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
Push-up harder to become a 10x developer.
Atlanta down. What a dump.
There is no control panel for most of the nodes. Like someone mentioned above, solusvm is broken since a long time. It says no support and tickets will be chargeable as mine is a BF special promo.
Hold the f*ck down. When Borta was floping everyone believed in him for like 2 months without to much wine. Now "top host" having some trouble and hienas already gathering.
Give virmach at least 30 days from now to correct their misstakes. And please, from all my heart, sue them.
MJJ HELP MJJ
And mine still is down and no response to ticket opened, this is indeed bad and frustrating :-\ I kept some websites under construction there without backing up, silly me...
the follow up to this is my server was down for like 5 days because the ip suddenly stopped working.
apparently they gave it to someone else, had to request a new ip and like 48 hours later i got it and reconfigured the network.
it worked then, but I already moved my site.
saving a couple bucks isn't worth having to use that fucking mess of a host.
horrible provider. run away
yeah anyone that uses them is a masochist to keep it a buck with you chief
Gosh, this is such a Sigma way to die. You should strive to end like this, Jonny.
lol OP got banned for calling mods Nazis...
Maybe try calling them Japanese imperialists next time. Same atrocities but much more politically accepted.
Alternatively, here's a revolutionary concept: why not stop calling mods names?
As mentioned in my PMS thread a few days ago, was any attempt made to communicate with the user about why said evaluation was made and what should be done in their opinion to correct it?
As I've just tried to say (above), not calling mods names strikes me as a much better (more considerate, safer, etc.) strategy than trying to decide which names to call them. Wouldn't you agree?
Unfortunately, not all people have the ability to provide feedback constructively (and I said as much in the PMS thread.) Can we at least get their opinion first and then choose to ban them should there be no useful outcome from trying to initiate a conversation?
You know, for a mod or an admin, it's a judgement call based on the circumstances. Sometimes, the circumstances are very clear, and sometimes, they are less so, and a mod or an admin tries to draw this distinction and to act accordingly, and sometimes, despite this effort, they get it wrong (and if pointed out, they should try to correct their decision)
Frankly, in the case of the user in question, I think that the circumstances were very clear and that the decision was appropriate. (In other words, by all signs, he knew what he was doing and couldn't care less)
My server has been down now, going on 72 hours. Hitting reboot does nothing. Putting in support tickets does nothing. Crazy.
Welcome to the family buds!
Already 3 days,sent ticket,nothing to do.
here is the issue:
Two virtual servers are set to the same IP.
Provider doesn't have sufficient firewalling (formerly ebtables, but it's being replaced by nftables) to prevent this from happening.
Upon reboot and every few minutes, each server emits an ARP request, and the router binds the IP to its MAC address.
Incoming packets would go to whoever sent the last ARP request.
If it's you, you can access your server.
If it's the other server, you feel you are "hijacked".
If my guess is correct, you should also have difficulty connecting to your server via SSH sometimes.
You have to wait for the other server owner to realize the problem and change their IP.
Or you can set an unused IP in the subnet, but this would break someone else if it's later assigned.
Or you can emit own ARP responses to the "offender" MAC address and make him stop using
the assigned IP, while setting static address in /etc/ethers
But if a provider doesn't nullroute or at least monitor such attempts it's a really bad practice.
FWIW: my @VirMach box (NL, AMS) was unreachable too. So I logged into the panel (which worked without any problem) to check. And indeed, my IP had also been changed by Virmach.
So I quickly changed my config(s) via VNC and then rebooted the box.
Everything is working OK again.
Unfortunately it seems that Virmach is not properly communicating the state of things in perticular whether one_s VPS has already been migrated to AMD Zen (Ryzen iirc) or not (but that might also be me not following closely enough and on all channels).
So, I'm currently doing a series of benchmark runs and will "diff" the result set against an earlier one.
A quick glance over the first results shows some amazing things. ProcMem is roughly on 26xx v4 or Epyc level "only" ("only" in '"' because I'll certainly not complain about that kind of a processor and memory - but not Ryzen as originally promised iirc - with a VPS costing just about $10/yr), but the real kicker is the almost 40 MB/s and a bit over 10000 IOPs (4k/4t). Connectivity seems to be quite nice too, DE_FRA ca. 900 Mb/s, UK_LON, FR_PAR, IT_MIL about 600Mb/s, East Coast 100 and 120 Mb/s, JP_TOK and SG_SGP about 45 Mb/s and even down under a not measly 35 - 40 MB/s.
