Reg365/NamesCo - web20 server issues - anyone else?

[harlemangel]

Hey all
First time to post here, even though I find the forum extremely useful when trying to find out the status of reg365/namesco issues.

The reason for my post is that I have about 15 sites on web20, and for the last 2 weeks ive been experiencing issues where all the sites go down every few minutes for about 60-90 seconds.

Ive been with 365 for about 6/7 years now, and like us all, have experienced varying levels of service. I was worried (like most im guessing) when reg365 was sold to Namesco, but have stuck with them as I feel the support has improved, albeit slowly, over the last while.

~However~, I raised a ticket on the web20 issue almost 2 weeks ago, and even though Ive been trying to inform them that the issue is with the server, I kept getting the response that its an issue with my sites. They even told me that I should signup to the new cluster, and manually transfer the sites myself, which I would have proceeded to do, wasting a lot of hours, until yesterday morning when I received :

"We've identified the cause - Apache is crashing. A CRON checks if httpd is running every minute, and if not restarts it, which is why there can be delays. "

And despite a request for more information, Ive received nothing.
As you can imagine, my clients are getting quite agitated and are wondering when this issue will be resolved. And because Ive received no indication from 365/Namesco, Ive nothing to tell them other that to keep saying that the issue is being looked into. However after 2 weeks, 3 of my clients are considering pulling their sites from my company to transfer to someone else.

I am extremely disappointed with the support Ive received on this. Im not one to go bashing people/companies as I believe that you cant forsee everything that can happen. But good customer service must be a priority.

So my question is : Has anyone else experienced issues with web20? Ive asked this of the support team in NamesCo, they told me no, but I fail to believe that Im the only one.

Thanks
HA

 

[johndineen]

Ditto

I have exactly the same problem.

this is the first reply i got after the support agent doubting me on the phone he only belived me when the site went down as part of the call

As of yet, we have been unable to replicate this issue that you are reporting, and the HTTP logs for your domain are not revealing anything of significance either.

We have also been unable so far to locate any performance or usage problems with this particular server.


Then after my reply questioning my fears of adding any more customers to this site i got this reply:


As has been previously stated, we haven't found any performance problems on the server you are hosted on so if you add new domains to your current hosting there shouldn't be any problems with them.

 

[jason]

OMFG!!!

I am in the exact same boat as you guys and I believe we are not the only ones on these forums either that have experienced problems with their "legacy" servers.

All along I knew that the performance issues I was seeing was something to do with apache going down and not network issues. I have spent a total of about 3 hours on the phone to them since January as emails gets me nowhere quickly. What really annoys me is that they initially blamed my code bases, and then I pointed out to them that phpmyadmin on their host was suffering the same problems.....my code I think not!!

Finally after much deliberation they accepted that there was something wrong. What really annoys me is that when you do ring up the people you get to talk to are just log checkers and can't actually fix anything. Then the people they escalate the problem to, you rarely get to talk to, all I get is an email several days later in the post saying ..... "we checked, and you should be glad to hear that everything is sound". I have sent them such hate mail regarding this problem as even today I have a client on web20 who is considering going elsewhere also.

They told me that they will activate a new cluster hosting account for me for one month. During that time I am to move my own site/apps over to the new cluster from web20. Once migrated, they will then update DNS and cancel my account on web20. They offered this to me for free. But I have many hosting accounts on web20 and they only offered to move one for free.

I asked them straight out, if I sign up to you as a new shared hosting customer do I get to go on to the new cluster. They said YES!. So then I asked them....Why as an existing loyal customer do I have to put up with an inferior level of service?. At this stage they turn in to a phone councellor and just agree with everything you say just to keep you from shouting at them.

The last thing I was supposed to get from them was an email clarifying whether or not there would be a fee to cover the cost of an upgrade to the new cluster initiated by me, or not. They have failed to get back in touch. This was after 2 - 3 days of phone fighting and me being put through to higher levels of management.

They told me that at some stage over the coming months all of us on the legacy hosts Web(x) will be automatically migrated to the new cluster.

I am actually considering opening up a new hosting account on their new cluster and moving accounts myself over. I know this will cost me but FFS they really do have us by the balls.

