issues with VPS

[Latona]

Support ticket raised for this.

2 emails received about cpanel:
cpsrvd failed @ Fri Jun 29 14:27:00 2007. A restart was attempted automagicly.
cpsrvd failed @ Fri Jun 29 11:40:14 2007. A restart was attempted automagicly.

The website hosting doesn't load up - I login to the WHM and get this:
The server was unable to lookup an an A entry for its hostname (server.myserver.com). This is generally because the entry was never added. However this could also be the result of your nameserver(s) being down. If you would like to attempt to automaticlly add the entry,

Found your hostname to be: server.mydomain.com
Found your short hostname to be: server
Found your domain name to be: mydomain.com
Found your main ip to be: IP.IPI.IPI.IP
If this looks correct,

IP=IP.IP.IP.IP Bind reloading on server using rndc zone: [mydomain.com] Error reloading bind on server: rndc: connect failed: connection refused Add Complete

Now what will cause this to happen? I changed nothing to do with server.mydomain.com - I followed the popup prompts in the WHM and it said it failed... however it should already have been in place since it was setup by you guys - I can only assume you set it up incorrectly otherwise it wouldn't give me the problems. Or you changed something today that caused the failure - as I made no changes at the time specified - as have not been on today until now - but once again cannot do anything because of downtime.

In the Virtuozzo status I see the following:

TimeDescending Old Status Status Obtained
Jun 29, 2007 02:26:39 PM VPS Hosting is starting up at the moment. It may take some time.(starting) VPS Hosting is up and running now.running
Jun 29, 2007 02:26:33 PM VPS Hosting is starting up at the moment. It may take some time.(starting,down) VPS Hosting is starting up at the moment. It may take some time.(starting,mounted)
Jun 29, 2007 02:19:12 PM VPS Hosting is stopped now.down VPS Hosting is starting up at the moment. It may take some time.(starting)
Jun 29, 2007 11:40:14 AM VPS Hosting is mounted now. It is not running and some operations are not available.mounted VPS Hosting is up and running now.running
Jun 29, 2007 11:40:08 AM VPS Hosting is starting up at the moment. It may take some time.(starting) VPS Hosting is mounted now. It is not running and some operations are not available.mounted
Jun 29, 2007 11:38:53 AM VPS Hosting is stopped now.down VPS Hosting is starting up at the moment. It may take some time.(starting)
Jun 29, 2007 11:36:38 AM VPS Hosting is setting at the moment. It may take some time.(setting) VPS Hosting is stopped now.down
Jun 29, 2007 11:36:35 AM VPS Hosting is stopped now.down VPS Hosting is setting at the moment. It may take some time.(setting)
Jun 23, 2007 01:30:46 PM VPS Hosting is destroyed.non-existent VPS Hosting is up and running now.running
Jun 23, 2007 01:30:40 PM VPS Hosting is up and running now.running VPS Hosting is destroyed.non-existent


However, it states that it is running - but the website hosting does not load?

What has caused it to have the status changes today? - I have only just now got checking it and whatever was done todayat 11:36:35am caused it to go down - I wasn't online and made no changes then so I can only assume you guys made some alteration?

At the minute I do not know how to get the website hosting up and running as I am trying to restart the server - however, it does not seem to have resolved the issue.

It really does seem like eukhost are really trying to fob me off with a highly inferior service to what they advertise as I never get an explantion to the cause of problems without many questions - and then I am assured it will never happen again - only for something similar to occur soon afterwards.

I posted this here as it is the same information sent in the support ticket - maybe all support tickets should be public so that people really find out what the customer service is like - promising things have been resolved only to have similar problems within a week or so.

EDIT:

If it is a result of the nameservers being down - then the single point of failure I had raised a query about for nameservers on a previous ticket was not addressed at all.
I am at a loss as to the reason because the stats tell me nothing apart from what is copy/pasted above.

[DPS Computing]

What have support said to you regarding this?

[flesso]

Do you have any custom confugurations on the server which maybe causing this? Could the scripts for your website hosting be causing a memory leakage or something? Just a couple of idea...

As for the Bind problem, thats simple to solve: you just need to start Bind, using the following code via SSH:

# /etc/init.d/named start

[Latona]

Originally Posted by flesso

Do you have any custom confugurations on the server which maybe causing this? Could the scripts for your website hosting be causing a memory leakage or something? Just a couple of idea...

As for the Bind problem, thats simple to solve: you just need to start Bind, using the following code via SSH:

# /etc/init.d/named start

Oh - forgot to mention numerous times I have informed support that I do not have the technical knowledge as I have never had a VPS Hosting before - normally just used cpanel from within shared server environment. I do not know codes to use via SSH. However, I can only go on the advice that support give me that EVERYTHING is fixed and won't be a problem in the future etc.

There is custom configuration from the default setup. I had included a full requirements of the script that I am using in a message to support - who then done the setup.

