Howdy, Stranger!

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


Down - OVH - SBG - Lots and lots of tears. - Page 12
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.

Down - OVH - SBG - Lots and lots of tears.

191012141519

Comments

  • @happyman said:
    This has had me thinking about my own servers.

    Does anyone know if RBX-2 and RBX-4 are geographically separate or on the same site ?

    Thanks in advance

    OVH RBX-2 59100 Roubaix Data Center is located at 59100 Roubaix, Nord-Pas-de-Calais-Picardie, France

    https://www.datacenters.com/ovh-rbx-2-59100-roubaix

    OVH RBX-4 59100 Roubaix Data Center is located at 59100 Roubaix, Nord-Pas-de-Calais-Picardie, France.

    https://www.datacenters.com/ovh-rbx-4-59100-roubaix

  • Will suppliers then state in their offer that they are fire resistant?

    Thanked by 1lentro
  • MechanicWebMechanicWeb Member, Patron Provider

    @cazrz said: People here blaming customers not having backups, yes its one fault on their end, but that does not excuse the company from the responsibility on the event. I pity the customers but not OVH.

    To make sure I understood what you wrote, you are saying that OVH should have backups of the data that was on those dedicated servers?

  • @MechanicWeb said:

    @cazrz said: People here blaming customers not having backups, yes its one fault on their end, but that does not excuse the company from the responsibility on the event. I pity the customers but not OVH.

    To make sure I understood what you wrote, you are saying that OVH should have backups of the data that was on those dedicated servers?

    Unmanaged dedicated servers*

  • @zafouhar said:
    OVH RBX-2 59100 Roubaix Data Center is located at 59100 Roubaix, Nord-Pas-de-Calais-Picardie, France

    https://www.datacenters.com/ovh-rbx-2-59100-roubaix

    OVH RBX-4 59100 Roubaix Data Center is located at 59100 Roubaix, Nord-Pas-de-Calais-Picardie, France.

    https://www.datacenters.com/ovh-rbx-4-59100-roubaix

    Thanks - Although that website also lists different certifications for the two - if they were at the same site surely they would both have the exact same certifications ??

  • cazrzcazrz Member

    @MechanicWeb said:

    @cazrz said: People here blaming customers not having backups, yes its one fault on their end, but that does not excuse the company from the responsibility on the event. I pity the customers but not OVH.

    To make sure I understood what you wrote, you are saying that OVH should have backups of the data that was on those dedicated servers?

    No, I was saying that the incident/event or what happened is their responsibility. Im not referring about backups or data here, but many people here are just blaming customers not having backups. I am simply referring to the event, not about customers and data. People who owns a building/infrastructure and a company knows this responsibility.

  • @coolice said: What about insurance? This is horrid.

    Generally speaking, the insurance company will need to check whether OVH is liable for the fire. If OVH is clearly at fault for the fire, then no compensation will be payable.
    Even if they do, the premium for the following year will be increased significantly.

    @zafouhar said:

    @MechanicWeb said:

    @cazrz said: People here blaming customers not having backups, yes its one fault on their end, but that does not excuse the company from the responsibility on the event. I pity the customers but not OVH.

    To make sure I understood what you wrote, you are saying that OVH should have backups of the data that was on those dedicated servers?

    Unmanaged dedicated servers*

    It is OVH's responsibility to make servers safe. Fire, Water, thief, power etc.
    No backup is customers' mistake. Fire is OVH's fault.

  • YmpkerYmpker Member
    edited March 2021

    Happy that, although Hetzner also is kind of a budget host, their DC park looks a bit more secure, overall (can't say for sure, ofc):

    Thanked by 2pike jsg
  • MechanicWebMechanicWeb Member, Patron Provider
    edited March 2021

    @cazrz said: Im not referring about backups or data here,

    but many people here are just blaming customers not having backups. I am simply referring to the event, not about customers and data. People who owns a building/infrastructure and a company knows this responsibility.

    Now that's clear, we can agree that it's not the time to point out who is to blame for the backups.

