A few months ago, the organization I work for decided to investigate new hosting providers: I spent several hours reviewing our possibilities, and we decided to go for the company ProfitBricks.
For a number of reasons, we prefer our data to be on servers in Germany, and ProfitBricks seemed to offer this at a good price. The company writes on its website that 24×7 support is included (unlike other cloud-based hosts): “When you have a technical question, who better to ask than a senior system administrator who understands your issues and can offer clear solutions directly and without delay?“.
I had some talks with ProfitBricks’ friendly sales team, discussed the architecture with them etc., and finally, my company decided to move our servers. I recall at our final IT meeting before signing up that I found it odd that I couldn’t find any neutral or negative reviews online. It just seemed a little too perfect.
I set up two servers just for testing, and my colleague set up a third server, which was the first we wanted to go into production.
After the server was set up, everything worked as it should for a little less than a week: On Sunday morning, I received an alarm that the server was having file system issues.
I logged on to investigate the problem, restarted the server to run fsck, which resulted in the server not coming back up.
I then called ProfitBricks, and after a brief conversation, I received the first email.
Sent: February 9th, 2014 09:27
Thank you for your Phone call and bringing your server issue to our attention. I have informed Sysops of the issue of your Server being in RO FS mode and we are starting an investigation into the issue. Thank you for your patience in this matter.
4½ hours later, I received a message with a proposal on some things I could do (booting up using a DVD image). I tried that, but it didn’t work. Called them again later in the afternoon to get a status. They promised to follow-up by email. That didn’t happen, and at night, I called again for a status. They told me that the ticket was pending their staff in Germany. It had been down for about 13 hours at that time, and I didn’t feel that the “core team” really cared about the system. I had called them 4 or 5 times during that Sunday to ask about the ticket, and kept receiving promises that they would get back to me, and that they were already working on the ticket.
I asked my German colleague to set up the server at a new host, but left the incident ticket open.
Then, 28½ hours after I opened the ticket, I received the first question that indicated that the server was actively being looked at:
Sent: February 10th, 2014 12:57
What file syustem is the OS using? ext3 ? ext4 ?
A few minutes later:
Sent: February 10th, 2014 13:00
Did you delete [server-name] server ?
And a few hours later:
Sent: February 10th, 201415:57
What Guest OS he used
To paste (if possible) relevant line (s) in the guest OS kernel log
Link and attach pictures to IO latency statistics for that volume 10 min prior and after going read-only
I had not deleted the server at the time he wrote, and these emails made me so mad that it was impossible for me to write back the same day. Instead, the following day, I asked for a full refund, which they never responded to.
Although this is generalizing based on a single case, I will still not recommend ProfitBricks to anyone. The offer of having system administrators on call 24×7 at a reasonable price was simply too good to be true.
ProfitBricks’ marketing and sales teams seem to be doing a lot better than the operations team.
Based on this experience, I have sent ProfitBricks’ billing department two separate messages asking to get a refund – they have not gotten back to me, which is not a big surprise to me.