affect all machines PXE booting to FOG.

Thanks for any advice or help,
D.L.

Hi,
Upgrading to 2074 fixed it, thank you for your assistance, and for the link!

Now I can and will check the SVN’s and code/bug history before posting anything, which is great.

FOG FTW!

Thank you again,
D.L.

I guess that I can just follow the link below, but is there are more chronicled version history online?

[url]https://svn.code.sf.net/p/freeghost/code/trunk/[/url]

Thanks,
D.L.

Okay, I’m on 2062, let me try that… I didn’t realize that there was a newer one.

Where is the SVN version history posted, or do you just request it through the command prompt using Subversion? I’ve looked around a bit but can’t seem to find a chronicle of SVN versions online…

Thanks,
D.L.