You can use socks server for download toorrents. Best choise insert socks traffic to wireguard connection and use sockd for outgoing and clean wireguard + port forwarding for incoming connections.
And you can use i2p network for download torrents in that networks. qBittorrent support it in experimental mode.
I use my home server as media library and cloud gaming device (kvm with sunshine). Also I hosted my friends web sites and some my sites.
hosting my home lab's server on hetzner would have been much more expensive I think.
But in reality, this will only allow you to receive incoming mail. In order for outgoing mail to work, it is necessary that the mail server and all the strapping go through the VPS to the Internet. This requires a rather complicated configuration of iptables, and I recommend that you simply either fill up the mailer on a VPS (there will be a maximum of gigabytes of mail. it's not that heavy), or buy a static address at home.
If you still decide to go the hard way, here's an approximate plan for what you need to do in the spirit of iptables, because setting it up in firewalld is a real torment.:
*mangle
:PREROUTING ACCEPT [0:0]
:INPUT ACCEPT [0:0]
:FORWARD ACCEPT [0:0]
:OUTPUT ACCEPT [0:0]
:POSTROUTING ACCEPT [0:0]
-A OUTPUT -m owner --uid-owner 924 -j MARK --set-mark 0x300
COMMIT
where 924 is the postfix user ID, you may have a different number. check it out
ip route add default via 10.8.12.4 dev wg0 table 100
adding the default route via the VPS address to the routing table 100. replace 10.8.12.4 with the address of your VPS and wg0 with the name of the interface for communication between the VPS and home. Then
ip rule add from all fwmark 0x300 lookup 100
We are sending all packets with the label 0x300 to the routing table 100. In other words, the postfix user will have his own custom routing table via VPS.
This creates several problems due to the fact that with this configuration, it may not be possible to connect to postfix via your server's interfaces. But in basic case all will work. Bypassing this problem will create even more complex routing rules and will generally be overkill. But if you're interested, write to me and I'll sign it.
Well... as I already wrote, my home server is literally on the Internet because I rent a static public IP address from the provider.
But if you have a VPS, then you just need to do port forwarding to your server with a VPS, and then add the following entries to the mx DNS server:
you.domain. 21600 IN MX 10 you.first.vps.
you.domain. 21600 IN MX 20 you.second.vps.
Where 10 and 20 are the server priority Or if the VPS is part of your domain then:
you.domain. 21600 IN MX 10 first.vps.you.domain.
you.domain. 21600 IN MX 20 second.vps.you.domain.
first.vps.you.domain. 21600 IN A 1.1.1.1
second.vps.you.domain. 21600 IN A 2.2.2.2
And if you also have IPv6, you can do
first.vps.you.domain. 21600 IN AAAA fd00::1
second.vps.you.domain. 21600 IN AAAA fd00::2
Where 1.1.1.1, 2.2.2.2, fd00::1 and fd00::2 are the addresses of your VPS
You also need to enter the address in the SPF:
you.domain. 21600 IN TXT "v=spf1 +mx -all"
What does it mean
v=spf1 is the SPF version.
+mx – it is allowed to send mail from the IP addresses specified in the MX records of the domain.
-all – prohibits sending from any other servers (hard refusal).
Also, in order for the signature to work on the mail server, you need to make several TXT entries (for a detailed explanation, see my links about DKIM):
keyname.__domainkey.you.domain. TXT "v=DKIM1; ...%DKIM params%"
and
you.domain. 86400 IN TXT "v=DMARC1...%dmarc params%"
And you need ask you VPS provider set PTR for you VPS IP address with first.vps.you.domain. Or some providers access that config in web panel.
Thanks, I'll give it a try sometime.
On my home server. My ISP gives me a static address and makes PTR records for only about $1.5 per month.
I have been using my own email for many years (to this day). Everything is working great. The main thing is to have a static IP and be able to specify your domain in the PTR record of the ip address.
In general, you will need: postfix (https://wiki.archlinux.org/title/Postfix) OpenDMARC (https://wiki.archlinux.org/title/OpenDMARC) OpenDKIM (https://wiki.archlinux.org/title/OpenDKIM) Dovecot (https://wiki.archlinux.org/title/Dovecot) Some interface to choose from (soGO, roundcube) Maybe graylists, ClamAV, SpamAssassin, or something else to protect your mailbox from spam and viruses. And if you want filtering functionality, then you also need Sieve.
It all depends on the greed of the campaign. I worked in a campaign where it was considered normal to keep a degraded raid without repair. Of course, data loss is a normal story in such companies. The raid guarantees data security only when one disk is being pulled (except for some raids), so it also needs to be monitored and replaced. On the other hand, with proper operation, you probably won't lose any data.
P.S. RAID0 - raid that can't be restored when degraded any disk in RAID. This is exactly worse choice for data save. STRIPE also writes blocks one at a time to the first disk and to the second, so that you would definitely lose exactly 50% of data blocks. Best choice raid10 for performance and raid5 if you need save money.
If you newbie linux user I really recommended create partition for /home and use LVM. That not so easy, but if you understand LVM Snapshots and partitioning that saved many hours for you. You can use partition manager for make /home snapshots and all system snapshots too if you have enough free space in LVM group. The downside of this feature is that you can't take up the entire disk with partitions, otherwise there will be nowhere to take snapshots.
If you want change distro for example:
If you want do momething risky:
ISC DHCP switched to KEA DHCP, They don't have package in Debian repo, but you can add repo and install: https://cloudsmith.io/~isc/repos/kea-3-0/packages/
ISC really deprecated... =( You can install dnsmasq of course, but he is much more slow. But nice for small networks.
Firewalld is much worse for small sustems. Who is really need mark ports? But in difficult cases you need write iptables rich rules anyway. So, as result I love old school with clean iptables without any upperlevel daemons.
as normal user but via systemd service (Linux with systemd system-wide)