Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The Synology End Game

lowendbox.com

Post reply on HN
Print view View original

331–340 of 360 posts

Re: The Synology End Game

#331

Post by lostlogin »

Hamuko wrote: post #125
lostlogin wrote: post #116

Earlier quoted context omitted.

I graduated from Synology docker, to Docker compose and my world got better. The weird quirks of Synology Docker are painful. Eg containers that won’t stop, or won’t start. It’s not easy to get into the containers (docker exec), recreating is tricky compared to copying and pasting docker-compose.yml.

Synology Docker aka Container Manager has support for Docker Compose under "Projects". It even explicitly gives you the option or either uploading a ready-made docker-compose.yml or to create a new one. Personally, I mainly use the CLI to manage my Compose files even on Synology DSM.

I’m surprised that I haven’t noticed this before. When I click on it it still isn’t obvious that this is what it does. It’s more than slightly embarrassing.

Thank you.

Re: The Synology End Game

#332

Post by dostick »

How ugly customer-hating synology is: Synology DSM os measures free space in percent. So when on your multi terabyte volume 200GB free remaining, you would receive notifications until it’s about 100GB free and then you will not receive notifications even it it runs out of space completely.

You can set the free disk space threshold percentage but it will accept number no lower than 5%. This is obviously a rudiment from old times, 20 years ago when volumes were 1-10GB, not TB.

I asked them to change threshold from percent to GB, over multiple times over few years - not going to.

I asked them to allow change threshold to lower value than 5% - not interested.

Not only you can’t know when disk drive runs out of space, but excessive notifications when there’s still lot of free space desensitises, so when actual disk failure happens you would ignore notification.

Re: The Synology End Game

#333

Post by robmsmt »

I’ve been thinking more and more with synology’s recent changes that it would open up the market for some new entrants that are like the system76 of NAS that are:

- low power cpu (arm?)

- uses latest kernel and flavor of Linux distributions

- makes own modular hardware in smallish form factor

- basic support package with slower SLAs to more professional SLAs for enterprises, I would happily pay per month to prevent enshitification if I had guaranteed support and even more so if it bundled some zero trust encrypted cloud storage.

Re: The Synology End Game

#334

Post by procaryote »

M95D wrote: post #232
procaryote wrote: post #216

Earlier quoted context omitted.

There are pci boards that let you hook up 4 sata ports to the pci3 on a raspberry pi 5. The drives will be a large part of the power draw, so if you want low power, going for 2 drives is probably better. That probably gets you into the 20-30 watt range for 10+ sata ports you might as well get an x86 motherboard as it's going to draw lots of power anyway Unless you plan to power down most of the drives most of the tim…

> Unless you plan to power down most of the drives most of the time I do. Read my other comments.

Yeah, it looks like you have somewhat niche requirements and would need to build your own hardware and software.

To run 10+ sata drives you'll either need to take a lot of care they're not spinning up at the same time, or getting used in parallel, or you'll need to dimension your power supply to cope. A beefy power supply will have a higher idle draw making the focus on getting the whole system idle down to 5w pretty hard

Re: The Synology End Game

#335

Post by procaryote »

whalesalad wrote: post #253
procaryote wrote: post #216

Earlier quoted context omitted.

There are pci boards that let you hook up 4 sata ports to the pci3 on a raspberry pi 5. The drives will be a large part of the power draw, so if you want low power, going for 2 drives is probably better. That probably gets you into the 20-30 watt range for 10+ sata ports you might as well get an x86 motherboard as it's going to draw lots of power anyway Unless you plan to power down most of the drives most of the tim…

rpi sux for longevity - be prepared to mirror and replace the SD card often

I've never had this be an issue. If it were an issue, you could probably put just /boot on the sd card and run the system off one of the real disks

Re: The Synology End Game

#336

Post by HighGoldstein »

