New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
Online.net and scaleway are for running attack or sending spam or torrenting ... Not for production because the support is really bad (but not as bad as OVH support), nearly as bad as their bandwidth outside of France.
Just move to hetzner and enjoy
People who like having available memory
Though I never really tried to tweak it much admittedly, might look better with tweaking.
Yes, does look a bit poor on here actually.
https://www.youtube.com/embed/EwAL7dFEUtw?start=94&end=103&version=3
I stopped using Online.net for production few years already.
NIC dead, I asked them to replace the NIC.
"No, you will need to reinstall everything on a new server."
You are a vi edit+ an initramfs rebuild + a reboot away from a 4GB ARC.
I can hardly tell my 2xHDD E3 is running on spinning rust. (just kidding, but it's running good for 3.5 yrs ; 3 proxmox major upgrades included)
I'm assuming they were using built in Ethernet and don't have any available slots to put a new NIC.
Take the disks to a new server, it's done
Raid is not disaster recovery but what it is.
Is a Hard drive failure recover in place toool.
It's good for if a drive fails then you can keep operating and swap out the bad drive as a hot swap.
So imo if it's a production server you need a backups and to do raid. Raid 10,5,1 etc something where if a single or more drives fail then you can keep operating and swap that baby out.
This is the singular purpose....
Not counting the game drives of 10 wd blacks or raptors......lol.
my approach is a bit different I prefer to set smart monitoring tools and act before my drives fail. along with offsite backups.
RAID controllers can fail (a funny thing I'm not aware of raid controllers monitoring tools), software RAID uses extra CPU for syncing especially in large arrays. I still think its a marketing gimmick or a luck tool for a lazy sysadmin who knows nothing about proactive monitoring !!
There are times where a drive can spontaneously fail. There’s a reason why large businesses and many corporations still use raid.
The reality is you can not expect nor act proactively on all drive failures as at any time you can have random failures.
Raid is not a lazy sysadmin tool cause they aren’t proactively monitoring Their drives. Many corporations replace drives before they fail.
A poor sysadmin will have no raid and think only proactive monitoring will be enough for a random drive failure when there is no indicators.
When keeping data intact, active and functioning is required you use raid so even random drive failures that aren’t able to be predicted do not harm the buisness
I made the mistake of uploading 3TB of data to C14 and setting lifecycle rules to archive it only to find they can't actually archive it in the the timeframe they specify and there's a hidden FUP which limits the number of objects per day that get archived to the cold storage.
I did actually manage to get a credit out of them for the unexpected time in hot storage.
But when it first came out of beta the C14 cold storage was nowhere near fit for purpose.
Only reason the data is still on it is I can't be bothered to move it at the moment.
Oh and any API keys you generate for object storage have FULL control of your account.
If you have a choice between online's object storage and backblaze/wasabi take the latter as either of them will likely lead to a better experience.
I have literally a similar terrible experience with them. Just posted.