Tony

The server has just been taken offline to replace the memory.

On Thursday August 23rd between 9:00am and 12:00pm PDT we will be replacing the memory on the Tornado server. This is due to the server repeatedly failing and we believe this problem to be caused by a bad stick of memory. Replacing the memory requires we turn the machine off and as a result we estimate the server will be down for between 30 and 45 minutes. If you have any questions regarding this maintenance notice please do not hesitate to contact our systems administration department. Date: 08/23/2011 Start time (PDT): 9:00am End time (PDT): 12:00pm Duration: 3 hours Estimated Down Time: 30 minutes

This is tricky because if we can write them as a blog post then they belong in our knowledgebase over our blog. I know for example a user recently had problems with building ruby gems after they had their username changed. It turned out the issue was with their .gemrc file in their home directory referencing the wrong path. This issue isn't well documented on the internet but if it's part of our KB when a user opens a ticket they'd see it opposed to finding some obscure blog post. They might be back soon we've been very busy and a lot has being going on blog posts about everything hasn't been a high priority issue as of late. We'll see if can get something up for the end of August though as a lot has changed in the past few months a lot of it not very customer facing. These changes however do make a huge difference in the experience of our customers and the reliability of our services.

We can assist with a dreamhost migration but understand it's not going to be near as smooth as moving from a host running cPanel. We'll try our best to get all the data in the right spots and set everything back up as far as databases. It will however require you to also test and make sure everything is working and understand we don't know your sites. That means it could take 24 hours for us to complete such a migration and possibly some time for you to test and if there are any issues point them out so we can attempt to correct them.

The server is now back online and serving traffic once again.

The file system check is done and the machine is currently booting back up.

File system check is 75% complete

File system check is 60% complete

File system check is 48% complete

File system check is 30% complete

File system check is 15% complete

Unfortunately it looks as though this will be longer than expected. The server is requiring a file system check on the /home partition which we are running now. We're estimating this will take approximately 45 minutes.

We will be performing an emergency kernel upgrade on the skyline server due to a kernel bug causing load averages to be reported incorrectly. As a result of load averages being reported as 2085233585 or higher various cPanel functionality is currently not functioning. We have consulted CloudLinux regarding this and it is a known kernel bug and will be fixed by updarting the kernel. We will be rebooting the server between 3:20am CDT and 3:50am CDT and we estimate the server will be down no more than 15 minutes. We are sorry about the lack of notice on this but unfortunately due to the nature of this issue it needs to be corrected as soon as possible. Date: 06/30/2012 Start time (CDT): 3:20am End time (CDT): 3:50am Duration: 30 minutes Estimated Down Time: 15 minutes

Based on what it's reporting it cannot find the domain as to why difficult to say does it provide more information? It's strange it cannot find the domain and we'd need more information maybe their trackers are getting blocked by us tough to say. Have you tried visiting your account when it claims it's down? Is it actually down during that time?

At this time the server is back online and has been for some time.

The file system check has been completed and the server is currently booting back up.

The file system check is 78% completed

The file system check is 57% completed

The file system check is 30% completed

It was investigated and the server was mistakenly restarted when disconnecting from the remote management system. We're sorry about any inconvenience the additional downtime may have caused.

The file system check is 6% completed

The Tornado server had crashed with a kernel panic relating to file system errors. The server has been rebooted and we've been forced to run a file system check on the /home partition. We estimate this will take about 1.5 hours to run.

The server is back online once again and we're awaiting word from datacenter technicians as to why it was brought offline after the memory upgrade has been completed.

The machine was online we're investigating now why it was taken by back offline after the upgrade was completed.

The server is now back online with 32GB ram and services are all starting back up.