So my question is this....Has anyone used their new cluster yet, and is it performing well? Is the level of services and configuration available as good as if not better than what we had through HSphere on the Web(x) machines.

Cheers

 

[paul]

I've had an issue with AWSTATS since January which has them miffed and they can't fix. They also offered to move me over to the cluster. Thankfully I will be moving my bigger sites to a better host very soon.

 

[jason]

They have finally offered me the option to transfer any/all hosting accounts I have with them on the legacy servers over to their new cluster.

I was managing my accounts through HSphere and actually liked it. I'm sort of dreading to see what the new system is like but looking forward to "100% Uptime" and improved performance.

We'll see, and I'll keep you posted.

 

[harlemangel]

Hey Jason, are they offering to do this for free? I am finding that performance on the older servers (web9 for instance) is dreadfull at the moment. Also, the fact that there are no backups being run on the legacy servers, I would like to move my sites also to the new cluster, where they have assured me that regular backups will be a feature. But if I have to do it myself manually, it will take forever (100+ sites).

 

[hosting365]

I understand that -all- sites on the legacy kit will be moved to the new clusters inside the next 6 weeks.

 

[letsmove]

I understand that -all- sites on the legacy kit will be moved to the new clusters inside the next 6 weeks.

A reply i got from R365ish when i asked about moving my sites
All hosting accounts will move to the new cluster automatically within the next year. If you would like to take advantage of our new cluster early, then please call out sales team and sign up for one of the new hosting packages. You will be able to upload and test your sites, once happy changing the name servers will complete the move.

Once you have completed the move of all the domains on your HSphere account, any remaining credit can be moved to the new hosting package.

Should you require any further assistance, please feel free to contact us.

If i have to move the sites I wont be moving them to another R365ish account!!!

 

[thebosdj]

I understand that -all- sites on the legacy kit will be moved to the new clusters inside the next 6 weeks.

That's not possible without the agreement of the client!

Where do you get that information from?

 

[moyda]

I'm unfortunately with this dreadful company also. I think you can't get ssh access on the "new" cluster either. Support tickets go unanswered and then eventually get closed. I've given up and am I'm leaving soon. This company is the web hosting equivalent of the Titanic.

 

[php.allstar]

I advised a client to go with them a few months back as I thought I knew their Apache setup inside out, based on my experiences through other sites I have with them.

It turned out the client ended up getting on a shared server with the ZEUS webserver.... YUCK YUCK YUCK!

Needless to say, the client got their money back and we moved to a different host. Not even going to look at them anymore.

*Forgot to mention I was on the phone with them a few days back about a hosting account we have with them where I work in the day job. They told me that they will be moving everyone onto the new cluster over the Summer, and if we like, we can initiate the changeover straight away via the HSPHERE control panel. They also said that there will be no Shell access. As a result, we're going to move all our shared hosting accounts to a new dedicated box, something we should have done when I joined the company!

I'm going to avoid them like the plague.

 

[Anotheridiot]

I'm another poor unfortunate on Web20 and had all the same problems that were mentioned above.
Being a network geek myself I always try to do as much as I can before logging a call. So when I log a ticket, its either something I cant troubleshoot, or something Ive already proven sufficiently is not due to my config or stupidity.

When I logged this with them on 24th March, I got a standard, its a shared platform canned response on the 27th and finally on the 31st after numerous mails back and forth they responded that the Apache process was crashing.
The ticket (according to their site) was escalated to 3rd line support where it remained idle untill being "closed due to inactivity"
I find this most annoying (as I work in a first line support role myself) and updates to the customer is the most important thing in my day. Too little feedback during the life of a ticket and nothing when the ticket is closed, there I was waiting here for them to come back to me but they never did.
A "we are still working on it" email would be a start.

As bad luck would have it, this particular sites database is on their SQL6 server, which has been having performance issues, confirmed by the hosts, for some time. The standard canned shared server response is all I am getting on this too.

While we are on the same vein, I have a ticket open with them at the moment where all the file and folder permissions on my site magically changed on the evening of may 16th (right around their 1 hour notice scheduled downtime). At the moment they are trying to tell me its the gumblar virus, which I have confirmed it is not.

Anyway, this is turning into a rant/gripe at register365 which I didnt really want it to, just wanted to share my experience of the same incidents.

 

[niall]

I'm unfortunately with this dreadful company also. I think you can't get ssh access on the "new" cluster either. Support tickets go unanswered and then eventually get closed. I've given up and am I'm leaving soon. This company is the web hosting equivalent of the Titanic.

In fairness, a lot of hosting companies no longer offer ssh on shared hosting accounts as it's more trouble than it's worth. The alternative is to get a VPS which you can do whatever you want on the CLI without fear of other users being affected.

Niall.

 

[harlemangel]

While we are on the same vein, I have a ticket open with them at the moment where all the file and folder permissions on my site magically changed on the evening of may 16th (right around their 1 hour notice scheduled downtime). At the moment they are trying to tell me its the gumblar virus, which I have confirmed it is not.

Ive been fielding complaints from customers since last Thurs about sites that went down due to this chmod of file permissions. Only after some querying through the support forum did I find out what happened, and why. They actually changed the file permissions themselves. It wasnt the virus that did it, but they reckon that 777 causes a security risk. I can understand this, but no notification was provided to us that this update was going to be applied. And the performance of web20 is diabolical. Bloody worn out with this company at this stage.

 

[Anotheridiot]

What support forum were you reading, Im looking at forum.register365.com and cant see anything mentioning this on it ?

Im still dealing with them via email support, again this morning they respond again with their Gumblar answer.
Ive asked for logs showing my ftp activity.

If they have confirmed already that this was carried out by them, then why wont they tell me the same thing ?
I had 777 on four folders, which is required for my application (phpBB) to run.

 

[mneylon]

I had 777 on four folders, which is required for my application (phpBB) to run.

Why does phpBB need 777 ? That seems a bit odd...

 

[niall]

If they have confirmed already that this was carried out by them, then why wont they tell me the same thing ?
I had 777 on four folders, which is required for my application (phpBB) to run.

To be pedantic, PhpBB doesn't require 777 permissions, that's just a quick hack due to limitations in the majority of control panels.

What's really needed is either for the directory to be owned by whatever user Apache is running as with 755 permission. The alternative is to have group ownership as whatever group Apache runs as and have 775 permissions.

Niall.

 

[harlemangel]

What support forum were you reading, Im looking at forum.register365.com and cant see anything mentioning this on it ?

Im still dealing with them via email support, again this morning they respond again with their Gumblar answer.
Ive asked for logs showing my ftp activity.

If they have confirmed already that this was carried out by them, then why wont they tell me the same thing ?
I had 777 on four folders, which is required for my application (phpBB) to run.

Anotheridiot, it was actually through the support tickets that I found out, sorry for the confusion. Below is an excerpt of one of the responses I received :

#####
The need to have any directory set with 777 permissions is a security risk to the site involved, and on shared hosting, this poses a potential threat to other hosted customers.

The reason for your site permissions being changed, is most likely as a result of some maintenance work that was recently carried out on Web20. There has been a growing threat posed by a web attack known as "gumblar", and any directories found to contain this threat had it's permissions changed to improve security of the server.
#####

No apology for the sites being brought down or the lack of notice, or any reassurance that this wouldnt happen again.

Hope this helps!.

 

[Anotheridiot]

Thanks harlemangel

Ive gone and set up some hosting with another provider. Im moving my test site as we speak, and once I get it running will be changing DNS over.

Enough was enough to be honest.

 

[Anotheridiot]

Why does phpBB need 777 ? That seems a bit odd...

Im just following instructions to be honest


I cant post links yet, so heres a quote from the article over on phpBB site.

Correct Chmod Values

* config.php
o 666 before installation
o 644 after installation
* All other files - 644
* All directories - 755
o There are some exceptions when it comes to directory permissions,
+ The files directory - 777
+ The cache directory - 777
+ The store directory - 777
+ The images/avatars/upload directory - 777

To be pedantic, PhpBB doesn't require 777 permissions, that's just a quick hack due to limitations in the majority of control panels.

What's really needed is either for the directory to be owned by whatever user Apache is running as with 755 permission. The alternative is to have group ownership as whatever group Apache runs as and have 775 permissions.

Niall

I will give this a try to see how it works out.