New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Common issue with most vps providers?
pioneernetworks
Member
What is one of the most common issues with vps providers?
I would have to say Support
Comments
Pinhdom caught vps being down, open ticket. Provider denies it.
pingdom is a big pile of crap, I do no accept it's results, show me something real for proof
Forgetting to update their websites.
I haven't really found support to be an issue for any of the providers I've found on here.
For example?
ping -c 5 or better results from some other location, and if I can then reproduce that from LA or Charlotte, or a traceroute that dies along the way, and please do not send screen shots, it's text, copy and paste it.
I'd have to agree with miTgiB. Several companies where I have managed the support departments have had problems with customers who "monitor ping" via pingdom.
I'm sure if someone put together a really good multi location nagios cluster and then built a front end for it, it'd be far far better.
We had a client who opened a support ticket because Pingdom said their VPS was down WHILE THEY WERE LOGGED INTO IT! We use BinaryCanary for our own monitoring and it always proves Pingdom wrong.
I personally use nagios for any monitoring.
The VPS being down or a service on your VPS being down? I get so many idiots complaining about Pingdom alerting them of their VPS being down and it turns out that they're just monitoring http and Apache crashed. Unless my Pingdom report for the entire node shows something, I'm not going to accept a client's report.
You're lucky you don't use solusvm. Solus used to have a really brutal monitoring setup for graphs that had the master do all the pulls, instead of nodes doing push, and graphs were always missing tons of data, even on idle nodes.
It still does it and people try to use the graphs at other hosts to claim SLA's, but at this point I don't think anyone trusts them.
Francisco
I stopped looking at SolusVM graphs last year. Complete waste of time. Every single one of our clients shows at least 1 Petabit of network transfer if you got 1-2 months back making the rest of the graph impossible to even read. Needless to say, we do not have a 1Pbps port on any of our servers.
I'd never heard of them until now - interesting. Why did you pick them?
I used them in the past when they first opened and when SD first started. Their monitoring and reporting are excellent and accurate. Their alerting is not so much, they'll send an alert almost 10 minutes after you tell them to so I only use them for monitoring and reporting and use my own systems for alerts.
Overselling is probably the most feared among the LEB buyers. If you want to generate a profit at low volume with the prices thrown around here, without owning your own hardware, overselling is probably the easiest way to do it and that can tempt people who get into this market.
Lack of company structure / business sense / common sense
"Hey I got a bright idea! Let's start up a hosting company, it'll be fun!"
And indeed it has!
Even without owning your hardware it is possible, There are many affordable server providers out there, VolumeDrive, DataShack, FDCServers, Worldstream.
Host time not sync or timezone incorrect.
There have two providers in my experience have this problem and solved until I submit a ticket to them.
I just think they don't ensure this simple basic server environment is not acceptable.
Host time not synced is a provider problem, incorrect time zone is not a providers problem.
Most common issue is inconsistent network performance. On my experience, sometimes it start great and goes south; sometimes it start not so well but becomes better after some (usually well advertised) provider network upgrade. Second common issue: some OpenVZ providers have very old templates, or have "experimental" templates that never become standard, tested templates.
i couldn't agree more.
sent an email notification for a server that doesn't have any issue, and did nothing when the server goes down. period.
vps down. I had pingdom monitoring my services port at 15 mins interval and only notifies me if its down consecutively (30 mins), not http though.ssh is one of it.Explain that i could;t vnc and ssh to it. Provider still denies it.
I noticed pingdom is not accurate, esp if it was to monitor europe location like germany, netherlands. Other locations they are quite accurate.
Sure if you want a cookie cutter it. VD is 'kind of' a unique location given how hard it is to get stock and or a reply from support :P
Francisco
We don't accept Pingdom complaints either. It shows a lot of drop-outs ( http://stats.pingdom.com/1bdrqltaboji/449948/2012/05 ) that I don't see mirrored on anything else we have.
One of our VPS customer has a quite busy website that gets monitored with Pingdom, AND has an automatic script to email us when it's "down". When it's only "down" because he's using Apache with DSO or suPHP or something slow and it can't handle the load.
We've tried selling him on our shared hosting, which for two dollars more per month, will handle hundreds of requests per second no sweat. We ended up having to add a filter to ignore his email.
I can't comment on other hosting providers but if you're not able to connect to your own VPS then that's usually a good sign that it's inaccessible, possibly even offline.
LoL! I would have done the same thing. Crying wolf to many times will bite you in the arse, automating the calling of wolf is even worse!
@KuJoe -
@Aldryic LOL! (Image isn't working so here's a link: http://oglaf.com/wolf/)
aww, that's a shame. Here's the direct link: http://oglaf.com/media/comic/wolf.jpg
@Aldryic Direct link doesn't work either (adds a dozen or so slashes, probably to prevent hotlinking).