    OVH used a WOODEN FLOOR in their data center! I can't imagine anyone close to having a sane mind would do anything close to that.

    Then it seems they didn't have any effective fire fighting equipment installed.

    And still the wise guys are more interested in free advice on backup!

  • How many of you who are saying all of these things are OVH fault (poor design, poor suppression, etc) are giving up servers in BHS (same container structure) or other OVH sites since they are negligent in your eyes - probably none

  • deankdeank Member, Troll

    Why would I give up my accounts at OVH? I maintain backups. It doesn't matter where you go, things like this can always happen anywhere. Hell, it can happen to your balls if you choose a wrong woman.

    Always maintain backups and you will be fine.

    Thanked by 1vpsGOD
  • Some people in the comments of the FB post where the Statement was published:

    "5 years of work gone overnight. "

    Backups? Anyone?

    Thanked by 2ariq01 Daniel15
  • cazrzcazrz Member

    @Ympker said:
    Some people in the comments of the FB post where the Statement was published:

    "5 years of work gone overnight. "

    Backups? Anyone?

    Maybe they have 3 offsite backups located in SBG1/3/4

  • coldcold Member

    @valk said:

    @cold said:
    lets hope this make the floors of plastik this time ! just in case....

    Plastik? No god please no

    i was ironic bruh !!!

  • @cazrz said:

    @Ympker said:
    Some people in the comments of the FB post where the Statement was published:

    "5 years of work gone overnight. "

    Backups? Anyone?

    Maybe they have 3 offsite backups located in SBG1/3/4

    Could be, but the guy said they had no Backups (at all).

  • deankdeank Member, Troll

    Meh, then his work must have had little importance.

    Thanked by 1sithrebel15
  • @cazrz said:

    @Ympker said:
    Some people in the comments of the FB post where the Statement was published:

    "5 years of work gone overnight. "

    Backups? Anyone?

    Maybe they have 3 offsite backups located in SBG1/3/4

    Offsite backups aren't properly offsite until they're on a different continent.

  • cazrzcazrz Member

    @ahnlak said:

    @cazrz said:

    @Ympker said:
    Some people in the comments of the FB post where the Statement was published:

    "5 years of work gone overnight. "

    Backups? Anyone?

    Maybe they have 3 offsite backups located in SBG1/3/4

    Offsite backups aren't properly offsite until they're on a different continent.

    Is it possible to be on different planet just to be sure?

  • deankdeank Member, Troll

    Mars and Luna are the only options right now. Sucks, I know.

    Pluto location would be nice.

  • yoursunnyyoursunny Member, IPv6 Advocate

    @deank said:
    Mars and Luna are the only options right now. Sucks, I know.

    Pluto location would be nice.

    summer host is planning an expansion to Jupiter. It'll open in next few years.
    It'll have enterprise grade CPU and storage, unlikely the crappy ARMv5 and CF cards in Antarctica.
    RTT would be 90 minutes or so, so you'll have to be patient.

    Thanked by 1webcraft
  • Make sure all of you add lowend DC engineers and LowEnd Firesuppression experts to your Linked in Skills. Its amazing the personal growth in skills and knowledge some of you have had. In fact, you can be thousands of miles away and 2 days experience and make all the conclusions fire investigators on site can't make, but they only have years of experience not guided by lowendgeniuses

    Thanked by 1Boogeyman
  • @cazrz said:
    Maybe they have 3 offsite backups located in SBG1/3/4

    Then they had missed the point about single points of failure. If you are careful enough to have several off-server backups then surely you are aware enough that having at least one of them with a different provider is a good idea? Even if SBG1-thru-4 were all in different cities, the company is still potentially a single point of failure.

  • @MechanicWeb said:
    And still the wise guys are more interested in free advice on backup!

    Who/what is/n't to blame for the fire does not change that a proper backup strategy would mean no data was lost (or at least would make it far more unlikely data is lost, cascade failures are a thing). If you have no good backups for data at any DC you take this risk, so the advice on backups is relevant to help prevent repeat losses. It is particularly relevant as, as mentioned in posts above, some who apparently had no backups are complaining that they've lost years of stuff.

