kix766 Posted January 20, 2014 Report Share Posted January 20, 2014 yes please keep it on ECHO ! so many issues if we are switched to another server again... Quote Link to comment Share on other sites More sharing options...
Michael D. Posted January 21, 2014 Author Report Share Posted January 21, 2014 yes please keep it on ECHO ! so many issues if we are switched to another server again...kix, We have two options:1. Try to bring accounts back online on another server, even if it's older data, to keep sites from being fully offline.2. Simply let everybody on the server be offline while we work on it. If the temporary server solution was an issue you could always have logged in and disabled your site on the temporary server [i.e. a maintenance page or something of the sort] where as if we didn't temporarily restore accounts to another server you would have no power to do anything but wait. I wish there was a better way to handle this that didn't affect our customers but at the end of the day we try to keep everybody online as much as possible. It's better to be online with slightly outdated data and possible slowness than to be offline entirely although we'd like to avoid both situations whenever possible. I do apologize for the trouble this may have caused you and I do appreciate your understanding in the matter. If you do have any corrupted data on your account [i.e. file that isn't working, content is garbled] open a ticket with technical support as soon as possible and we can restore the latest valid and non-corrupted copy of that data that we have. Quote Link to comment Share on other sites More sharing options...
Michael D. Posted January 27, 2014 Author Report Share Posted January 27, 2014 We identified the bad drive and removed it and are using the other SSD for read cache We enabled the read cache at 5 PM today ET [17:00] and it is apparent by our I/O graph below. It does take some time for the cache to initialize and populate which is the reason for the 2 hour delay on the I/O improvement.http://www.screen-shot.net/2014-01-26_19-30-26.png Anything under 10% is acceptable and anything under 5% is excellent. Currently the server has 240GB of SSD based read-cache. Once we have the replacement drive on-hand we'll be upping it to 480GB of read cache. We're going to formally be marking this issue as 'resolved.' Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.