Shank wrote: post #72
tecleandor wrote: post #48

Not only that, but their security situation is terrible. Their OS is full of EOL'ed stuff. On products you can buy TODAY, you find: - Their Btrfs filesystem is a fork of a very old branch and doesn't have modern patches - A custom, non standard, self built, ACL system for the filesystem - Kernel 4.4 - PHP 7.4 (requirement for their Hyperbackup app) - smbd 4.15 - PostgreSQL 11.11 - smbd 8.2p1 - Redis 6.2.8 - ... They…

The encryption is also broken. If you use encrypted shared folders, you have an arbitrary filename limit ( https://kb.synology.com/en-ro/DSM/tutorial/File_folder_path_... ). If you use volume encryption, your encryption key is stored on the NAS itself, which is capable of decrypting the data, unless you buy a second Synology NAS ( https://blog.elcomsoft.com/2023/06/volume-encryption-in-syno... ) to act as a key vault…

> but in what world would the drives, and not the NAS itself, be stolen?

Not to defend Synology, but popping a drive out of the NAS so that it won't be noticed (or noticed much later) is a much easier way to steal data than carrying off the whole NAS. I assume they're guarding against the kind of scenario where an employee steals steals drives rather than ski-masked thieves breaching the office and making off with the NAS.

Re: The Synology End Game

#337

Post by wtcactus »

Tajnymag wrote: post #245
wtcactus wrote: post #182

I’ve bought an Ugreen NAS right when they came out. Installed TrueNAS and it’s been rock solid and with a lot of hardware features Synology doesn’t offer all this time. There’s really not a lot of reasons to use Synology anymore (only thing I miss was the sync solution they had. It was indeed better than Syncthing and the likes).

Does UGreen support ECC memory? As far as I've read online, it's crucial for Truenas.

In part that’s an interesting question. Some of the processors on the Ugreen line are Xeons, so they should support ECC, but I’m not sure the MB does.

But you don’t really need ECC for ZFS. That’s a kind of myth that has been debunked several times. Sure, it’s always good, but ZFS works just fine without it.

(I’m on my phone so it’s dificult to find a good source for the claim, but if you search a bit, you’ll find some good blog posts talking about it).

Re: The Synology End Game

#338

Post by Tajnymag »

wtcactus wrote: post #337
Tajnymag wrote: post #245

Earlier quoted context omitted.

Does UGreen support ECC memory? As far as I've read online, it's crucial for Truenas.

In part that’s an interesting question. Some of the processors on the Ugreen line are Xeons, so they should support ECC, but I’m not sure the MB does. But you don’t really need ECC for ZFS. That’s a kind of myth that has been debunked several times. Sure, it’s always good, but ZFS works just fine without it. (I’m on my phone so it’s dificult to find a good source for the claim, but if you search a bit, you’ll find so…

Well, that's the thing. I've also found some posts saying it's a myth and you'd be fine, but the ratio of posts recommending ECC to posts debunking the myth is like 10:1.

Re: The Synology End Game

#339

Post by wtcactus »

Tajnymag wrote: post #338
wtcactus wrote: post #337

Earlier quoted context omitted.

In part that’s an interesting question. Some of the processors on the Ugreen line are Xeons, so they should support ECC, but I’m not sure the MB does. But you don’t really need ECC for ZFS. That’s a kind of myth that has been debunked several times. Sure, it’s always good, but ZFS works just fine without it. (I’m on my phone so it’s dificult to find a good source for the claim, but if you search a bit, you’ll find so…

Well, that's the thing. I've also found some posts saying it's a myth and you'd be fine, but the ratio of posts recommending ECC to posts debunking the myth is like 10:1.

The important key point is that you will not lose anything by using ZFS without ECC compared to using any other file system without ECC. The myth is that you specifically need ECC if you’ll be using ZFS, not that ECC is a really good idea for any file system.

https://jrs-s.net/2015/02/03/will-zfs-and-non-ecc-ram-kill-y...

Post reply on HN