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.
★ VirMach ★ RYZEN ★ NVMe ★★ $8.88/YR- 384MB ★★ $21.85/YR- 2.5GB ★ Instant ★ Japan Pre-order ★ & More
This discussion has been closed.
Comments
@VirMach
Is it possible to upgrade SPECIAL 1536 to SPECIAL 2560?
TYOC039 (384MB RAM) + Template: Linux Debian 10 X86 64 Gen2 V1
After upgrading by apt
I don't know if the cause is in the Debian 11 template, but at least on a 384MB machine, this flow looks good.
MOAR
MOARRRRRRRRRRRRRRRRRRRRRRRRRR
My vm crash / got offline when try yabs (right on fio disk test)
Must boot manually to make online again
same here, vm got offline at random time. and have to boot manually at web panel.
TYOC039 (384MB RAM)
TYOC039 (384MB RAM) +Linux Debian 11 X86 64 Gen2 V1（with default kernel and cloud-image-amd64）
system auto shutdown many times.
I try Almalinux 8 X86 64 Gen2 V1 today
TYOC040
Disk speed got improvement but not for speed, still same like before
iperf3 Network Speed Tests (IPv4): --------------------------------- Provider | Location (Link) | Send Speed | Recv Speed | | | Clouvider | London, UK (10G) | 61.1 Mbits/sec | 179 Mbits/sec Online.net | Paris, FR (10G) | 91.7 Mbits/sec | 464 Mbits/sec Clouvider | NYC, NY, US (10G) | 98.0 Mbits/sec | 160 Mbits/secI have some honest questions for those of you with "system auto shutdown" comments and 384MB VPS.
How much swap do you have?
What do your logs show around the time of shutdown?
What services are you running ?
What optimizations, if any, have you done so these systems can run comfortably in 384MB without OOM'ing ?
Thank you in advance for answering my questions. As I run only one 384MB KVM in Amsterdam and it is a bit of a PITA at times for me.
EDIT: I run Ubuntu 18 on this 384MB and have run Alpine in the past because running Debian or Alma or Centos gives me too many OOS problems.
top - 10:21:03 up 11 min, 2 users, load average: 0.01, 0.00, 0.00
Tasks: 74 total, 2 running, 72 sleeping, 0 stopped, 0 zombie
%Cpu(s): 0.0 us, 0.0 sy, 0.0 ni, 99.3 id, 0.0 wa, 0.0 hi, 0.7 si, 0.0 st
MiB Mem : 348.5 total, 185.2 free, 63.2 used, 100.1 buff/cache
MiB Swap: 256.0 total, 256.0 free, 0.0 used. 274.8 avail Mem
"Linux Debian 10 X86 64 Gen2 V1" update to debian 11 with 5.16.0-0.bpo.4-cloud-amd64 #1 SMP PREEMPT Debian 5.16.12-1~bpo11+1 (2022-03-08) x86_64 GNU/Linux
everything is cool
@VirMach Feedback a problem, running the yabs script sometimes automatically shuts down the vps.
about ipv6
try https://www.tunnelbroker.net/
Until VirMach provisions the native IPv6 this is a great recommendation.
set your swap
looking forward to native IPv6 and hong kong
2G has been added, but it only occasionally triggers the vps shutdown.
and if you do not run YABS does VPS still shut down automatically?
opss....
fortunately, dont use this scripts every day.....
I think the problems are:
what it is mean Booting from hard disk after rebulid my VPS,and can not connect ion SSH.thanks(Node40,768M,debian10and11)
will there be an email notification when machine in HK is going to be sold? thx.
I was actually kind of surprised when VirMach came out with 384MB as an option in the new offers. Then I thought about it and think I understand. (although nobody can truly understand the mind of VirMach) If you have a 128GB RAM maximum on Ryzen, big gen 4 NVMe disks, plenty of reasonable cost IPv4, and you want to limit electric cost. How do you maximize revenue per node and still make the LE market happy?
The price difference between 384MB and 768MB is not much, so the increased node density makes more sense. As the limiting factor for density would seem to be RAM. I expect that is why the 8GB RAM VPS quoted earlier in this thread was not as appealing as one would normally have expected from VirMach in the past when running E5 nodes.
We can return to our LE 384MB roots, or buy 2.5GB RAM with Ryzen at a very, very good price, subsidized by the 384MB crowd. In the end I think VirMach made a very smart choice on how to keep almost everybody happy. (except the support staff
)
/End of ramble.
VirMach does not normally send out marketing or sale notification emails. You should monitor LET to find out all the sales and new location information from VirMach.
We hear about it here first !!
HostLoc is a close second.
P.S. Congratulations on your first comment.
I'm running wireguard on Debian 11 with the default swap, but it should never have been shut down.
(Excluding troubles on the hardware side)
well it does seem like debian 11 can run with 384mb ram , and wireguard is as lightweight as it gets
https://www.debian.org/releases/bullseye/amd64/ch03s04.en.html
Nope.
I see tons of OOM errors for 384MB and 768MB packages where they hit something like 440MB or 800MB+ and get shut down. It sounds like that, not related to the disk in this case.
How often are you running YABS?
Fine I'll do it. Maybe. Soon. But don't think that I enjoy it one bit!
Yuck, IPv6.
Sorry, it's been all the years with CC, I'm conditioned.
I'm already regretting doing Hong Kong, I think I'm going to reverse that decision. I can't handle another 200 people asking when.
You think it's only gonna be 200 people considering it's a perfect target for your Chinese clients?
But to be fair, a VirMach Hong Kong location would also be interesting alongside Tokyo for the Asia region.
Here is Debian 10 running wireguard/znc on 128MB RAM (OVZ though).