Results 1 to 14 of 14
-
05-15-2007, 12:52 AM #1Junior Guru Wannabe
- Join Date
- Oct 2006
- Posts
- 34
Best uptime possible....100% anyone? :D
I am in need of a reseller account that has the best uptime along with attractive bandwith/space. I am currently with steadfast, they have been fairly good, just a slight problem with one of there servers but was quickly moved over....which recently suffered raid problems causin my sites to be down for over 4hours!. Now im not bashin them, there support has always been excellent and quick but i need uptime, for I almost lost several clients that day. I have just signed up with a reseller account with resellerzoom who are also known to be good and were my second choice when i decided to go with steadfast. I am with their failover plan ....for obvious reasons, however i've been reading some post on here and their support forums and found several people experiencing down time which scares me even more. I like their white labeling but uptime is more important to me.
I am kind of a bit confused now, i've been so busy researchin for a good host that i am loosing developing time on my web application. I've looked at mosso but they seem to have too many constraints...and well rackspace wants min of $400/month.....
Can anyone suggest any advice? uptime and support are very important to me, features should be the same as resellerzoom & steadfast's reseller plans along with php5 and mysql. After long research I found these 2 companies to be the best, has anyone found better?
-
05-15-2007, 01:15 AM #2Web Hosting Master
- Join Date
- Dec 2002
- Location
- Prince Edward Island
- Posts
- 2,289
Rackspace is well worth it if your looking for high end hosting.
-
05-15-2007, 02:43 AM #3Web Hosting Master
- Join Date
- Jun 2003
- Location
- Nova Scotia, Canada
- Posts
- 4,128
100% uptime is tough to offer and maintain, even Rackspace lost their ability to market that they have never had downtime, and that's their major selling point.
My advice would be to find a company that has 1) a good history and 2) a solid 99.9% uptime guarantee to go with that history.
If uptime is important to you, and it appears to be, then your best bet would be to draw up a shortlist of sensible, time-tested companies and slim down your list from there, to get to one provider.
Sensible uptime guarantees are not the hardest things in the world to maintain, support staff are a major part of that. Some companies maintain 100% network uptime but then drop the ball on the human side of things: server management. Be careful, also, when choosing a provider that boasts 100% uptime (and states they have always maintained it). I don't know of a single company in existence that can honestly say this without fibbing. There's always going to be little blips here and there, humans are still involved. The most important factor, though, should be to research how exactly the providers in question handle downtime. Do they provide live updates, do they constantly pass the blame, do they have an open book policy, etc.
Simon
-
05-15-2007, 02:51 AM #4Temporarily Suspended
- Join Date
- Dec 2006
- Location
- East Coast, USA
- Posts
- 458
100% uptime i dont think any host can do it. They are server upgrades, updates need to be done every now and then.
You would be best to have a look in the reseller offer forum to see what kind of host meets your needs.
Also send in pre-sales emails asking about there uptime to see what they tell you.
Best of Luck.
-
05-15-2007, 03:00 AM #5Retired Moderator
- Join Date
- Feb 2005
- Location
- Australia
- Posts
- 5,849
I believe a good reseller host should achieve 99.9%. This is up to 3/4 hour per month downtime on average, but most months you'll do much better than this - then just occasionally something will happen to cause several hours downtime. In essence, sh*t happens.
Leaving a host that you're otherwise happy with because of one such incident doesn't make sense to me.
Realistically, what uptime are you looking for and what are you willing to pay for it? (Hint: 99.99% will cost you many times more than you're paying now)Chris
"Some problems are so complex that you have to be highly intelligent and well informed just to be undecided about them." - Laurence J. Peter
-
05-15-2007, 11:45 AM #6Junior Guru Wannabe
- Join Date
- Oct 2006
- Posts
- 34
thanks for the quick reply
thanks for the quick reply, and I know it is unrealistic to ask for 100%, but I would like the best company close to it. I would like to start with a solid network/company where I get the same uptime status no matter what my plan is, this would be a company i would gradualy move up with. Can anyone suggest companies of such status, where they are known for there uptime?
Maybe folic is right, I can't really blame them too much for such an incident, but for more than 4hours! in one day, and several random downtimes the day before....it just really annoys my clients. Especially since it was done during the day (when my audience is more likely to visit). Can any one give advice on which is better; steadfast or resellerzoom? i've only been with rz for a week or so now.
All in all i guess im just as a typical web developer/designer that just wants his site to always be available.
-
05-15-2007, 02:03 PM #7THE Web Hosting Master
- Join Date
- Jan 2003
- Location
- Chicago, IL
- Posts
- 6,957
I just wanted to note that the issue with the web server you were on was fairly bizarre and has never happened before. Normally with a drive failure we can just take out the drive and rebuild the array in the background and no one will really notice. Here the issue was more complicated as the array had still been rebuilding from a previous error, since if we allowed the system to continue rebuilding, with another potentially bad disk in the array we would be risking data loss, so we took the system offline while the array was being rebuilt, swapped the drive when that was done, and then needed to run fsck on the system, which actually took most of that 4 hours. There was really no way around that.
