Status as of Saturday, June 24, 2017

[Steve Sokolowski]

Here's a few comments on today:

• The number of outstanding customer support requests is now above 30. As a result, development has been placed on hold and all resources have been devoted to resolving the requests. Chris will resolve the issues that require the least work first, so that he can make the biggest difference to the most customers. Right now, it's difficult to gain an understanding of what, if anything, is actually wrong because there are so many outstanding requests. I think that many of the outstanding issues have been resolved without even having read them first, but we need to confirm that.
• After the backlog has been resolved, our next task will be to set up a ticketing system. The major problem that is currently hindering progress is duplicate issues, where E-Mails come in that have already been submitted to the forums and in private messages. Chris will set up a ticketing site where tickets can be entered by customers. E-Mails will continue to be accepted and will cause the automatic insertion of tickets. Tickets will be able to be merged so that when five customers report the same issue, Chris can track them as one issue and make one response.
• Once the ticketing system has been brought online, forum private messages will be eliminated as a method of support, so that everything is in one place. The "support" forum will continue to be a location where customers can talk about issues and Chris can reply to general problems that affect lots of people. Our goal is to get outstanding support requests down to zero as soon as possible.
• The only time support requests are not answered immediately is on Mondays to Fridays, during "normal business hours." Once the ticketing system is available, issues Chris is working on will be visible by me so I might be able to help. We'll also investigate hiring someone who can work during these hours to resolve tickets.
• Chris purchased two additional solid state disks to further reduce the share insertion delays. The current delays continue to plague the site for a few reasons: first, the huge number of support requests has blocked our being able to release the changes that will reduce the data written to disk. Second, it will take four weeks before all the old share data is archived on the normal schedule. Finally, the existing share data is on a RAID 5 with slower disks because of the unnecessary data. Eliminating the unnecessary data will result in 1/5 the data being written; the random share exclusion resulted in 1/8 the data being written, and the new disks will result in 10x the performance, meaning that we expect the share insertion capacity to be 400x larger by August.
• The disk installation will occur the evening of July 6, and will take 15m. Four weeks later, after all the old data has fallen into the archive, we will move the shares data to this disk, resulting in 3 hours of downtime in early August.
• A post summarizing the attacks and their resolution, which was written earlier in the week, will be made at some point later this weekend to close out that chapter in the site's history.

[GregoryGHarding]

sounds great guys, will you be using a system like zendesk?