Results 26 to 50 of 71
-
04-11-2014, 04:56 AM #26Junior Guru Wannabe
- Join Date
- Jul 2010
- Posts
- 69
-
04-11-2014, 11:23 AM #27Web Hosting Master
- Join Date
- May 2003
- Posts
- 1,708
You aren't the only one, all of our call backs break often through their updates and you have to go back in and re-save them. We do not get the MySQL errors, would be easier if there was an error generated somewhere on the callback issues, but Hostbill is full of stupid stuff like this that makes it a pain.
-
04-11-2014, 03:53 PM #28Newbie
- Join Date
- Jan 2014
- Location
- Los Angeles, CA
- Posts
- 12
We've been using OA 3.2 since shortly after release with IS. After the upgrade we had some hiccups, but all of those seem to be solved now.
-
04-15-2014, 06:55 AM #29Web Hosting Master
- Join Date
- Sep 2002
- Location
- Earth
- Posts
- 2,533
I am thinking of setting up an unofficial mail list where customers of OnApp can post to, for discussion around Onapp. I'd be a technical list for answering questions such as "have you upgraded", etc. Just so your not 100% reliant on Onapp.
I also believe it would add some piece of mind because it freaks me out when we are having issues and Onapp come across that nobody else is having the same issue.
Thoughts?
-
04-15-2014, 06:56 AM #30Disabled
- Join Date
- Aug 2011
- Location
- Dub,Lon,Dal,Chi,NY,LA
- Posts
- 1,839
There is an onapp group on LinkedIn
-
04-15-2014, 01:55 PM #31Newbie
- Join Date
- Jan 2014
- Location
- Los Angeles, CA
- Posts
- 12
We recently started the group "OnApp Users" on LinkedIn for just such a reason. Please feel free to join.
-
04-15-2014, 06:55 PM #32Junior Guru Wannabe
- Join Date
- Dec 2013
- Location
- Sydney, Australia
- Posts
- 47
-
04-30-2014, 12:39 PM #33Junior Guru Wannabe
- Join Date
- Jul 2010
- Posts
- 69
-
04-30-2014, 05:06 PM #34
had them upgrade us a couple weeks ago.
WORST upgrade ever. No, really
Tech came in, with no notice, no warning, just started upgrading. We knew it was going to happen, just not when.
When VMs started not responding (due to load caused by upgrade), they were rebooted. Tech cancelled the reboots
When confronted, tech just said
Calm down and let me finish
Adding to all that, things just haven't worked the same since the upgrade.
-
05-01-2014, 05:45 AM #35Junior Guru Wannabe
- Join Date
- Feb 2014
- Location
- UK
- Posts
- 60
An upgrade should not impact the operation of virtual machines, most of the actual effort is on the control panel itself and Hypervisor upgrades are extremely quick and straight forward, ultimately it's primarily package upgrades, it sounds like something else was contributing to this issue?
The Cloud Simplified (http://www.thecloudsimplified.com)
Public & Private IaaS Cloud Solutions, N+1 Public Cloud Powered By IBM & OnApp
Enterprise High Availability, IPv4 & IPv6 Mission Critical Platforms As Standard
Follow us on Facebook & Twitter @CloudSimplified
-
05-01-2014, 02:51 PM #36
-
05-01-2014, 03:53 PM #37Randy
- Join Date
- Aug 2006
- Location
- Ashburn VA, San Diego CA
- Posts
- 4,615
First step in upgrading any cloud infrastructure is temporarily disabling HA... I'm surprised the VMs were auto-rebooted during the upgrade - or is it your own separate system that bounces VMs based on load?
Fast Serv Networks, LLC | AS29889 | DDOS Protected | Managed Cloud, Streaming, Dedicated Servers, Colo by-the-U
Since 2003 - Ashburn VA + San Diego CA Datacenters
-
05-01-2014, 03:59 PM #38
Nothing automated about it. It's all human (even at 3am )
I received notifications that they were unavailable due to load
I attempted to reboot when they couldn't be reached via SSH. It's what you're supposed to do, keep client sites online.
-
05-01-2014, 05:16 PM #39
Thats weird, all the updates we've done
3.08 to 3.1
3.1 to 3.1.2
3.1.2 to 3.2
No virtual machines were affected whatsoever, and no hypervisors were rebooted.
We do use KVM , i dont know if its different with other platforms, i believe with Xen its a different scenario if im not mistaken.
On a side note, we do our own upgrades, we prefer not to rely on Onapp schedules which most of the time dont work for us, this since you cant really instruct them at what time to do them.█ RACKNATION: Costa Rica DDOS Protected Servers & Colocation - [ https://www.racknation.cr ]
-
05-01-2014, 07:58 PM #40Aspiring Evangelist
- Join Date
- Nov 2012
- Posts
- 428
I did my own upgrades for awhile but when I ran into issues, I was told "the upgrade team works between blah blah blah" and I would be dead in the water until the next business day or until I really pushed for someone to look at it.
-
05-01-2014, 09:47 PM #41Web Hosting Master
- Join Date
- Jan 2005
- Location
- Toronto, Canada
- Posts
- 588
one thing we found with onapp outside of the bugs is its a hardware hog on everything. We are testing other cloud solutions as well right now, my techs are not 100% happy with onAPP, i find the support good for most of the time.
