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
How do you even achieve this? Having two servers with same files+config? so that we can manually change dns record in case on fails?
Or maybe its something that can be fixed using load balancers?
Yes. You don't have to manually change DNS though. You can write a monitor script and change the DNS settings through name servers API.
You can also use smart DNS services like "Azure Traffic Manager" to achieve auto fail-over.
123system, almost a decade
It depends on what kind of server it is of course, but yes, dns and loadbalancers are common ways of dealing with it.
Do both. And you don't have to run load balancers like haproxy or maxscale on a separated vps. Install it on each of your production server and put your all productions under each. And then point your DNS to each haproxy.
Poor Conceptboard. It looks like the dns for their eu side are all hosted in that dc.
https://conceptboard.statuspal.eu/incidents/164060
When OVH DC burnt down
Wrong. When you are selling three nines or above, it is unacceptable. 99.9% equates to 8 hours and 46 minutes downtime a year. Five nines is no more than five and a half minutes downtime a year.
I think LowEndHosts really do exaggerate their stability and availability which is often overlooked. Claims are made that they are on par with HA providers, some even going as far as to say they're cheaper. A lack of SLAs and it's clear why certain customers would never touch them.
Sounds like my racknerd service is back since 37"
Better weekend ever
And not one customer has complained ... Surprised I am
Racknerd France 100 hours of downtime - longest I can recall for me with any host
https://app.screencast.com/ZIdr6080g9W1C
One time our upstream rebooted their router for scheduled maintenance, we lost $300,000 during the 3 minute outage.
This. I thought it would be mentioned much earlier in this thread
Hi @tdwuk -- Most clients should have seen affected services come back online throughout the past 24 hours, and as of a few hours ago, everything in Strasbourg should be 100% back online from an infrastructure standpoint.
To recap and out of transparency, this was a larger facility-wide issue at the DataDock SXB facility, which was beyond our immediate control. On the morning of March 21, GoDaddy's section on the first floor of DataDock SXB experienced a small fire in a battery-backup room, which triggered/caused a water leak. The fire was quickly contained, and all of RackNerd's equipment was free of water damage. To avoid additional damage and to comply within the limits of local authorities, power was then subsequently cut for the first floor, which impacted roughly ~30% of RackNerd's infrastructure (as the majority of our infra is on the second floor).
We took proactive measures within our power to restore affected services as quickly as possible. Our datacenter partner has successfully relocated our affected equipment from the impacted 1st floor to the unaffected 2nd floor, avoiding the need to wait for the facility to resolve the issues with the 1st floor (they worked hard to move racks of equipment, not only for us but many others). Every step of the way we were committed to keeping our customers in the loop with our progress, as well as sharing frequent updates via the status incident page. Ultimately, this situation could of played out a lot worse but we're glad that it's all behind us now.
At this point, we are showing everything to be back online, so if your service in France is still offline, it may be an isolated issue. If anyone is still facing issues please open a support ticket and we can investigate/troubleshoot accordingly
Did anyone get any refund? I was using the host for barely 1month+ and this happened. Moved my sites to hetzner. I asked the support for partial refund, but rejected with a response that they will compensate "later".
Hi @hassancent - to add clarity by default RackNerd doesn't provide refunds, as outlined within our terms of service. So if you're requesting a full refund, our team is trained to deny these requests. Though we work closely with our customers to ensure we help solve outstanding technical issues they may be facing.
From what you're commenting, it does sound like terrible timing (as a new customer of RackNerd) and the duration of an unexpected outage.
We are most known for our solid service, backed by solid earned reputation for the service level we provide. This event was rough for everyone, and not something you should expect of RackNerd.
We're still working with our upstream and gathering our side of things to issue SLA compensation along with the RFO.
Cool. Should have read t&c, that's my fault. btw i asked partial refund.
Well I wouldn't use racknerd for even hobby projects after what happened. I was expecting max 8-10 hours of downtime, maybe once a month. This was way to much.
I know this incident sounds like it might not have been in hands of racknerd, But seems there was no backup plan in mind if something like this happens.
Regardless of the provider, you are responsible for your own backup plan. Unless you're paying extra for management and an SLA it's always going to be 100% on you.
If you are doing anything important you should always host your DNS servers on two continents (or at-least east/west) and two different external (not you) providers.
Its also one of the few things that even if you can, you shouldnt put on your own network. Or any single network.
Yes.
All i meant was that this incident made racknerd not even worth being considered as backup server.
When you claim to have companies like Amazon, Walmart, HP and FedEx as customers you do not host everything on a single Racknerd vps. No offence to Racknerd but they are a lowend provider, nobody with a brain would host something vital there without backup. Someone is either lying or is stupid as fuck!
Also, spinning up a new dns somewhere else and point the domain to that should take 30 minutes, max. Instead they moved everything from .com to .eu, there is no way that could have been faster, cheaper or easier.
I'm pretty sure some still hope that @cociu will come back 🤣
OVH SBG, my dedi even never come back
Service down again today for approx. 2hrs so far. Not ideal. I had to put in a cancellation request. It's not worth the headache. Luckily I only had one landing page company site on the vps. Moving that asap.
Same Strasbourg datacentre, ironically I am paying extra to host there
All's fine for me now with Racknerd
Here are the last data:
Don't hope too much, after a ticket I had a refund of $8 for a $36 annual plan,
I've learned how his business could be profitable now :->
All's well that ends well, and you handled the situation very well. Nobody is immune from a data center incident, nobody.
I'm delighted to renew my confidence in you.
Hi @tdwuk -- We sincerely apologize for any inconvenience. There was a routing issue that impacted several nodes in this location earlier this morning, which has since been resolved. As of now service in France should be restored. To recap everything, a more detailed email will be going out to all affected customers next week (including compensation details), once we receive an official RFO from our datacenter provider.
While most customers should be back online, and everything is online from a power perspective, it's inevitable that some isolated issues may still occur after a broader outage incident like this -- our team is still working with customers to resolve any isolated cases.
Although we fully expect our France datacenter operations to stabilize back to how it was pre-incident once these individual cases are finished being sorted out, we understand if some customers would have better peace of mind with a migration to another location all together. Therefore, we are providing our France-based clients the option to migrate their services free of charge to any of our other available VPS datacenter locations, including our other European location, Amsterdam. @tdwuk -- if interested, please submit a support ticket, and our team will assist you with the migration process accordingly.
Hi @commercial -- Thank You so much for your understanding -- it is true that not even the best of us can be immune from any sort of incidents in the world of IT/technology. We always strive to address issues transparently, learn from them, and continuously improve our processes and infrastructure resilience - while most importantly keeping our customers in constant communication/in the loop.
We look forward to working together for many more years to come 👊 Thank You again for the understanding and renewed confidence.
I mean, it's kinda still happening
(45.x subnet)
Outside of that, and seeing that you offer migrations to AMS, any plans on providing that location for the yearly deals? even if it costs extra?
(just wondering, would support allow me to buy a similar server that I have in FR but in NL instead?)