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
negative, last delivery i had in FRA was HDD with no upgrades
Hit a 1650v4, 64GB RAM, 2x 1.92TB NVMe in GRA.
If you’re looking for stock LE-Es with HDD in GRA, I have 3 that I’m looking to get rid of at cost. I gambled a bit hoping for bigger drives but didn’t win.
FRA keeps delivering hdd zero upgrades only v6 cpu
FRA was a big disappointment this time around compared to last year
Thought its just me.
Hope is fading away soon
But it will be fulfilled by the cancelled non lottery orders on Monday™
Daddy betrayed 💔 😢 😔 😞 😪
Newbie here. Can someone tell me the time of the day Kimsufis are released? I got my KS-C a few months back when they released a couple of them in the early morning UTC. The stocks were over by the time I finished checkout and rechecked.
Looking for an upgrade.
Randomly as stock becomes available
I improved this more with claude and gemini
, let me know what you think:
https://github.com/loayai/yet-another-hardware-info/
Basic usage:
curl -sL https://raw.githubusercontent.com/loayai/yet-another-hardware-info/master/yahi.sh | sudo bash
Last GRA deployment on Friday evening, after 10 minutes. On Thursday a friend got an 1200G/64G/1270 v6 config. I payd yesterday a LE-B NVMe in GRA but still no delivery.
1--2 delivery will be happen but I think that indirect cancelations will be started soon. They might promise long deployment time and offers the moneyback.
did you got any servers during the weekend?
They mentioned to wait 15 days for my BHS. but they did that in 3.5 days. So don't lose hope.
Anyone here know the steps to fix the automatic renewal? I think it's a case of where 1 of the components is marked as manual renewal, but I cant remember the steps to change it
https://lowendtalk.com/discussion/191096/how-to-fix-ovhs-broken-renewal-system-since-they-don-t-give-a-shit
Bloody eye (pun intended) made me miss all the B-rated fun in doctors office, scripts failed of course, so I had to build my own chicken from sticks and mud (EPYC 7551P, 256G ram, 2× Tesla P40, 2Tb NVME)
What a nice water cooling loop with P40s.. that loop might cost more than 2xP40.
$7 / year?
https://github.com/adns44/ovh-renew-manager
It's priceless in terms of fun labour: I had to use drill with abrasive disk to file off few millimeters of copper block (and that's not the first thing I tried to fit it in), because it didn't fit though the cut in original cover plate. And I didn't like idea running it without heatsink on vmem.
This setup was for 4× P40, but since I'm very lucky their price tripled while I was fucking to find out if I can do water cooling on them at all. But all in all it's running about 45ºC on 100% load.
It's got second 1.2kW PSU btw just for teslas. Famous DPS-1200 with PCIE breakout board. I can power my kettle with 12V while talking to my imaginary llm friend
Hardware itself is not expensive, but everything else cost arm and leg. Overall making server hardware to run silently at home was very fun project!
More like a day, when it's burning close to 1kW
Sorry for going back in time to this post. I don't use proxmox regularly but there is some sort of a paradigm shift at work and, as a result, I may want to play with it in the near future, why not on the couple (read 3) of KS-LE-B I happen to have handy. My question is, can you really build a proxmox cluster with 3 nodes spread over 3 datacenters, in 3 different subnets? I asked Copilot the question and he wasn't a big fan of the idea. Should I just try and have my KS-LE-Bs in the same datacenter since vCrack is not an option with these boxes? ... and even if I do these probably won't be in the same subnet.
I tried it with Hetzner nodes over multiple datacenters in 1 location and it's a no-go.
Cheers, I'll forget about it then, too much unlikely trading involved to have all 3 LE-Bs in the same datacenter (and in the same subnet) just to see them burn like a baguette forgotten at the botton of the oven next time OVH starts a bonefire.
For true HA in Proxmox you need a shared storage like CEPH and low latency as corosync requires it. In my opinion on KS-Line the best way of HA is on application level (inside VM).
If configured right, same location is fine. I operate a cluster of 3 LE-Es and 2 LE-Bs at RBX
Thanks for the great explanation. Let's see if my few pending orders will allow me to make this happen with Monopoly style trading.
All baguette DC should be fine.
As long its below 10ms latency, according to Proxmox.
Problem is rather no vCrack, so you limited to 500Mbit shared uplink.
Which in the case of a DDoS takes a giant dump.
I guess depends on what kind of application needs HA.
They rarely die or go down, nightly backups will do for me at least.
If the mobo dies, its replaced before you wake up again.
But OVH is OVH, they already burned down a DC, so better be not in the same.
My Baguette looks like this:
CPU: Intel(R) Xeon(R) CPU E3-1650 v4 @ 3.60GHz
RAM: 64GB DDR4
Storage: 2x4TB HDD + 2x480GB NVME
Bandwidth: 500 Up, 1G Down
Location: RBX
For 17,99 € per month
I like it
lovely baguette indeed.
Thats it, I'm paying one of my orders