Re: [cpx] tech support and CPX issues - cross post

[suzanne newman <snewman@xxxxxxxxxxxxx>]

I have been experiencing this problem since November on various Linux VPS
servers. The problem, as explained to me is that on reboot, dovecot does not
start up. Mail comes into the server, but does not get placed into the
mailboxes. I was told (in November 09) that this is a known problem and that
a permanent fix was being worked on.

When this problem happens (so far 7 times on 4 different servers) I call in
with the same case number so there is a record of how many times this has
happened. I have not had a problem with tech support on this.

Verio, if you are listening - it would be helpful to have a place in the
back room listing these known problems and tracking what is being done to
fix them. This would be helpful to resellers, as well as to tech support, so
we don't all go crazy trying to figure it out. A documented known problem
with a known fix (even if short term) would take less tech support time.
Also, a documented list of things to be fixed would help these problems not
fall through the cracks - maybe even, a person could be assigned to oversee
this list.

Suzanne


On 4/23/10 11:23 AM, "Brian White" <bwhite@xxxxxxxxxx> wrote:

> (cross posted from the business@xxxxxxxxxxxxxxxx list - meant to do
> this with the original post and forgot)
> 
> ------------------
> 
> Recently there were threads on the Provider talk lists about both CPX
> and its long time issues, as well as certain issues regarding tech
> support. Because I know that some Verio employees that both care
> about and can possibly help with these issues from a larger
> perspective monitor this list, I'm going to explain something that
> just happened to me that involves both of these issues. Also, maybe
> another reseller reading this will be able to save some time and
> headaches if they encounter the same thing in the future. My intent
> is NOT to complain about any individual at Verio or a specific
> support rep, it's simply to help these issues come to light because
> they both need addressed/fixed.
> 
> A Linux VPS client contacted me the other day with two separate email
> related issues:
> 
> 1. No one could connect to their POP accounts, Outlook returned an
> error indicating that the connection was being rejected because there
> were "multiple devices" trying to connect at the same time. We
> quickly ruled out quota issues and malformed email files as the
> cause. We also verified that email was still arriving on the server
> correctly, it wasn't bouncing or being rejected for any reason.
> 
> 2. At the same time, each user that logged into CPX and tried to view
> their email via that interface was shown a message by CPX stating
> that their inbox was "empty" (even though it wasn't), and looking at
> the folder display list showed zero folders (which also wasn't true).
> 
> After some initial troubleshooting I called in to support and the
> first tech 1 that I talked to tried for about 15-20 minutes to figure
> out what the problem was before passing me off to a tech 2 (I think
> it was a tech 2, I admit my notes don't indicate that).
> 
> After explaining the issue to him, suddenly we got cut off. No
> problem, I thought, since they always ask for "your callback number
> in case we get cut off", he'll call me right back. Nope, never
> happened. (HELLO, someone at Verio in management, am I wrong in
> thinking/expecting that I SHOULD have gotten a call back?)
> 
> So after waiting a little while I called back in and got another tech
> 1 (3rd support rep I've talked to at this point), who while he was
> waiting for some info to come up asked me to explain the problems,
> which I did (again), and he too tried to figure out what was going on
> and couldn't, and he then told me he'd need to "escalate the issue"
> to the tech 2 queue.
> 
> So the issue gets pushed into the tech 2 queue and I wait. About 4-5
> hours later I haven't heard anything or received an email from
> support, so I call back in and get the 4th different tech rep for
> this case. He looks at the issue and tells me that no one at tech
> level 2 has looked at the case yet (hello Verio management again -
> just so I know, is tech level 2 so overloaded that a case where no
> one on a VPS can get their email should have to sit and wait that
> long? What if I hadn't called back in then, how much longer would it
> have been?), but he'd be happy to look at it for me. So I once again
> explain what's going on, and he puts me on hold and comes back a few
> minutes later and the problem is fixed (!). Major kudos to that
> support rep.
> 
> Unlike the 3 other tech reps that looked at it before, he knew right
> away what to look for (a CPX/dovecot memory error that was preventing
> any email connections from being made, the Web site itself on the
> server was unaffected) and he simply rebooted the server and
> everything was back to normal.
> 
> Granted, maybe you can say I should have thought of just rebooting
> the server myself (like when I unplug my HD cable box so it will
> reboot as soon as I notice the picture starting to break up, but
> that's another tech support story), but if so, why didn't the first 3
> support reps think of that 4-5 hours earlier? For whatever reason,
> this 4th support rep was familiar with CPX memory problems (and the
> others weren't?), and as he said: "*especially* on the Linux servers".
> 
> Unfortunately, as many resellers have known for a long time, this
> kind of hoarded specific knowledge is all too common in the support
> dept. overall. I know they have their own knowledgebase, and maybe
> sometimes it does help them, but it's been my experience over the
> years that how fast you get a support case resolved often depends on
> the luck of the draw on who answers the phone or first reads your
> email, and many, many times there's been cases like this where it
> seems like what should be common knowledge (or at least easily
> discoverable) is not, and that's not even getting into the issue of
> why CPX is still having these problems in the first place.
> 
> One last question for Verio management: I certainly realize that in
> any tech support dept. everyone can't know everything, but this kind
> of incident is not uncommon, so do you at least agree that it should
> be uncommon? Understand that my frustration comes from how many times
> over the years something like this has happened, not just this
> individual case. Maybe I'm wrong about something, or simply just have
> a misunderstanding involving expectations, but if so, I welcome the
> education.
> 
> Regards,
> Brian White
> ======================================================================
> This is <cpx@xxxxxxxxxxxxx>      <http://www.groupmail.org/lists/cpx/>
> Before posting a question, please search the archives (see above URL).


======================================================================
This is <cpx@xxxxxxxxxxxxx>      <http://www.groupmail.org/lists/cpx/>
Before posting a question, please search the archives (see above URL).