But - WARNING - keep in mind that this is based on just a single run and it might just have been during a lucky moment or with still only a handful of VMs on the node.
I'll keep you updated.
Oh, and @Virmach: you really, really need to work on your communication! The nicest hardware and connectivity aren't worth much when you p_ss off customers in droves by communicating very poorly!
OK, here's the comparison "old" (as originally purchased) vs "new" (current "updated"?).
The processor string changed from "Qemu ... cpu64-rhel6" to "Qemu ... v2.5+" which doesn't tell us much other than the VPS has been moved to another node.
As for the benchmark results the "original" benchmark was done with a significantly older version of the benchmark software so I can only compare what's comparable.
(please forgive my crappy formatting)
Funny, when doing the computations single core the "new" system is quite a bit slower, but when doing it using multiple threads the "new" system magically gets quite a bit faster. Note that that is normal, what is not normal though is (a) that the "old" system did not show that, and (b) how much difference there is in the "new" system.
What really somewhat shocked me though is the AES result which is way below even older Xeons while at the same time the RSA result is about what one would expect from a 26xx v 3 or 4 or an Epyc, but with pretty much all relevant flags - needlessly - disabled; you get SSE but only up to sse3 and no AES hardware support, no popcnt, etc.
My personal summary is that while it seems that the new node's processor seems to be a better and more modern processor than on the old node it was configured by a heavily drunk newbie technician.
The new node however has a roughly 2x faster disk and a 4k4T result of 37.56 MB/s (min 33,27) and 9616.5 IOPs (min 8517,87) with a quite low spread of 88.6% to 109.8%. Those results (btw, based on about 35 runs) are excellent, impressive, and surprising. I've seen many VPS costing 3 or even 5 times more whose drives do not achieve that performance.
As for the network I saw many changes over almost all targets. Most towards the positive and a few towards the negative but most in the range of +- 10% to +- 20%. Unfortunately the targets who are slower now mostly are asian targets that were already slow before like India, Singapore, and HongKong.
A few targets however show changes surpassing 50%, gladly all of them to the positive. Moscow (RU) is almost 90% faster now, Frankfurt (DE) is 120% faster and Milano (IT) even 200%.
TL;DR I'm impressed by the new node's disk/NVMe, OK with the changes in connectivity (although I would have preferred to see weak targets getting boosted a bit instead of seeing already strong targets getting even stronger) and kind of tending towards the positive wrt the new node's processors - mostly however I'm stunned and disappointed by how you chose to configure it. Example: no AES hardware support, are you kidding us, Virmach? This is the 21st century! Being at it, also kindly stop the "qemu ...v x.y" BS and tell us the actual processor; your customers deserve to know that information and they also need to (hint: spec. exec. vulnerabilities).
But probably even more important than the technical points: TALK TO US, @Virmach !!!. The way you are behaving is a recipe for doom.
Your incompetence has been doubled.
We have SSE 4.2 and AES and SHA-NI.
Hint: reinstall a "gen 2" template in SolusVM.
Alternatively, send one push-up per CPU flag to @VirMach along with your IP, and VirBot will enable them for you, without needing to reinstall.
So? All that means is that you looked at a different node . Mine doesn't have AES hardware support and sse4x and that's what I was looking at and writing about, my VPS.
Please, spare me your pushup BS.
"Top Host" lol
I did push-ups for 24hrs and this host still sucks
On my fourth day without my server. From what I am reading, all of you have networking issues. Does anyone have a vps that simply won't boot? Have you all reinstalled the os?
The price, especially the BF deals, is probably worth the migration as long as you don't use them for anything too serious, but I do not envy the provider or anyone with lots of VirMach systems after this migration. I've spent a fair bit of time updating doc, keys, and configs, but I'm most worried about the inconsistency now.
For example, I have a few systems that show one IPv4 address in the control panel, but another on the system, even after a reconfigure. I had one system get a second IPv4 address, but this was not part of the plan or reflected in the control panel. It went away after reconfiguring, but I'm worried that there is a pretty big mess that may never be quite untangled.
My service is even being terminated without any reasons. Opened a ticket, waiting for their reply. Sighhhhh.