As a note, we were concerned about the potential for data loss, but even if that event happened and we lost all the data we still do keep daily backups. We simply do not want to resort to using those backups unless absolutely necessary.Karl Zimmerman - Founder & CEO of Steadfast
VMware Virtual Data Center Platform
karl @ steadfast.net - Sales/Support: 312-602-2689
Cloud Hosting, Managed Dedicated Servers, Chicago Colocation, and New Jersey Colocation
-
05-15-2007, 02:25 PM #8Location = SoapBox
- Join Date
- Oct 2003
- Posts
- 6,564
Excellent response from Karl @ steadfast.
If this was the only incident and you have been happy with them, I would surely stick with them.
Your only other option is to pay more for the little things. ie) smaller disk drives mean less disk space allowance, but, much faster times to recover (in case of a raid failure as outlined above) and much faster for fsck to run, etc...
This really is a balance between the resources you want and the reliability/uptime average you want. As others have indicated "stuff" does happen - but, high availability plans (at least good ones) do account for "stuff".
For example - steadfast above have described a situation which can be recovered in 4 hours and more then likely they will still maintain 99.9% for the year on that server or very close to it. If this sort of "worse" case scenario is not adequate, you will need to pay more for an environment where they can recover and maintain a 99.9% average for the month in which this failed array occurs. This will obviously cost more and will most certainly give you less overall disk space to work with.
Anyway, based on what you have said, and based on the excellent response from Karl @steadfast - I would probably stick with them - if you absolutely must have more security then a 4 hour recovery of a failed RAID array (which is certainly WELL above average), then you should look to be paying ALOT more for disk space and you should be asking potential providers how they handle items like raid array failures and what sort of recovery time can be expected. You can be 100% certain that the more disk space you get with an account, the longer the mean time to recovery will always be.www.cartika.com
www.clusterlogics.com - You simply cannot run a hosting company without this software. Backups, Disaster Recovery, Big Data, Virtualization. 20 years of building software that solves your problems
-
05-15-2007, 02:28 PM #9Disabled
- Join Date
- Jul 2006
- Location
- Detroit, MI
- Posts
- 1,962
@OP:
budget?
-
05-15-2007, 04:58 PM #10Founder - Jaguarpc
- Join Date
- Jul 2000
- Location
- Colorado Springs, CO
- Posts
- 2,288
If you arent doing load balancing and failover with geographically separated clusters you can never deliver 100% uptime. Networks are fairly easy to do but hardware is a the red headed step child of uptime.
Greg Landis | Director of Growth Jaguarpc - Unlock Superior Hosting with JaguarPC
Managed Servers - Virtual Private Servers | AMD EPYC Dedicated servers
Follow us @ Facebook.com/Jaguarpc | (888)-338-5261 | greg @ jaguarpc.com
-
05-15-2007, 05:08 PM #11Web Hosting Master
- Join Date
- Feb 2004
- Location
- Sacramento CA
- Posts
- 3,513
I have to say that at your budget you have to be realistic about downtime. You could be paying a great deal more and still experience a 4 hour downtime. Even the host with the best uptime can not control things outside of his environment, things happen. If this is the only significant issue you've had with Steadfast (who as far as I'm aware of has an excellent rep) I would not be so fast to jump ship.
Other then that have you taken a look at clustered services like Cartika offers? Not an expert but thats one way of minimizing your downtime.
-
05-16-2007, 11:40 AM #12Junior Guru Wannabe
- Join Date
- Oct 2006
- Posts
- 34
info
Thanks Karl for the detail explaination, I wish it was stated in an email or on the steadfast website for those that wanted to know what really happened.
I guess your right, the only draw back I see so far is the H-sphere cpanel (which I think is amazing btw) but if i need to move to a vps or dedicate server I will most likely use cpanel or something else other than hsphere, and migrating the accounts seems to be difficult.
Is there a tool to aid migration from hsphere to other control panels?
-
05-16-2007, 01:00 PM #13THE Web Hosting Master
- Join Date
- Jan 2003
- Location
- Chicago, IL
- Posts
- 6,957
Karl Zimmerman - Founder & CEO of Steadfast
VMware Virtual Data Center Platform
karl @ steadfast.net - Sales/Support: 312-602-2689
Cloud Hosting, Managed Dedicated Servers, Chicago Colocation, and New Jersey Colocation
-
05-16-2007, 01:14 PM #14Location = SoapBox
- Join Date
- Oct 2003
- Posts
- 6,564
I guess your right, the only draw back I see so far is the H-sphere cpanel (which I think is amazing btw) but if i need to move to a vps or dedicate server I will most likely use cpanel or something else other than hsphere
Is there a tool to aid migration from hsphere to other control panels?www.cartika.com
www.clusterlogics.com - You simply cannot run a hosting company without this software. Backups, Disaster Recovery, Big Data, Virtualization. 20 years of building software that solves your problems