enotch vps down 4 days now

cristina78 · #1 03-13-2010, 05:24 AM

Hi all,
I'm desperate looking for someone that may have had the same problem.

I have a Enotchnetworks VPS Uranus plan, with 1GB RAM, with the recommended Centos5.4 install. Everything was going fine, the machine is pretty responsive, until suddenly the VPS stopped working: no network, but the control panel would say it's booted but no RAM was being used. After powering it up, the disk usage graph would keep growing from 13% until it reached 100%. Reboot, it goes back to 13% and then it starts to grow again. No network, no RAM. I think it's an OpenVZ machine.

Their support sends an average 1 message a day with the "try now" thing, but nothing has been solved. I've called the 1800 number 24x7 support, but a answering machine picks up everytime.

Yesterday they told me it has something to do with a failed udev startup.

Maybe someone here has seen this before? Or anyone has a solution for getting my backup out of there? (unfortunately my last backup didn't work and I'll have more trouble if I start it over from scratch)

It's really sad, I was a happy Servebyte customer for a year, then switched to Enotch so I could run more apps and now 4 days and no VPS...

Thanks and regards,
Cristina

nice-price · #2 03-13-2010, 05:59 AM

If you upgrade Centos (or other distributions) you might run into problems with udev. Here´s a small howto for solving that issue: http://service.welcome2inter.net/kno...yarticle&id=30 . You might set this by yourself via a rescue-shell or your vps-provider can do that via host-access.

tsj5j · #3 03-13-2010, 11:41 AM

HyperVM used to have a fix-udev option.
Anyways, as mentioned earlier, this is a likely culprit.

Enotch support isn't spectacular, but... its cheap.
Don't place mission critical stuff on it.

cristina78 · #4 03-15-2010, 03:33 PM

Thanks for the tips. They don't offer a rescue shell tho, so I'm left to the mercy of their absent support.

One week and no solution yet. 3 days without even saying they're trying...

I would not recommend Enotch for ANY application. It's a total rip off!

-Cristina

**123Andrew** · #5 03-15-2010, 03:45 PM

This happens with alot of VPS Panels, If you interupt a back-up or something near that order with a reboot/shutdown/start up, Then the container will completely become unresponsive until it is rebooted by the administrator, Ask Enotch you reboot your container, Tell them that it HAS to be rebooted before it will be responsive again. close the ticket asking why it's down and just simply ask for a container reboot.

enotchnet · #6 03-15-2010, 06:09 PM

There is a lot more to this than the OP has posted. The support guy asked for more information that you originally provided and finally you provided this info below:
--
The only "weird" thing I may have done was some fake swap I needed to trick install oracle-xe with this script, but it was a run once script and finally I didn't even install oracle-xe:

#!/bin/bash
SWAP="${1:-512}"
NEW="$[SWAP*1024]"; TEMP="${NEW//?/ }"; OLD="${TEMP:1}0"
umount /proc/meminfo 2> /dev/null
sed "/^Swap$Total\|Free$:/s,$OLD,$NEW," /proc/meminfo > /etc/fake_meminfo
mount --bind /etc/fake_meminfo /proc/meminfo
free -m
--

This has nothing to do with service quality since all of the methods to typically fix / correct a udev issue are not working for your VPS container. This is also not a managed service. We offer managed VPS plans that go through a different support process and are on different hardware and are powered by Xen. So let's try to evaluate the current situation and categorize it as a Enotch problem. There are 100's of other VPS's that are not having this issue.

I believe that there is a possible solution that is being tested now as the issue is clearly more than the udev error message and more aligned with the configurations that you implemented.

tsj5j · #7 03-15-2010, 10:30 PM

That's pretty irresponsible of OP to blame his issues on Enotch, especially since the problems are self-induced and on an unmanaged provider.

Pathetic.

usera · #8 03-16-2010, 01:49 AM

how irresponsible in trying to install oracle xe with such a large fake swap. Very unprofessional to say the least on an openvz platform.

**123Andrew** · #9 03-16-2010, 02:01 AM

Would would have been nice to know what you done to the VPS before hand instead of blaming it on enotch, As if WHT doesn't have enough hate threads =\

cristina78 · #10 03-19-2010, 08:15 AM

Still no solution. The machine responds to a ping, but ssh does not respond.

Looks like they can't control their VPS, they don't have enough people, or they are just using my memory or bandwith for someone else... how would I know?

You can't offer an unmanaged VPS and not give the user ways to manage it: rescue shell or access to files, at least. All I have at Enotch is start, stop, reboot, rebuild. That's all.

-cristina

usera · #11 03-19-2010, 08:28 AM

on the bright side, some vps providers here would have automatically cancelled your account END OF STORY for trying to create a fake swap and corrupting your vps. Either rebuild or pay for managed hosting in which case I'm sure a managed openvz provider would also cancell your account if you did the same thing over there. Also don't try to pull this on a xen host. You would probably be cancelled trying to change your swap size.

enotchnet · #12 03-19-2010, 08:35 AM

I see that you are still not providing all of the details.

For starters we created a second VPS for you and the details have been in your account since the 15th.

Second, there is a response to your open ticket that clearly shows the error which is also the result of your configuration which is reason for creating the second VPS.

Also, you never asked for console access. We have it available its not provided to everyone since we are moving to a new control panel shortly and the login combination will change at that time. No worries.

I just updated the ticket with instructions on how to access the SSH console. This could have been requested in the ticket and we would have gladly provided it to you.

Lastly, we have no reason to use your assigned memory or bandwidth for another account and in fact we would be quite foolish to do so considering the fact that you are paying for 1 VPS and we created a second one for you out of courtesy so in theory we are providing you twice the assigned resources already for the price of 1.

linuxclark · #13 03-19-2010, 08:36 AM

Quote:

Originally Posted by cristina78

Still no solution. The machine responds to a ping, but ssh does not respond.

Looks like they can't control their VPS, they don't have enough people, or they are just using my memory or bandwith for someone else... how would I know?

You can't offer an unmanaged VPS and not give the user ways to manage it: rescue shell or access to files, at least. All I have at Enotch is start, stop, reboot, rebuild. That's all.

-cristina

cristina78, you are turning this thread into a positive review for Enotch now.

cristina78 · #14 03-19-2010, 08:44 AM

But that's not something I've *done* to the VPS, its a *temporary* mount I did when I was installing things in the beginning, because oracle-xe goes looking for swap using the "free" command during installation. It's NOT A FAKE SWAP SPACE. It's a fake report on swap to programs that ask for it, and only for a brief period of time.

After that, I've restarted the VPS many times and never used the script again, and free was back to it's original state. Just try it on any VPS or in a CentOS vm and you'll see that it goes away after reboot.

-cristina