Suzune

This is an old PC (Intel i7 3770K) with 2 HDDs (16 TB) attached to onboard SATA3 controller, 16 GB RAM and 1 SSD (120 GB). Nothing special. And it’s quite busy because it’s my home server with a VM and containers.

The question is how do you get a bad performance with ZFS?

I just tried to read a large file and it gave me uncached 280 MB/s from two mirrored HDDs.

The fourth run (obviously cached) gave me over 3.8 GB/s.

There is nothing to refurbish in drives. They are just second hand devices. You can check if they are fine pretty easy and you need to take a look at the age (power on hours). I replace drives at 50k-60k hours, no matter if they are fine.

Most of these observations are subjective. I’ve had some Seagate drives that worked well but were very hot and wasted energy. On the other hand WD was crap so far, starting with 3 TB. Not because of quality, but because of power saving features that were a major annoyance to me (green and some blue drives). Red drives I had were mostly fine, even they wore out pretty quickly (Load_Cycle_Count bugs). They ran at 0% health left for a few years and had other awful SMART and on-drive controller bugs.

Since Seagate and WD are essentially the same company and they lied about SMR before, I wouldn’t buy either of them.

Are you looking for something like cached credentials?

Many manufacturers offer product sheets. You can also use price comparison websites. They sometimes offer an easy way to look at the specs or even compare them side by side.

Some hard drives are built for 24/7 operation. They have higher MTBF ratings and longer guarantees.

Hard drives are very different. Many of them waste energy, lie in the SMART log or just are weird (spin up and down, lose speed, get incredibly hot etc.)

I’ve been self-hosting Postfix for several years and it’s not difficult, if you’re absolutely confident what you do. I don’t recommend it if you don’t know basic behaviors and internals of SMTP and relaying. Also you need to know how to secure your server so you don’t get spammed a lot and getting hammered with brute force attacks.

From time to time you need to react to delivery problems. Most interesting one is perhaps Microsoft, which you need to ask to whitelist your server or your email won’t be accepted.

I recently removed my 25Gbps PCIe dual port cards from my 2 servers because they were using 20W more. My entire rack including 2 UniFi PoE connections uses 90 W now (so 110 W just for having 25 Gpbs).

There is some heat from such cards, but usually it gets transported outside fine. The ones I bought did not come with a fan. I think you cannot operate them without one. The heat sinks get very hot.

Did you use iperf? It makes sure that HDD/SSD is not the bottleneck.

You can also check the statistics and watch for uncommon errors. Or trace the connection with tcpdump.

The only problem is sync across devices. There are not many good options. Maybe it’ll get better with DB support later.

Maybe it’s not what you want to hear, but I was looking these types of apps, too, some time ago.

Then I noticed that most of the apps are useless for me and found an interesting type of apps that implement a “second brain”. There are a few of them, but I chose Logseq.

Any errors on the interfaces? Does tcpdump show something interesting? What do the system logs say?

Logseq is great. It’s still in early development. Only sync is not so great. I use Git and wrote two scripts (pull/push) for Android which I start manually. The desktop application is very powerful and extensible. The app only supports the most common features without any plugin support.

It’s also pretty great if you’re able to code.

I’ve been updating Nextcloud in-place (manually) for multiple major versions without any flaws. What is the problem?