I decided to write this today after our companies web server hosted by 1and1.com has been down for more than 72 hours and we have called them 7 times in the past 3 days only to get a should be up in few hours message every time.  We have been in business since 1994 and have never experienced such a long down time in our history.  Usually a few hours is acceptable per year not days.  Had we known the server would be down this long we would have pointed our domain names to another server and kept our customers happy.

-- Summary - Their uptime and bandwidth numbers are basically untrue.  They are probably pretty good for home use isp but for business we'd recommend yahoo.com or amazon aws

I'll be honest, I was once one of 1and1.com's biggest fan.  Having tried other services they were for one time the best provider from a price/value point of view.  We tried them for a free trial loved them so much that we moved most of our business to them.  Upgrading to a developers account, then adding a unix managed server and adding a 2003 windows server.

Although something happened around late January or early February this year 2007. Their service started a downward spiral.  Here is an excerpt of a letter we wrote to them that is still un answered.

We hope that you use caution when deciding to use their service and really hope that they return to the levels of support they provided prior to february this year.

============== 

Dear Sir,

We have been a very loyal customer for 3 years and have been promoting 1and1 any chance we had, we could not say enough about the great service, great company and great reliability at 1and1.com

We have 3 accounts a hosted developers account where we host our company website, a managed unix server and a windows 2003 server.   We have built up around 50 domain names over the past 12 year and manage 95% of them at 1and1.

In February our customers started complaining that their files were not getting downloaded, they were getting corrupted from our managed server.  Our 1and1 server ran for about 2.5 years flawlessly, without a problem and no complaints from our customers, but something happened in late January or early Feb.   We tried too, sure enough files being downloaded from our managed unix server just stopped in mid stream.  We checked the error logs and called 1and1 technical support.  When they tried from their network they could not reproduce the problem, we would expect your connection to the internet is faster than ours and our customers.  Finely after a few weeks of technical support back and forth 1and1 said they made a change to the global variables that we could not override, since we had a managed account, they told us that we needed to upgrade our server to a more expensive one with more memory.   Needless to say being a unix system developer in the early 80s, I started to lose faith in 1and1, unix and apache are very capable of handling the demands of our business without the global variable.  It was almost like 1and1 sabotaged us in an effort to make more money, at the risk of losing business. We were not sure we wanted to spend hundreds more and hundreds of hours moving our configuration to another server.  Also the explanation we got from 1and1 did not make much sense.  We were not sure an upgrade would even work.  According to the 1and1 support rep's explanation they said we were running out of ram, but things worked great for the 2.5 years prior to them setting a new global variable.  Upgrading to a more expensive system with more ram would eliminate the problem if one person tried to download a file but if two people tried to download a file at the same time we'd run out of memory even on a bigger system.  The caching variable needs to be set lower, eliminated or allow users to over ride it.  Let us upgrade servers when our demand drives it.   Companies like ours start small and we grow, let our growth drive upgrades.  We make more money we can spend more, changing variables or system configuration to force upgrades premature only hurts your reputation for solid quality and performance.

We have lost faith in 1and1 on our managed server but continued to experience good luck on our developers account.  We moved the critical files to our developers account and users continued to have reliable service until this past weekend when our developers account when down.  We called for support and the guy (you need better phone service to india, the connection is really bad), said they were working on it.   Now 72 hours later my website is still down.  We have lost thousands of dollars in business and have to fend calls from customers that they are working on it.  It's making our company with our 12 year history look bad like we don't know what we are doing.   My boss has asked me to look for other solutions as I completely understand his frustration.   I managed several large data centers including fermi labs server farm, there is nothing that could cause an outage beyond 4 hours if you have your plans in place.  Hardware can be swapped out and re-imaged in a matter of hours.  Domain addresses can be moved to servers that are up.  This outage should have lasted no more than 4 hours.

Can someone please tell me what is going on and when the system will be back on line, what is the problem so we can tell our customers, telling our customers over and over that it should be up in a few hours is making us look like idiots.  Every time we call support we get the they are working on it, it should be up in about 4 hours response, although its usually timed until the indian guy gets off of work and sticks the next guy with the problem, not very helpful for us.

We currently spend about $250/mth with 1and1 and would really like it if 1and1 could find a way to get back to jan 07 level of support and reliability.   If you intend to continue your down hill trend we will need to take our business elsewhere.   Please have your people look at the things that have gone wrong since Jan and develop a plan to get back to that level of reliability and please let us know so I can share that information with my management and customers.  Our business is growing and it would be really nice to grow with 1and1 and not have to move elsewhere.

If possible it would be nice to get some sort of credit for problems with our managed server and developers account once you get the problems resolved.  We would like to be able to override the cache variable you set on the managed server, not everyone who buys a managed server uses it for the same applications, we have lower amount of files but they are larger in size, your variable assumes that we have more files that are smaller in size.  Even removing the value and letting unix and apache process transactions would be better than not being able to use the server for the applications it was purchased for.

-----   

Randy