Strange Version of Nethsecurity

NethSecurity
1

Hi

Today customer call me cause not having internet connection.

I’m not able to connect to the Nethsecurity VM running on Proxmox that is what i see on the console if i try to do any input doesn’t receive anything.

image
image1035×649 86.4 KB

So i “Stop” the VM and power-on again and after that the WebUI is responding, but in the Dashboard i see a strange version:

image
image530×141 12.5 KB

Now i Download the image of system from the official nethsecurity website and upload then install and reboot now the version seems ok:

image

Waiting your response about it.

Thanks.

2

That’s really strange. Could it be that there was a testing version installed initially and it worked without error until today?

Is there a subscription on that device?

I checked my NethSecs and with the stable version 8.7.1 there’s no update shown.

3

Hi

I’m prettu sure that there is not a testing version, idk what is the reason behind that, i checked to for comparison to another customer firewall and the latest stable version is the 8.7.1.

Wait for the developers maybe he has some idea why this is happen.

4

That is 100% a pre-release version, nothing to add there :man_shrugging:

1 Like
5

Hi

Idk why had a version of pre-release, anyway now i have the latest stable.

Thanks.

6

Hi, today morning already three customers they called me because they don’t have a network/internet both of them the firewall run on Proxmox VM, and they’re all stuck on this screen.

image
image1035×641 68.1 KB

After the reboot i see that one of them had this version(see screenshoot below) and the other: 8 24.10.0-ns.1.5.1 i update all of at the latest version.

image
image542×144 7.77 KB

The only thing I noticed in common that these firewalls have is that they are all virtualized on proxmox

Maybe is occurs only when the nethsecurity is virtualized and running on Proxmox? I update the Proxmox of my customer at the latest version and kernel about 1 month ago, so if it depended on that strange one that only gives the problem now.

7

AFAIK we talked about an update script using sysupgrade some time ago. Could it be the issue?

8

We only talking :slight_smile: never implemented.

1 Like
9

It does not seems to be stuck, just hit Enter as described in the log and you get access to console to do more debugging.

1 Like
10

Hi

I try to press enter on the Proxmox Console of the VM but not take any input.

11

Hi guys today the problem back again all the customer have nethsecurity firewall virtualized on proxmox has the VM totally blocked/freeze.
I have noticed that i have one customer that they have nethsecurity running on a VM on Proxmox too but instead of ZFS they have LVM(default proxmox installation) and they don’t have the problem, so it seems that is some bug releated to proxmoxvm+zfs.

As i said above i can hit Enter on the console nothing appears is totally blocked, which LOG i can look at for understand which is the possible cause?

UPDATE:
From the Beszel Monitoring i see all of them except one goes down today at 12:22-12:23 am except one at 1:22am

In /var/log i don’t see log for previous boot but in “messages” only for the current boot.

image
image1051×105 27.3 KB

Thanks

12

You could setup a persistent storage, see Storage — NethSecurity documentation

13

I’m migrate the virtualization firewall to physical

14

using Proxmox virtualization with display = serial terminal and serial port = socket

should work

image
image1920×1080 245 KB

1 Like
15

Thanks for the tips, next time if the problem still i try with this settings.

16

Off topic, but did you upgrade that cluster from 8 to 9 and did you have any issues with any nodes upgrading?

17

not a single problem on all 9 nodes in the cluster and 6 more servers not in a cluster

looking in to bash history - yes i had to remove for all nodes a iSCSI storage (ZFS over iSCSI)

and uninstall a third party package proxmox-truenas (should never use it in first place)

now I’m using only standard NFS in addition to Ceph

this is all what i found in the history (i had already the latest ceph version):

apt update; apt full-upgrade -y; apt autoremove -y
pve8to9 --full # just info about any possible problems
sed -i ‘s/bookworm/trixie/g’ /etc/apt/sources.list
sed -i ‘s/bookworm/trixie/g’ /etc/apt/sources.list.d/pve-enterprise.list
sed -i ‘s/bookworm/trixie/g’ /etc/apt/sources.list.d/ceph.list
apt update; apt full-upgrade -y; apt autoremove -y
echo ‘grub-efi-amd64 grub2/force_efi_extra_removable boolean true’ | debconf-set-selections -v -u
apt install --reinstall grub-efi-amd64
apt modernize-sources

recently i had to reinstall some of the nodes

and it took maybe less time to remove a node from the cluster,
clean install Proxmox and join again the cluster
the only thing you have to setup again is the network
and I did save a copy of the file…
and if you reinstall a node the ceph rebalance is increasing the disk weareout

1 Like