Since eukhost moved servers I have done nothing to the setup - it has all been done by support - due to the week+ of downtime I thought I would let the 'experts' sort out the problems - however, not even 1 week of having it sorted out and things starting to go the same way again.

I don't know about a memory leakage from the script - considering it is used by various people on shared servers and they haven't reported any problems about their hosting. Honestly, I do not know - it is possible - but I do not know where to check and support have never mentioned it as a possible cause of a problem... it has always been setup issues with things not bein setup correct (from the explanations I received after several requests).

[DPS Computing]

If you havent got the technical knowledge it is best to leave it to support as if you start to enter codes willy nilly and dont know what you are doing and things go wrong then the blame could fall on your side.

What have support said the problem is? And what is the cause?

[flesso]

Considering that you have even been moved over from another node, I think it has to be something you're doing; whether it be a faulty script or custom configuration issue.

As for not being able to use SSH, you need to learn this, you will find it very helpful in the future - you can't just rely on a pretty screen to do everything. Since your having so many problems I would consider semi-dedicated or a high level shared package - if one of these can accomodate your website hosting.

[DPS Computing]

Originally Posted by flesso

Considering that you have even been moved over from another node, I think it has to be something you're doing; whether it be a faulty script or custom configuration issue.

As for not being able to use SSH, you need to learn this, you will find it very helpful in the future - you can't just rely on a pretty screen to do everything. Since your having so many problems I would consider semi-dedicated or a high level shared package - if one of these can accomodate your website hosting.

I think Latona used to use shared hosting but moved to a VPS Hosting because it had outgrown the shared hosting.

[Latona]

Originally Posted by DPS Computing

If you havent got the technical knowledge it is best to leave it to support as if you start to enter codes willy nilly and dont know what you are doing and things go wrong then the blame could fall on your side.

What have support said the problem is? And what is the cause?

That is precisely WHY I have left it to support - and from the feedback - it all appears that it is on THEIR end that the problem lies ... never did they suggest it was the script I was using... it was always a problem/mistake/oversight to something they had done/not done.

If it was to do with the script I was using - I would have expected to have been informed when the problem started and continued - not be told that it was problems on their side and that it should all be resolved now etc.

Considering that you have even been moved over from another node, I think it has to be something you're doing; whether it be a faulty script or custom configuration issue.

As for not being able to use SSH, you need to learn this, you will find it very helpful in the future - you can't just rely on a pretty screen to do everything. Since your having so many problems I would consider semi-dedicated or a high level shared package - if one of these can accomodate your website hosting.

Well, I wasn't online when the problem occurred - so it couldn't have been anything that I done specifically that I am aware of. I had been offline several hours (asleep then at a few meetings today) - when I got online the website hosting had been down/inaccessible for 8hrs+

I do have the ambition to learn how to use it - and is a VPS Hosting not better than a high level shared package? - surely VPS Hosting would be better resources than a shared package - considering it costs considerably more per month.

[DPS Computing]

VPS Hosting is better than high level shared - but it means more responsibility on your part unfortunately.

However support should be able to identify whether the problem is at server end or yours if you ask them - which will bring you a long way to helping solving your problem .

[flesso]

As for the scripts thing, what kind of things do the scripts on your website hosting do? Have you had support re-create your VPS Hosting at all, that should be the next step.

A VPS Hosting may not always be better than high level shared hosting, for example the semi-dedicated package for Linux probably has more stuff than the l-VPS Hosting-01.

[Latona]

Originally Posted by DPS Computing

I think Latona used to use shared hosting but moved to a VPS Hosting because it had outgrown the shared hosting.

No. I have only EVER had a VPS Hosting from eukhost.

My previous experience is in hobby website hostings using shared server with control panel with different hosting providers - for hobby website hostings mainly that a basic database etc would be used (maybe wordpress or some common script etc).

The reason I am using a VPS Hosting is that the website hosting has not launched - it is in development, however much of the development time has been lost because it has not been accessible - already explained in a previous post that the script requires a license to operate and cannot be developed offline.

I am looking at launching a business website hosting that will be a primary focus for other business owners. I am using the VPS Hosting as I was advised that a smooth transition to dedicated would be undertaken when required. The script has been run by other people on shared hosting servers without causing problems for those people - therefore a VPS Hosting is a much wiser choice at this stage than a dedicated server... I was advised by others that a VPS Hosting is scalable up to a decent number of members then after that a dedicated or split dedicated server would become more worthwhile, although fully dedicated was definately not needed until the website hosting was receiving regular decent levels of traffic. (That advice was from other users of the script - with most of them using it in a shared environment).

[Latona]

Originally Posted by DPS Computing

VPS Hosting is better than high level shared - but it means more responsibility on your part unfortunately.

However support should be able to identify whether the problem is at server end or yours if you ask them - which will bring you a long way to helping solving your problem .

