A SWAP file?
On ZFS?
Can have issues…
Better would be:
During Install, when chossing ZFS mirror, remove 8 GB from the usage from each ZFS used disk. 2 Disks, as in a mirror, would leave you with 16 GB swap, more than enough.
Here on my Home Proxmox LAB, this one running on a ODroid H3+ with 64 GB RAM:
That’s good to know… 
So far the swap issue hasn’t affect me. Only a Samba container failing. If only I knew how to restart the samba container.
The cavallery is needed:
In NS7 that was:
systemctl restart nsdc
Hi
I lost the link to the NSDC restart in NS8, do you have that command handy?
My 2 cents
Andy
:-} Already tried that…
Failed to restart nsdc.service: Unit nsdc.service not found.
Do you have access to /cluster-admin/ ?
Can that application be restarted there (three dots)?
Hi Andy,
Unfortunately not. It seems it’s one service they don’t want us novices playing with… I think it would be great if I could run a restart on it… for logging purposes to see exactly why it’s failing on reboot.
I think more an overisght in admin interface design…
BTW, out of curosity (And Time Zone issues):
Where are you located?
I’m in northern Switzerland, within walking distance to a german beer if it was daytime!
LOL… I’m in New Zealand. I’m wondering if I should change the NTP servers in Rocky Linux to pint (point, now beer is on my mind) here. Maybe it’ll help if it realises it’s upside down.
North Island, eg Auckland?
Andy wash your mouth out with soap I’m from the South Island… Auckland is another Island (and country) as far as us southerners are concerned.
No insult to southerners intended!
I have heloed out 3 users from the northern island, but you’re the second from the southern isle!
Thanks Andy. Hopefully I’ll get an an answer regarding this error. I’ve got my backup’s running now… and hopefully will be home before midnight.
As always, when it’s an IT issue: Good Luck!
And no, even the best know-how can’t compensate major hardware failures!
My 2 cents
Andy
Do you use PBS yet?
If not, you’re missing out major!
It’s fixed!!!! I ran a systemctl restart chronyd.service on the proxmox >_ Console right after rebooting… I didn’t wait around, logged straight in and changed before system had a chance of logging in Samba… Checked the status, and saw time was still out, ran systemctl restart chronyd.service again and checked. Time was correct. THEN I tried logging into the domain cluster. It’s working. It must be a time issue which is strange as the Samba container is on the same machine as the rest of the containers. A container must be running it’s own NTP and not updating the main cluster clock.
My “Time Server” Setup:
I get a Stratum1 time server feeding mine (Stratum2). All local hosts use my OPNsense as master NTP service.
Swiss are known as clock builders…
Proxmox, but also NethServer are somewhat lacking in this respect…
I’ve installed pfSense… I was wondering what OPN looked like… I may flip over to it one day. pfSense has a few oddities I’m getting used to. I was spoilt by the old NS7 system.
pfSense - they really made a bad show when OPNsense was forked…
And: The provided Source Code won’t compile, due to miissing BLOBs in the Repo.
But everywhere on their page it says OpenSource.
Ask twice, and youre locked out of their Forums!
I can help with OPNsense. It also works great as a VM…
Large parts of the config file can be re-edited and reused on the other platform…
I started with m0n0wall, pfSense forked from m0n0wall, then OPNsense forked from pfSense.
I. moved directly to OPNsense, as the creator of m0n0wall suggested…
Never regretted it!
Sounds like a plan… I’ll look at installing OPNsense later this month. I’ve had no fortune trying to get wireguard working on pfSense… I’m sure it’s just an out bound tunnel issue. pfSense didn’t handle my rule for forwarding port 25 to a dedicated gateway, until I told it to forward everything from the NS8 VM to the specified gateway, and then I modified it back to port 25 only. Then pfSense behaved itself. Multi-WAN’s are fun.
