Server downtime - details please!!!

In reference to this thread:
http://support.jodohost.com/showthread.php?t=7194

I’ve posted on this subject before. Here we have a downtime of 30 minutes for a server. The update is “IIS is not responding on win6”, then “Win6 IIS issue has been resolved”.

While technically correct, this kind of “information” is NOT sufficient for us as resellers in order to give our clients the impression we know what is going on. I could have told you that “Win6 was down” and “Win6 was resolved” without reading the forum, and indeed even before I got the email that it was down in the first place!

The router maintenance over the weekend was supposed to be a downtime of 15 minutes (granted, the entire system), but we got constant updates throughout the process.

We can at least have some explanation to give to our clients of why a server was down for 30 minutes. From the perspective of our clients, or indeed our own service domain, it’s no less important.

Please don’t get lazy with this, it’s very important for us resellers - it’s bad enough that, by the time we even SEE the issue in out inbox, clients have called asking why their site is down and we have to find something to tell them because this idea of posting urgent issues on a forum is inherently flawed.

If you detect some stress in my tone, it’s been passed down from a client to their “hosting provider”, so in turn I pass it down to mine. :slight_smile:

Please give us something intelligent to tell a client, even if you don’t want to send network outage emails immediately via a more efficient and immediate means. At least we can call them after the event and have some explanation.

Thank you.

[edit] Just has to put some more bold tags in… :stuck_out_tongue:

I’ll check out what was the issue. Win6 has not had any such outage in a long time, I am suspecting its abuse but I’ll talk with the admin who took care of the issue and update the thread

Please give us something intelligent to tell a client, even if you don’t want to send network outage emails immediately via a more efficient and immediate means. At least we can call them after the event and have some explanation.

We have no means to make server-specific announcements via email. Hence, if its an unexpected outage, it would be announced on the forum. Prolonged and scheduled outages are always sent via email.

If your client reports an outage, the best thing to do is load up this forum and you’ll find any information about a server-wide outage on this forum. Not many companies give you up to the minute notification on anything that happens, on a public forum.

However yes, a reason is lacking in that notification and I’ll see to that
I acknowledge resellers are in a difficult position when it comes to an outage, but we are doing our best to make information available quickly on this forum and via email if possible

I’m a new Jodohost reseller. The fact that I got 2 emails on 3/30/06 telling there will be down time and an ESTIMATE on how long, to is me great. I host a dedicated server with another company (cyber------) and am never informed about down time. Believe me they will soon be history.
As for what to tell your customer, be non-technical, and make it look like you are improving their service. I am sorry for the down time due to the upgrade to our servers which will improve our service there may be a down time of (whatever amount of time) …
As for your service account the same applies. 1. you were notified by jodohost and 2. this is an improvement to your service. No down time no improvements etc.

Sorry as a new reseller with BAD experiences with other companies I do not understand your problem.

Thanks for the response, Yash. Sorry for the note of frustration!

I guess one thing I can do is set up a Cron job to ping the servers I use, say every 2 minutes, and email me if there’s a problem. If it works well, I can set up something for other other resellers to subscribe to the notices for their servers…

LOL got you worried now? hmm, might make that a 30-second ping interval just to be safe… :stuck_out_tongue:

[QUOTE=antic]

I guess one thing I can do is set up a Cron job to ping the servers I use, say every 2 minutes, and email me if there’s a problem. If it works well, I can set up something for other other resellers to subscribe to the notices for their servers…

[/QUOTE]

Ping rarely goes down. What goes down is web services like it did in this case. We have a system that monitors ASP, ASP.NET and PHP on all our web servers, and notified us within 120 seconds of an outage..

The live status of servers, you can make available on your website. Someone posted code for (check announcements section)

[QUOTE=jhpweb]
As for your service account the same applies. 1. you were notified by jodohost and 2. this is an improvement to your service. No down time no improvements etc.
[/QUOTE]

jhpweb: Totally agree, Jodo is at the front of the pack in terms of service in comarison to the vast majority of hosts. I do know at least one other that uses forums exactly this way to post notices and for client discussion, but they are definitely the exception to the rule. But hey, no poking no improvements either. :boink: :slight_smile:

Still, it’s always best to let a client know what is happening instead of waiting for them to call you. And I never fib to my clients… that’s opening the door to bigger problems. If you tell them it’s intentional downtime for “making improvements”, they’ll just ask you why you didn’t notify them in advance, the same as we’d do if Jodo didn’t tell us about a router upgrade.

I think this issue will continue to be raised in the future too. It would take the techs seconds more to add some details into the posts, so I don’t think it’s a big request. Why would anybody do something half way? Make little sense.

That advice about lying has to be the worst advice ever.

Well honestly, I was not around for win6 issue, but we don’t know WHAT it was, it was not working on termserv, and the KVM was showing nothing, we had to APC reboot it.

It seems the server just quit responding, I don’t know what else to say, no error logs, no logs at all during the period it was not responding.

The techs on shift were trying to contact me to check the server, but the one time in MONTHS I forgot my phone I was away for 30 minutes with my family eating lunch on sunday, had I been around the outage probably would have been 10 minutes or less.

I am very sorry about that matter.

I will be discussing with the weekend staff the proper ways to handle this if I am not available, as sometimes one simply can not be.

No problem Stephen, and thanks very much for the response! These things do happen.

If it’s the case that you don’t know why it wasn’t responding, then it’s cool to just say that in the outage post instead of saying nothing at all, because then of course we’re left to wonder, and some trouble-maker comes along and complains about it. :wink:

It’s just a procedural thing and really only amounts to a few extra words, even if it’s simply, “the server is back up, reason is unknown at this stage so we will be keeping a close eye on it.”

That makes everyone feel better, and personally I wouldn’t be afraid to tell a client that, as long as they know I’m on the ball.

Thanks again… didn’t mean my post to cause a big stir. :slight_smile: It’s a small point but an important one for us.

Cheers!