I am getting ready to order my server through a reputable company and am new to the colo thing and would like to here from some others letting me know if this server will be plenty for what I want. I have been using a dedicated server with Liquid Web for the past two years. The server I have with them has the following:
The server I am ordering is listed below. This has more memory and hard drive space then my current dedicated server. I am pretty sure that the 2 dual Xeon Woodcrest 5150’s will blow away the 2 dual Opteron 242’s. Am I correct in assuming this?
I would like an opinion on with this is a good server or is it to muck or to little. My train of thought when building and ordering this server was to build one that had plenty of memory and storage along with a good processor that I would not have to worry about for a while. I am thinking of changing my hard drives to 250G hard drives. So was this a good build for my train of thought?
2 dual Intel Xeon Woodcrest 5150 socket 771, 4x 2.66Ghz/2M L2 cores
Supermicro X7DVL-E dual socket 771 server board, Intel 5000V chipset
8G (4x 2G) ECC FB-DIMM DDR2-533, 2x open
2x Western Digital 160-Gig SATA-II RAID Edition, 7200rpm, 16M Buffer
2x Western Digital 160-Gig SATA-II, 7200rpm, 16M Buffer
on-board Intel ICH7R 4-port SATA controller
on-board Intel dual-port gigabit NICs
on-board ATI ES1000 Video, 16M
24x slim CD-ROM, no floppy drive
SuperMicro SC813MT-420B 1U Rack-mount Chassis, Black
4x Hot-swap HDD Carriers & 1x4 SATA Backplane
supermicro 1U 420w Power Supply
1U Mounting Rails/Brackets Kit included
I'm considering building the following 1U rack server for colocation. I'm completely new to rack servers and colocation so I would be very grateful to hear your opinions:
Intel Core 2 Quad Q6600 CPU ASUS P5B-VM SE Motherboard 8GB PC5300 667MHz Non-ECC RAM 500GB HDD 345W Power Supply
I also have a few specific questions:Are there any differences between the Q6600 and it's Xeon equivalent? (especially power wise?)
Are there any disadvantages using a desktop motherboard in a (1U) server? (again, especially power wise?)
I've read that non-ECC memory is a lot more acceptable for servers now than it was a few years ago, is this true?
Does a 500GB drive use much more power than a 160GB drive?
How many watts of power can I expect the server to use?
Does a 1U chassis usually come with uATX mounting points?
I have come across this vps/virtual-grid solution, which at a first glance seems something like a vmvare esx infrastructure: thegridlayer.com does not provide very detailed information but for the little I understood they guarantee you almost 100% uptime because if your vps hardware fails it will be moved to a new node in minutes withouth any intervention. The grid structure also seems to easen scaling.
I am yet to understand what kind of virtualization they offer, but they claim the ability to build your own kernel, so it would seem a complete one.
Prices are quite cheap too: 59$ for a 512mb ram, 80gb disk, 2TB/bw virtual machine... Unfortunately they only seem to run a single prehistoric centos 4.4 distribution, but if you really can compile your own kernel and do all the tweakings would not be a big deal.
I have 2 dedicated servers and one VPS with liquidweb. Lately I am facing quite an issue with support and I think I am posting 2nd time about liquidweb support.
Its all about my VPS, which use to hold 45 clients consumming around 50 % disk space and around 30 % bandwidth.
On 3rd of April VPS started to shoot emails about services failure and continuously went for 2 days. Then on finally 5th April, it went down. So I reported this to liquidweb support by creating a ticket(I created the ticket by shooting an email to support@liquidweb.com not by logging into PIMS). I got an reply in around 1 hr 20 mins that server is up. So i checked it and yes it was up, but with lots of issues. Still were getting service failure emails as they are failing and above all my root password stopped working. So couldn't login into WHM that time.
I reported this back on 6th April. No reply on 6th of April. On 7th april, I sent a reminder about the issue. They repiled by asking for my last 4 digits of credit card for verification purpose (Designatation of that person was written Web Designer) . Now when I ordered this using paypal, there is no question of last 4 digits of credit card. So I politely replied them back within 4 hours that I have used paypal for ordering the VPS.
They didn't reply back on 7th, not on 8th. Finally on 9th, when I lost few customers because I can't login into WHM, I wrote a bit stringly worded post in that support ticket.
To quote myself.....
"OPPS!!!....This has been a tremendous lack of response from you guys. It seems that you don't have support guys anymore. I have lost few customers because I can't login into my WHM. And it seems that you have changed my root password and now not letting me know about this.
Suddenly one web designer is replying my ticket.
Whats is happening there?
Let me know my root password immediately.
Thanks Krish"
Now after posting this reply, I decided to call them. So within 30 mins of this post I called them. Jason, a system admin replied and we talked politely about the issue. He accepted over phone that it was big mess.
Why big mess?????...because when he restored my old password and I logged into WHM....I don't see any account left . All data lost...everything gone as it appeared there is nothing left.
So he took up the issue of restoring account. He claimed that server was rooted. Now how come after so many days and communications they discovered my server was rooted?
Then he started restoring accounts....all accounts came back but with a loss of around whopping 80% loss of data. Most of customers started complaining about loss of emails and data. On 10th again I reported them about such loss and mysql which was not working. They fixed the mysql issue, but no database was left.
Then I got busy on 11th, could not follow it up as I should have been doing. But still as I was waiting replies from my customer about how much is the loss (I was telling them that we should be able to get them back from backup).
So I reported back to support on 11th as follows...I quote myself...
"I am waiting for reply from them. I noticed one thing. That the content is not restored properly...as there are only 13% of space used right..which should be around 50%."
Our company uses ProductCart made by Early Impact (www.earlyimpact.com) for our shopping cart system. People on Early Impact's forums have recommended JoltDigital.com for web hosting for ProductCart.
Has anyone else used/is using JoltDigital.com or know of someone who does?
In your opinion, how are they as a hosting company: Customer support Uptime/reliability Feature reliability (ie- database management, scripting & server languages, etc.) Easy to use...
1)I bought a vps from them, got it promptly, except it was set to the wrong port.
2)I requested it reinstalled without there 'extras' so my network admin could do it himself. They do not offer a reboot in there power panel.
3)They reinstalled it, but changed my root password, also it is still set to the wrong port.
4)I then spoke to there live support, they started talking then disappeared for about 10 minutes to chat on the phone without even telling me, leaving a paying customer sitting there thinking 'what the hell'.
5)I requested the VPS be set to my original password by email, and they haven't I just wanted a clean VPS to do my have my own install done on... how was that so much to ask of them?
I have now sent in another password request, but my advice to you all, steer the hell away!
Brand new to posting on this forum, but been browsing for a while. Also brand new to VPS, but i recently decided to take my first plunge into the VPS world. I've been using Iron Mountain VPS for about 3 weeks now, and since there seems to be very little talk about their services on this forum, I thought it would be useful to post my opinions, and maybe get the opinions of some others about this provider.
My Background: I began my search for web hosting because I had been hosting my own email server at home, but began to have several issues keeping the server up and running (power outages, network outages, power supply failures, etc...). My wife was complaining about her email being down, so I began looking for a more reliable solution I'm also a software developer, so I also had some interest in running some "hobby" type j2EE programs, so I began to look at VPS solutions. Since I am pretty technically inclined, I was just looking for a low cost unmanaged VPS solution with at least reasonable reliability that was powerful enough to run j2ee applications, host email, and run a couple other services like hosting my digital photos and setting up a versioning control repository (svn). I actually took the time to make a speadsheet to compare over 10 VPS providers, and eventually found Iron Mountain to be the best solution for me.
Setup: Their sales department was VERY quick to answer my questions, and very accomodating to my requests. I purchased their Pearl package online, and once the transaction was complete, I recieved an email that I could already log into my control panel. Here is the timline that things happened:
11:35am - Purchase transaction complete 11:36am - Recieve email on my billing info, and how to get into my VPS control panel. 11:48am - Recieved email that my VPS has been initialized, and I can no SSH to my instance 12:22pm - Recieve email that they have already setup a MySQL database for me (on their servers), and it is ready to use (without me even asking mind you). 4:31pm - I send an email (not even a support ticket) saying that I actaully prefer PostgresSQL 4:40pm - Recieve email that they will setup a Postgres instance for me 5:34pm - My PostgresSQL instance is up and ready to go (on their servers!)
So in summary, that is 13 minutes from purchase to having my instance up and running. 34 minutes additional to have a database ready for me to use (without me even asking for one). And another hour of time it took them to get a Postgres instance up for me... Not bad
Perfromance/Reliability: Let me preface this by saying that I am a father of 2 with a full-time job, so I really don't have time to run benchmarks and such. Also, my VPS is more of a personal site, so I don't have to worry about reliability as much as say a reseller or ecommerce customer. So I'm only basing these comments on preception and my knowledge of the technology Iron Mountain uses. In genereal, I have experienced NO downtime, and performance seems VERY quick. My J2EE application almost seems to run faster off these servers than it does on my localhost development environment... which is good enough for me.
Now let me rant about why I am very comfortable with their ability to continue to have great perfromance and reliability. One key here is that this is clustered web hosting! This means they have (for example) 30 servers working together like 1 giant computer. This giant computer runs all of their customer's VPS instances. If one server were to fail, nobody is losing service, it just means that there are now only 29 servers making up this one giant computer until they can get #30 back up and running. This not only protects us customers against one possible cause of downtime, but it also allows Iron Mountain to do server maintanence and upgrades without interupting any of their customer's service. My employeer has a similar setup in their datacenter, so I am semi-knowledgable on the benefits of a cluster. Now their service could still go down because of things like network issues (which is often out of the control of the VPS provider), but at least it removes risk at one point of failure. If you need anything more reliable than this, you should consider other alternatives such as having a mirrored service with another VPS provider on a completely different network.
Another HUGE selling point for me is that they host my databases, email, and spam filters (and DNS if you need to use it) on THIER servers! All three of these can potentially hog system resources (disk space, memmory, and bandwidth), so to not have them taxing my personal instance is HUGE. This also spares me from having to configure and maintain these services on my own. Oh, and to top it off, the perfomance of these services are VERY fast. However, the flexibiblity is there if you decide you want to host your own database and email on your own servers (obviously).
Support: As I mentioned earlier, I only really needed an unmanaged provider, and I haven't experienced any issues to this point, so my opinion on their support is limited. However, I have sent about 5-7 emails their way, and have received a response from them within 10 minutes every time! This gives me quite a bit of comfort should a real issue arise.
The funny part of all this is that I can not tell whether Iron Mountain's plans are managed or unmanaged. I guess they are at least semi-managed, as they have setup a database for me without any questions. I guess my question is whether they are fully managed (ie will do backups and install custom apps for you) or not... Maybe an Iron Mountain rep can chime in here.
Summary:
VERY smooth sailing so far. I have email, databases, subversion, apache, tomcat and more running, and still have plenty of resources left over for more. Sales/Support has been VERY friendly, and I have no concerns their. For what I need out of a VPS, who could ask for more. Their prices are also very reasonable, and their new Quartz plan is an amazing deal for a hobbyist like myself. Not really much to critisize at this point, but I will keep this forum up to date if anything arises.
We work with a hosting provider in Sweden. We rent a dedicated server there. The traffic was uneven, from time to time. Sometimes there was no traffic, and sometimes it reached 90Mbit. We bought a configuration with 10,000Gb. Now they sent us the traffic invoice. That has 18Tb if to count according to the 95% system. And in total we used about 12Tb. We didn’t know they counted according to 95%, that wasn’t mentioned in the agreement, it said only about 400sek per TB excluding VAT***. That’s why we thought that we had to pay exactly for the used traffic because if to count according to the 95% system, which means it’s counted per a channel, and the agreement informed only about the Tbs.
Lets say you're a customer looking for web hosting, but do have technical experience - you know, you develop your own websites, you've had experience in this sort of thing before.
What if you came across a provider who seemed to offer a good service, they're high quality, they can host your website on their brilliant setup etc... but they do not provide any e-mail accounts with your hosting?
We're developing our own shared hosting setup, our own control panel too. Regardless of the control panel though, we wouldn't feel comfortable hosting peoples e-mail. We have plenty of experience in every other aspect of general shared hosting - but not looking after e-mail accounts nor the associated software.
To be honest I don't think that many shared hosting providers truely handle e-mail properly, and that job should really be left to the professionals.
We could of course guide customers or potential customers on why we won't offer e-mail accounts (i.e. not wanting to offer something we know we can't provide to a high enough standard) and instruct them on how to setup e-mail with another provider (such as Google, who will do this for free with limitations).
The alternative to the above is that we mask in a third party to look after e-mail, i.e. resell someone elses e-mail services as part of our hosting packages. The third party would require API access to setup/remove accounts..
What do you think? Are we just acting stupid trying to provide web hosting without e-mail hosting included? I noticed a while back Dreamhost encouraged their customers to use an alternative e-mail provider!
I am looking for a new VPS and once again I think I am going to ditch Servint VPS in preference of a cheaper VPS. I know Servint is well known and they have a very good reputation here in this forum. I been longing to sign up with them but Lunarpages' offer looks very attractive.
What about Lunarpages?
It is $10 cheaper than Servint.
Lunarpages has 1000 GB bandwidth but Servint has only 500 GB.
Lunarpages has 20 GB storage but Servint has only 15 GB.
Lunarpages has 500 MB RAM but Servint has 384 MB guaranteed RAM (1 GB burst RAM).
Is the 500 MB RAM here http://www.lunarpages.com/virtual-private-server/ GUARANTEED RAM or SHARED BURSTABLE RAM?
The only thing that Lunarpages lost to Servint here is the number of IP address. Lunarpages comes with 1 IP but Servint offers 4 IP which is good for SEO.
I wanted to contact Lunarpages but couldn't find their contact email. Anyone know their contact email?
[mysqld] datadir=/var/lib/mysql socket=/var/lib/mysql/mysql.sock # Default to using old password format for compatibility with mysql 3.x # clients (those using the mysqlclient10 compatibility package). old_passwords=1
I have Plesk 11.5 (service provider mode) on a Windows 2008 server IIS7.Most of my sites are developed in .asp and therefore i use a custom 500-100.asp error page that check s the IP of the visitor then displays either a friendly error, or if its my IP a full error of what has happened (it also emails me the error). This allows me to debug pages easily whilst developing and to keep an eye on anyone trying SQL Injection hacks on my sites (as the error and email also have session variables and IP address).I dont have root access to the server as it is a Webfusion dedicated server.I have following the Plesk documentation -
1) Switch on custom errors for the subscription 2) Look in virtual directories and navigate to error documents 3) Find the error in question (500:100) and change it to point at either a file or URL
FILE - I had the data centre add in the 500-100.asp error page in to the virtual template so that my page is available in the list of virtual files - this didn't work but that maybe because its not a static page??
URL - when i add the path it says its incorrect, if i add a fully qualified address, it accepts it but it doesn't work.give me a specific example of the URL that can be entered relative to the root as the format in the documentation isn't accepted. The last step is to restart IIS which is also an issue as i cant seem to do this from the Plesk panel..It is as if it isn't catching the 500:100 error, and only catching the general 500 error??
I am currently running Google Analytics/Urchin 5 (v5.7.02), on a server, the server has started to act up, (on its last legs etc) and now I am trying to transfer the Urchin Software to a new server, where it would work effectively.
However upon installing the urchin software on the new server and running it (localhost:9999), I am presented with An Action Items Page, and these following choices
This is the scenario, domain.com are setup on server1, however server2 also has the same profile of domain.com as we use ns3 and ns4 using domain.com. This works fine with the nameserver setup on server2.
However I encounter problems as the emails from server2 won't reach server1 as there are duplicate profile on server2.
My question is how do I setup the DNS in cpanel/whm from server2 so the emails from server2 will reach server1?
I just want to use a server for file sharing, it will have nginx and that's it. I'm looking at centos, or freebsd, but I been using centos forever now and I'm not sure how to use freebsd, should I just stay with centos?
Do I tell my hosting provider to just install the OS and give me ssh action and that's it? Don't install any control panels or any other stuff? I want one domain and one subdomain on it though and ftp action.
I'm wondering whether it is possible to perform a full server migration to a new Plesk server with the same hostname or will Plesk give an error about the hostname being the same?
The new server would not be accessible by hostname (only via IP) until DNS and glue records were changed after the migration.