Rebel Networks
Geek Powered Hosting - The Cloud Made Simple
SuperHero 24 |7 | 365 Support
Please Visit us @ www.rebelnetworks.com
-
05-01-2014, 10:04 PM #42█ RACKNATION: Costa Rica DDOS Protected Servers & Colocation - [ https://www.racknation.cr ]
-
05-05-2014, 06:46 PM #43Aaron Wendel
Wholesale Internet, Inc. - http://www.wholesaleinternet.net
Kansas City Internet eXchange - http://www.kcix.net
-
05-05-2014, 09:18 PM #44Junior Guru Wannabe
- Join Date
- Feb 2014
- Posts
- 49
Onapp = ONSHIT
Wont say much but I wont recommend it to no one. Not even my worst enemy.
-
05-06-2014, 12:18 AM #45Web Hosting Master
- Join Date
- Oct 2001
- Location
- Ohio
- Posts
- 8,535
-
05-06-2014, 04:11 AM #46Web Hosting Master
- Join Date
- Sep 2002
- Location
- Earth
- Posts
- 2,533
-
05-06-2014, 12:09 PM #47
Yes it is a great platform ...
Sad to see people making generalisations in such a way instead of elaborating what in specific happened that you are so unhappy.
Weve had our bumps in the way with onapp as well as any other platform we've used, is simply the nature of the beast.█ RACKNATION: Costa Rica DDOS Protected Servers & Colocation - [ https://www.racknation.cr ]
-
05-06-2014, 04:24 PM #48Web Hosting Master
- Join Date
- Apr 2004
- Location
- New Hampshire
- Posts
- 773
-
05-06-2014, 05:52 PM #49Web Hosting Master
- Join Date
- Jan 2005
- Location
- Toronto, Canada
- Posts
- 588
I think there are just way to much and to many issues to list with onapp, I think that summarizes it. You can reduce hardware requirements and improve over all resources with Apache Cloud stack and for a paid product VMware vSphere. We are going though testing modes right now for both products
Rebel Networks
Geek Powered Hosting - The Cloud Made Simple
SuperHero 24 |7 | 365 Support
Please Visit us @ www.rebelnetworks.com
-
05-06-2014, 07:54 PM #50Junior Guru Wannabe
- Join Date
- Feb 2014
- Posts
- 49
Every beast has its nature; however, our onapp is simply a beast.
Some may say hardware is the cause of problems, or network. True, but when the hardware requirements were given to you by onapp, and have an entire DATA CENTER network at your disposal. What's to say then? Specially when Onapp made the setup for you?
To keep things short I will go straight to some annoying points that have hurt our reputation.
Random HV kernel panicking due to memory swapping. We were asked to implement more HV which will be use for swapping purpose. for those that use Onapp and use the requirement hardware you know how expensive it is to add one HV. Not to say how annoying it is having to deal with something that could had been prevented from the very beginning.
Kernel panics due to "nf_conntrack: table full, dropping packet". The nf_contrack were too low by default which lead to the HV to overload and kernel panic. Series of dev fixes and sysct.conf changes reduced the nf_conntrack crashes. We still have them but are being worked on by the devs.
Not to say how long it takes to get all of the VM machines back online when a HV crashes.The onapp panel cant' handle high amounts of queues, so if you try starting 10 machines at once you will guarantee locked up the queue. Then having to deal with degraded vdisk, or zombie process due to the crash.
Network stops responding on VM machines, reasons unexplained. We were asked to wait two months for an upgrade that supposedly patches the random loss of network. Not to say how fun it was to implement the updates, took us almost two months to update to then be told that we need to upgrade again. LOL
The original datastore implementation done by the onapp admins was horribly wrong we ended with a degraded datastore that was not redundant with extremely poor performance. We had to purchase 4 more HV to make the setup redundant, and still no performance although improvements have been made.
Forgot to say the extremely high LOAD the datastore cause, and how it cuts the VM performance to a fraction due to the series of reads and writes it makes.
I have a whole entire book of problems I can keep posting but the truth is that Onapp is the better product out there for fast Virtual machine deployment. Its great if it works I suppose. Support is okay, until you have to argue with them that the problems you are having are not server related but HV related.
Not to say that you don't have control of the product problems, I bet a datastore node stops responding, or the datastore API stops running theres no way you will fix the issue unless you worked for onapp or happen to know their product better than the employees, or open a ticket.
These are the problems I deal with on daily basics, and to be truthful I wish they fix all of our problems. I just want to move on to more important matters than having to deal wit the same o BS.Last edited by Photonhelper; 05-06-2014 at 08:07 PM.
Similar Threads
-
[FEATURED] OnApp Cloud: Hardware SAN v OnApp Storage (SANity)?
By cwl@apaqdigital in forum Colocation, Data Centers, IP Space and NetworksReplies: 212Last Post: 04-24-2013, 05:59 PM -
Webbynode : beta testing program , up to 50% discounts for early adopters
By ctaborda in forum VPS Hosting OffersReplies: 0Last Post: 11-03-2008, 02:58 AM -
webbynode : beta testing program , up to 50% discounts for early adopters.
By ctaborda in forum VPS Hosting OffersReplies: 18Last Post: 10-30-2008, 02:52 PM -
Is it too early ?
By inquisitive in forum Web Hosting LoungeReplies: 5Last Post: 12-31-2001, 06:55 PM