That was the question I had asked way back when I first had a problem - at the end of may when the server move took place. I was assured it was a problem at eukhost - not with the script I was using.

I am on the L-VPS Hosting-03 - overall cost works out at £47 per month if I remember.

[Latona]

Originally Posted by DPS Computing

What have support said to you regarding this?

In reply to the original query you had DPS, support have responded. The website hosting is now up - and this was the response:

Hello,

I have made the corrective changes in the DNS and now your website hosting is working fine. Feel free to get back to us for any further issues.

However, dns to me relates solely to the domain name - maybe why it was showing that the VPS Hosting was running - but the website hosting failed to load. I made no changes to the DNS - I attempted to correct the dns but when I checked it looked as it should look - how it had looked before when it was setup by support and how it looks now after it was 'corrected'.

To me it appears to lose the dns info somewhere when the server gets mounted/fails - but won't let me set the info (if that makes sense - I have never seen any different info in the WHM from when the website hosting was inaccessible to when it was acessible again).

[eUKShane]

Hello,

I have monitored your VPS Hosting resource usage, cpanel logs, average load and everything looks normal.

1) Regarding "cpsrvd failed" problem I have done following cpanel tweaks :

-- Increased the limit of maximum memory a cPanel process can use before it is killed off (in megabytes).
-- Disabled "Use native SSL support if possible, negating need for Stunnel"

This will definitely help you. Just to let you know that after every VPS Hosting reboot attempt cpsrvd runs to restart all services.

2) Regarding the "A entry for the hostname" I have tried same from WHM and not getting any error, the output is as follows :

Found your hostname to be: server.yourdomain.com
Found your short hostname to be: server
Found your domain name to be: yourdomain.com
Found your main ip to be: IP.IP.IP.IP
If this looks correct,

>> Add the entry

IP=IP.IPI.IPI.IP Bind reloading on server using rndc zone: [yourdomain.com] Add Complete

Now hostname is properly resolving to your VPS Hosting ip, RDNS has been set too. You can check this using following command :

#host server.yourdomain.com
#host IP.IP.IP.IP

3) I have checked logs and doesn't found any reboot attempt at 11:36:35 today. You can check this using "last reboot" command.

Your VPS Hosting was rebooted today, but it was a manual reboot. Please check following logs :
------------------------
Jun 29 22:05:10 server portsentry: Starting portsentry -udp: succeeded
Jun 29 22:05:11 server init: no more processes left in this runlevel
Jun 29 22:36:35 reboot
Jun 29 22:36:35 server shutdown: shutting down for system reboot
Jun 29 22:36:35 server init: Switching to runlevel: 6
------------------------

I am still monitoring your VPS Hosting for next few hours and if I find any problem then will definitely get back to you.

[Latona]

Originally Posted by eUKShane

Hello,

I have monitored your VPS Hosting resource usage, cpanel logs, average load and everything looks normal.

1) Regarding "cpsrvd failed" problem I have done following cpanel tweaks :

-- Increased the limit of maximum memory a cPanel process can use before it is killed off (in megabytes).
-- Disabled "Use native SSL support if possible, negating need for Stunnel"

This will definitely help you. Just to let you know that after every VPS Hosting reboot attempt cpsrvd runs to restart all services.

2) Regarding the "A entry for the hostname" I have tried same from WHM and not getting any error, the output is as follows :



Now hostname is properly resolving to your VPS Hosting ip, RDNS has been set too. You can check this using following command :

#host server.yourdomain.com
#host IP.IP.IP.IP

3) I have checked logs and doesn't found any reboot attempt at 11:36:35 today. You can check this using "last reboot" command.

Your VPS Hosting was rebooted today, but it was a manual reboot. Please check following logs :
------------------------
Jun 29 22:05:10 server portsentry: Starting portsentry -udp: succeeded
Jun 29 22:05:11 server init: no more processes left in this runlevel
Jun 29 22:36:35 reboot
Jun 29 22:36:35 server shutdown: shutting down for system reboot
Jun 29 22:36:35 server init: Switching to runlevel: 6
------------------------

I am still monitoring your VPS Hosting for next few hours and if I find any problem then will definitely get back to you.

Hi,

Thanks - yes - I did try a manual reboot to try and resolve it before raising the ticket. Obviously when the ticket was resolved the website hosting and everything was back. I was simply posting the contents of the ticket here so maybe others could advise on possible reasons - in addition to support so that in future the problem shouldn't happen.

The other logs were from the virutozzo status changes - that was at 11:30 and 02:19 or around then. However, as stated everything is back working again now and has been since the ticket was resolved last night. Hopefully with the tweaks made that this problem doesn't occur again - all I can do is cross fingers and hope as I have no idea what caused it :P

[Latona]

I spoke to soon.

Now I cannot access ftp - the error is: