Educate Technical Support Staff on downtime involved in modifications

[stressfree]

Hi there,

First up, I've been a customer for over a year, and had a dedicated server for six months + now - overall I am VERY happy with the service you provide - 95%+ of the time it is great, so thumbs up on that!

One major problem I do have, which has now come up several times and despite my requests, does not seem to be improving is your technical support advisors understanding of the downtime involved in some upgrades. I was told by your staff on Friday that a member of management would call me regarding this problem, but as I am yet to hear thought I would post it on here, as the suggestions I have sent through technical support helpdesk seem to have fallen on death ears.

The basic issue is that on several occassions I have asked for certain updates to be made, and asked whether there will be any downtime longer than a simple restart or not. My reason for wanting to know this is that if there is a downtime involved, I would always choose to have the update made overnight so that it doesn't affect my customers as much.

On several occssions I have been advised that there will be no downtime, when in actual fact the updates have resulted either in a specific site being down for several hours (e.g. in the case of SSL cert installs) or, as happened on Friday, my whole server being fairly useless for a couple of hours due to lack of memory whilst installing MSSQL server.

I do understand fully that sometimes downtime may be required, but all I ask is that I can be told about it, which is the very reason I ask. To be told by support staff that no downtime will occur when this turns out not the case is not on, and really is a thorn in your side as it is the only issue I have come up against repeatedly with you.

I would really appreciate a response on this, as it has cost me time and caused my customers un-necessary downtime (and the subsequent angry calls to me) on severa occassions now, when all that is really needed is some better education from you to your staff as to the implications of various updates.

Thanks,

Dan

[Ryan]

Hi Dan,

Please accept our apologies for the inconvenience caused let me explain what had happened on Friday :-

The MS XML component which is required and necessary for the MS SQL Services to function properly was in use and was of a higher version which isn't supported by MS SQL. Hence it was removed from your server.

We then installed a lower version of MS XML component and later installed SQL Server which needed a reboot. You server was down for five minutes until the reboot process was complete.

We have briefed all our support staff members about this and they have been told to inform the customers in case of a possibility of a downtime and also mention the duration till the outage is going to last.

Please be rest assured that you will be provided precise information with regards to updates and downtimes (if any) by our support team from now on.

[stressfree]

Hi Ryan,

Thank you very much for the explanation - none of this was forthcoming on Friday, so it's good to get a full run-down.

Although the downtime was only five minutes for a re-boot, my server was effectively down for a couple of hours, as all but the most basic html websites were too much for it - whatever your admins were having to install was using a lot of resources - I don't know what exactly, support told me that it was due to the backup software I was using, but this doesn't really hold water as that usually runs fine, and I gave permission for it to be stopped whilst the update was in progress anyway to free up resources for you, yet this didn't seem to be done.

It was really this long spell where, though not technically 'down', the server was not doing it's job that was the big problem.

Anyway, thanks for the response, and good to know it's been taken into account

Thanks,

Dan

[eUKhost.com]

Hi Dan,

Installation of SSL Certificate should not result in any sort of downtime. We install many SSL certificates every day for many customers and hardly few customers complain as they experience temporary local outage due to change in the IP address of website. SSL certificate can be installed on a Dedicated IP only. We assign dedicated IP to your website and new users see your website from the new IP, as a customer you may find your website inaccessible for some time as your local machine tries to load your website using old IP address.

MSSQL installation process is very simple and it should not result in downtime even on a server with 2GB RAM. This MS XML dependency was unforeseen and our staff should have explained to you what went wrong during the installation process.

I will talk to the Windows Team leader and make sure we give proper explanation to you if something goes wrong due to unforeseen circumstances.

[stressfree]

Hiya,

Thanks, I do think it is really just an issue with communication, and it's great to hear that some progress can be made in this.

I do understand now with things like SSL certs, where the downtime is only due to the IP address changes meaning local problems, but to the customer it still looks like their site is down - the first couple I had installed I was told there would be no downtime, and of course there was due to the IP addresses changing. Of course the server didn't technically go down, but to my customers it looked the same.

Now that I am a bit wiser on all this server stuff, I understand what happens and will always make SSL requests late at night, so it doesn't affect my customers...little details like explaining this initially would though take you guys from A to A++ in my opinion

Thanks,

Dan

[Ryan]

Thank You for your understanding Dan

You can certainly expect proper explanation and complete details from us henceforth in such scenarios.

