Tony Posted October 16, 2011 Report Share Posted October 16, 2011 The sea005 server had crashed earlier today due to several sticks of memory failing on the system. Upon the server booting back up it was noted that the server was now reporting 16GB total memory opposed to the total 24GB. We've had four total sticks of memory fail which will need to be replaced. We will be replacing these on Sunday October 16th between 9am and 12pm PDT. This is the most off peak time for this server based on load average which is why we're doing it during this time. We estimate there will be about 30 minutes of downtime to replace the failed sticks of memory. If you have any questions regarding this maintenance notice please do not hesitate to contact our systems administration department. Date: 10/16/2011 Start time (PDT): 9:00am End time (PDT): 12:00pm Duration: 3 hours Estimated Down Time: 30 minutes Link to comment Share on other sites More sharing options...
Cody R. Posted October 16, 2011 Report Share Posted October 16, 2011 We're performing this replacement now due to instability issues. We'll update this ticket once it has been completed. Link to comment Share on other sites More sharing options...
Cody R. Posted October 16, 2011 Report Share Posted October 16, 2011 It appears the CPU wasn't seated correctly in the machine causing issues posting. The machine was booted successfully however half of the RAM (12GB) isn't showing so we're going to be bringing the machine down once more to swap out the RAM. Link to comment Share on other sites More sharing options...
Cody R. Posted October 16, 2011 Report Share Posted October 16, 2011 After replacing the RAM again the server is having trouble posting. We believe this is caused by bad CPU's so we're having these swapped out. Link to comment Share on other sites More sharing options...
Cody R. Posted October 16, 2011 Report Share Posted October 16, 2011 Everything is now online. We'll be monitoring the server closely for the next 24 hours to ensure there are no further issues. To wrap up the delay / what happened this morning: * The machine crashed unexpectedly earlier on the 15th. We were able to get online and schedule a memory replacement of 8GB worth of RAM * The server crashed unexpectedly early in the morning of the 16th * We swapped out all of the memory ahead of the maintenance window due to crashes * After swapping the memory the machine had issues posting until the processors were re-seated. Once booted the machine showed only 12GB of RAM * After replacing the RAM again the machine failed to post a second time. We had the processors swapped and the machine posted / went online without issue Sorry for any trouble this may have caused Link to comment Share on other sites More sharing options...
Recommended Posts