Topics created by arnaudrigole

@arnaudrigole I am glad you found a solution that worked for you, though I’m still interested in trying to at least guess at why the other partition scheme gave problems.

Was there a difference in disk size in the SATA vs nVME SSDs?

Perhaps the nVME SSD is smaller and the OEM partitition at the end of the disk was out of range, though we’d expect a different error for that.

@george1421 Thank you very much for your time George. I’ll try today and back to you asap

@george1421
Hi George,
Thanks for your reply.
I had no infos in the apache logs, so i think you right there was a redirection.
So, i just uninstalled manually fog (mv /var/www/fog to fog.bak & /opt/fog to opt.bak), uninstall mysql with apt-get purge mysql-* and reinstall completly fog from the git dir.
Now it works.
Arnaud

@arnaudrigole No, it doesn’t happen any more. It happened in the past yes.

The GUI can also be used to add “filtered MAC’s” this is simply a comma separated list (e.g. mac1,mac2,mac3,etc…)

The GUI accessible portion to add the filtered mac is:
FOG Configuration Page ->
FOG Settings ->
FOG Client - Host Register ->
FOG_QUICKREG_PENDING_MAC_FILTER

The most important thing to attempt recovery of (other than images) is the ssl directory. If you have fog clients out in the field running, they will absolutely not trust nor obey any fog server that cannot authenticate. The directory is /opt/fog/snapins/ssl
There’s a walkthrough here:
https://wiki.fogproject.org/wiki/index.php?title=FOG_Client#Maintain_Control_Of_Hosts_When_Building_New_Server

If you can do that, and then re-use the old IP, then all the existing clients will report in to the server as pending.

@arnaudrigole Please ensure that if you have STP enabled on this “test” port, you must ensure one of the fast STP functions are enabled either RSTP, Port Fast, or Fast STP (depending on what your building switch maker calls it). Or just disable STP on the “test” port to see if that fixes your problem. I would suggest leaving STP on and making sure one of the fast protocols are enabled.

We have also seen GreenEthernet (802.1az) cause problems with the FOS engine.

Out of curiosity, what make and model is your building switch?

@arnaudrigole Cool. Glad you got it resolved.

I finally found the problem…
After the reboot, the old filezilla server restarted with windows, and bypass the built-in ftp server …

@arnaudrigole As you can see there are no udp-sender processes running. So multicast just cannot happen. Those should be started by the FOGMulticastManager service (which seems fine on your server). So please check the logs as I suggested earlier.

@arnaudrigole That’s one way, albeit incredibly labor intensive.

Since you have a 2008 server, maybe you give a shot at setting a vendor class filter in it? There are some screenshots in the below article from 2008 to help you along your way, but I’ve not been successful yet. That’s not to say it can’t be done, it’s just saying it hasn’t been done yet. I was working on a community member’s production 2008 DHCP server and messed with it for a while but no luck. I’m not entirely sure his environment was even setup correctly…

If you are successful, we will be wanting screen-shots and steps for the Wiki.

@arnaudrigole https://wiki.fogproject.org/wiki/index.php?title=Location_Plugin

This might help:
https://wiki.fogproject.org/wiki/index.php?title=Location_Plugin

@Wayne-Workman said in Image hosting on NFS server, process uncompleted?:

@arnaudrigole I have some methods documented.

using lftp:
lftp -c 'open x.x.x.x; user UserGoesHere PasswordHere; mirror -e /images/TheImagePath /images/TheImagePath; quit'

using NFS:
mount x.x.x.x:/<remote directory> <local directory>
cp -R /mbimages/Optiplex9020UEFI /images/Optiplex9020UEFI
umount <local directory>

Using Samba:
mkdir /tempMount
mount -t cifs //x.x.x.x/ShareNameGoesHere /tempMount -o username=YourUsernameGoesHere -o password=YourPasswordGoesHere,noexec
cp -r /tempMount/ImageNameHere /images/ImageNameHere
umount /tempMount

using SCP:
scp -r /images/ImageNameHere root@x.x.x.x:/images/ImageNameHere

using rsync:
rsync -a /images/ImageNameHere root@x.x.x.x:/images/ImageNameHere

wiki worthy

Added to the wiki here:
https://wiki.fogproject.org/wiki/index.php?title=Migrate_images_manually

@Tom-Elliott
Oh ok.
I updated the kernel to “Official Published Kernels
Kernel - 4.1.2 TomElliott” and it works!
i just don’t see the “TomElliottUSB” on the kernel i selected before…!

SOLVED 🙂 !

@ch3i
Salut !
Ecoute ce matin ça fonctionne en upload/download, je ne sais pas comment.
J’ai toujours le message d’erreur “Device busy” à l’upload mais ça passe outre et ça fonctionne…

@Wayne-Workman
Still have the error, after redefined storage node 😕

@Sebastian-Roth
Yeah, i understand. Trust me i tried to do a lot of times, analyse logs and permissions and work on issues with an unix professor but we don’t found for NFS…
No problem, still working!

@Wayne-Workman

Hi,
I tried to follow this procedure 3 or 4 times, it never worked… 😞
i got blocked when trying to access /images2, IO error everytime!

So i wanna try everything else to export fog storage…

@Quazz ah, yes you are right. I didn’t think about that, enterprise only. And not installed by default. But the FTP test should work from Windows. I would try connecting to the FTP share from the CentOS that should give you an indication as to if the server with NFS is accessible.

@Wayne-Workman
Ok, thanks for your precisions, i’ll work on NFS storage later.