Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!


HAZI.ro | Maintenance in progress (19 Aug 2023)
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.

HAZI.ro | Maintenance in progress (19 Aug 2023)

FlorinMarianFlorinMarian Member, Host Rep
edited August 2023 in Outages

Hello!

Since the temperature of the CPUs on the storage node exceeds 65 degrees Celsius after switching from 'low consumption' to 'performance' mode, we decided to stop all VPS servers for a few tens of minutes to replace the thermal paste of the processors of that server.
Since the storage node is used on all VPS servers, even pure SSD servers will be stopped because the configuration for the cloud-init feed is still on the storage server, regardless of the storage medium and we want to avoid any data corruption.

Thank you for your understanding and apologize in advance for any inconvenience caused by this extraordinary maintenance.

P.S. In a negative scenario, the server would have shut down by itself due to the high temperature and there could have been corrupt data at the VPS level for the data that was not yet written to the disk but only in the cache memory.

Thanked by 1yoursunny

Comments

  • This is the third time my VPS has gone offline within a week. Please cancel my order and refund my money !! Order Number: 6116979130

  • KrisKris Member
    edited August 2023

    Did you not stress test these machines before putting them into production?

    In the future I'd suggest you upgrade the BIOS, make sure it's in performance mode, and use stress-ng for a few days before putting them into production, this is a bit of a rookie mistake.

    Hoping it's the end of you trashing Calin as a less than stellar setup than yours.

    Kinda both par for the course, expect you're* selling at normal datacenter pricing.

  • zGatozGato Member
    edited August 2023

    @Kris said:
    Did you not stress test these machines before putting them into production?

    In the future I'd suggest you upgrade the BIOS, make sure it's in performance mode, and use stress-ng for a few days before putting them into production, this is a bit of a rookie mistake.

    Hoping it's the end of you trashing Calin as a less than stellar setup than yours.

    Kinda both par for the course, expect your'e selling at normal datacenter pricing.

    Calin has been running his DC for a while now, and he experienced all the issues Florin is going to start to now. Even Calin is going to start Colocation on a real DC for some of his dedis.

  • KrisKris Member

    This is just sort of basic setup for servers, despite where you host them.

    Also what gives with the single point of failure for every machine on one storage node.

    Not feeling that ChatGPT generated + picture quality with oversights like not putting a server into performance mode and stress testing it pre-production, especially if it's a SPOF for every machine you host.

    That's just my opinion.

    Thanked by 1yoursunny
  • yoursunnyyoursunny Member, IPv6 Advocate
    1. Migration in progress
    2. Maintenance in progress
    3. Unsustainable sale in progress
    4. Involucration in progress
  • 200 score.
    Let me put performance settings
    700+
    Bad airflow, fans going crazy 65+ degree
    Big mistake
    Shut down
    Change paste
    Open post 1st before clients notice

    Alt leat cociu went radio silence, you are still projecting your incompeten're...

    Thanked by 1yoursunny
Sign In or Register to comment.