    If your data is that valuable you take care of it. If you want your provider to manage and be responsible for backups pay them to (though I prefer to backup off-provider to reduce single points of failure). If you want your data in better DCs, perhaps look into the DCs before using them instead of just looking at the price and server specs.

    If you don't want to listen to us "wise guys", feel free to not listen. But be aware that we'll enjoy saying "we told you so" if you are affected by a problem like this in future.

  • FalzoFalzo Member

    @MeAtExampleDotCom said: single points of failure

    to be fair SPOF is a concept that always relates to a scale.
    - if you look at a harddisk, having two removes SPOF.
    - if you look at a single server, having two removes SPOF.
    - if you look at a whole DC, having two removes SPOF.
    - if you look at the world, having two... wait.

    maybe we should blame providers, if they don't account for earth extinction. it's a single point of failure ;-) ;-) ;-)

    Thanked by 2yoursunny Saahib
  • cazrzcazrz Member

    @MeAtExampleDotCom said:

    @cazrz said:
    Maybe they have 3 offsite backups located in SBG1/3/4

    Then they had missed the point about single points of failure. If you are careful enough to have several off-server backups then surely you are aware enough that having at least one of them with a different provider is a good idea? Even if SBG1-thru-4 were all in different cities, the company is still potentially a single point of failure.

    Well there's Scaleway 25 meters underground data bunker in case there's more chaotic disaster than storm and fire. Backblaze in Phoenix, Cali and Amsterdam. BuyVM in Lux, LV and NYC. TransIP bigdata dual copies in Amsterdam/Delft.

  • xaocxaoc Member

    I keep my backups in Uranus. ;)

  • xaocxaoc Member
    edited March 2021

    Comment by OVH - Thursday, 11 March 2021, 11:52AM
    The first cleaning and securing works of the datacenter have started this morning.
    Our teams are building a document summarizing the impacted services for each building. This document aims to give you visibility on the recovery status of your services and data in the coming days.
    A FAQ will be available soon to determine the location of your products and answer your questions.

    http://travaux.ovh.net/?do=details&id=49484

  • Anyone hosting in OVH SBG after 2017 is either ignorant or reckless. Here's what they said about a multi-hour outage in November 2017:

    http://travaux.ovh.net/?do=details&id=28247#comment35250

    Emphasis should be given to

    SBG's power grid inherited all the design flaws that were the result of the small ambitions initially expected for that location.

    In other words, they built a small DC on the cheap, then they tried to ride the wave until it broke, and they were lucky that it was for a few hours only.

    It's impossible for an outsider to say if they kept their promise to improve the electrical installation, or even if the electrical installation was at the source of yesterday's fire. The attitude alone made me write off hosting anything in OVH SBG, and left a bad feeling about the other OVH DC as well.

    Thanked by 2OhJohn Daniel15
  • xaocxaoc Member

    @quicksilver03 anyone not hosting in OVH SBG is either ugly or smells. ;)

  • MakenaiMakenai Member
    edited March 2021

    @quicksilver03 said:
    Anyone hosting in OVH SBG after 2017 is either ignorant or reckless. Here's what they said about a multi-hour outage in November 2017:

    http://travaux.ovh.net/?do=details&id=28247#comment35250

    Emphasis should be given to

    SBG's power grid inherited all the design flaws that were the result of the small ambitions initially expected for that location.

    In other words, they built a small DC on the cheap, then they tried to ride the wave until it broke, and they were lucky that it was for a few hours only.

    It's impossible for an outsider to say if they kept their promise to improve the electrical installation, or even if the electrical installation was at the source of yesterday's fire. The attitude alone made me write off hosting anything in OVH SBG, and left a bad feeling about the other OVH DC as well.

    In my case there actually was no option to select a DC - just the country when buying through SYS. Only after the server is delivered you get to see the DC (SBG2 in my case :smile: ). At this point you've already paid.

    What are you supposed to do at this point? Ask for a refund and try the lotto again?

Sign In or Register to comment.