[Nige]

I'd like to add to this - I don't think Tech Support give enough information out, and it's not always accurate.
I've just migrated to DNP servers - the "Welcome" email had an incorrect name server (mistyped) AND the IP address for the control panel was incorrect.
I've had to go to Live Chat to prise information about webmail log in - and now one of my domain mailboxes is returning mail and can't be access through the new webmail (using ip address) and I don't know if it's a technical problem or just because I've no been given the correct information.

Attention to detail is crucial!

[Rock]

I'd like to add to this - I don't think Tech Support give enough information out, and it's not always accurate.
I've just migrated to DNP servers - the "Welcome" email had an incorrect name server (mistyped) AND the IP address for the control panel was incorrect.
I've had to go to Live Chat to prise information about webmail log in - and now one of my domain mailboxes is returning mail and can't be access through the new webmail (using ip address) and I don't know if it's a technical problem or just because I've no been given the correct information.

Attention to detail is crucial!

Hi Nigel,

Apologies for any inconvenience this earlier errors may have caused at a primary stage of this newly setup DNP cluster. I've replied to your ticket [GSU-60476-615] with the complete explanation on what went wrong with the domain's mail services. Let us know if you need any additional information..

[Nige]

Thanks.
I've received your email and explanation (attention to detail!).
However, webmail still not working from here.

[Nige]

Still broken.
I was assured "no downtime" too.
Note that all of the issues so far have been to misconfiguration by Eukhost and giving me incorrect or incomplete information.

[Nige]

And what really drives me mad -

"We have tested the connection from this end and found no errors"

I get this over and over again. No resolution to the problem, just this.

Not helpful. The farce continues - "No downtime" migration = problem after problem.

[Rock]

Thanks.
I've received your email and explanation (attention to detail!).
However, webmail still not working from here.

Still broken.
I was assured "no downtime" too.
Note that all of the issues so far have been to misconfiguration by Eukhost and giving me incorrect or incomplete information.

Hi Nigel,

Webmail is based & works on an IP address [ Under Construction ] which doesn't require DNS interference. As I mentioned in the ticket earlier, emails were tested & found to be working through webmail. What exactly do you mean by webmail being broken here?

And what really drives me mad -

"We have tested the connection from this end and found no errors"

I get this over and over again. No resolution to the problem, just this.

Not helpful. The farce continues - "No downtime" migration = problem after problem.

WRT Ticket ID GSU-60476-615 : You say that emails are working in webmail, but not in your email client, which of course is an issue with the mail configuration settings within the email client only. What mail servers (SMTP & POP3) are you using in that email client? Did you try using the mail server's IP address (109.203.96.7) instead of mail.domain.com for SMTP & POP3 instead within your email client?

It's requested to either update the tickets opened with our technical support dept or approach our 24x7 Live Chat services to get these issues sorted ASAP.

[Nige]

Jack,
I shall respond to the contunuing issues through the ticket system.

But to answer the webmail/email issue:

I was told that my email client settings wouldn't need to change.
I didn't change my email client settings.
Email client stopped working after migration (couldn't connect to POP3 server - SMTP server timed out)
I removed the secure option from POP3 and I can receive emails through the email client
I've changed the SMTP server to the IP address but it still times out.

I keep being told that a particular webmail account works fine. When the user or me tries to log in to webmail using the links sent by eukhost it doesn't work (error message is email address or password is incorrect - but they're not!).

Either I'm missing something or you are!

Thanks

[Nige]

Ticket updated.
I've solved one issue myself already - The Webmail Mailbox - DNP wasn't registering a changed password properly. Hence the "incorrect email address or password" when trying to log in to Webmail. I used the password as originally sent by eukhost, successfully logged into the WebMail mailbox and changed the password from there - which has been successfully retained.
Significantly, the only account with this problem was the one that had been "migrated" by eukhost and given a password on creation. All the other accounts I created myself on DNP and haven't (so far) subsequently changed the password since creation.
Make of that what you will.
Anyway, progress has been made - I just need the full POP3 SMTP settings (I know the domain - I need useable ports and security settings) for the new servers (please don't post them here!) and I think I'm finally good to go!

[Nige]

Oops!
Again spoke to soon.
Will keep it to tickets from now on!

[Nige]

Yet more issues..
To echo the OP's sentiment - Don't promise what you can't deliver! "No Downtime" does not mean faddling around for days changing incorrect settings. If you'd have got it all right in the first place then I'd be lauding you to everyone. As it is, I'm just fed up with it all.

[eUK-Ralf]

Hi Nigel,

We have updated you about the email feature requested in ticket BQT-52974-572.
Please kindly reply with further instructions.