WATCH LIVE, OHH NO Cisco UCS Blade Failed DIMM Modules

[@liveitloveitlife]

Wouldnt it be easier to use the UCS console to find out.

[@sysadmintutorials]

Hi Darren, on the KVM the server was only showing memory test at which point it wouldn't progress any further. Within UCSM health it also didn't show any errors on the RAM. Only till I found the memory module causing the issue, did the server boot. Unfortunately this time that was by process of elimination. This happened to me twice before as well. Usually the DIMMs show an error in UCSM health though which is much easier to troubleshoot :)

[@eugenealejandro8627]

@@sysadmintutorials hey David! Nice content as always, especially for those who don't have any clue what UCS looks like in real life. Did you replace just that faulty mem module? ofcourse with the same identical specs. Really appreciate your quality content! Hope to see more in the future!

[@sysadmintutorials]

Hi Eugene, thank you for the great feedback. Yeah that's why I thought it would be good to add some hardware content to the channel as well. Exactly, I had to log a Cisco support case and then they sent out RMA replacement DIMMs. After replacing them the server booted. Funny thing is, next day a SSD disk failed in the same server, however it was in RAID 1 so server remained up. We put it in maintenance mode as it had ESXi on it under vCenter management, and then repeated the same process by logging another Cisco support ticket and getting an RMA replacement part :)

[@lostphotographs3936]

The Log output is also a good place to look once you figure out where to go to pull up the internal hardware listings. It will show exactly what dimm is in each slot, along with a slew of other details. Who made it, serials, speed, etc.

[@jrzvi786]

Login in to cisco management console you will get there health status of all memory.

[@sysadmintutorials]

Hi Javed, unfortunately there were no errors in UCSM health. On the console of the server upon boot, the server was performing memory tests at which point it wouldn't boot any further. Had to find the faulty ram manually by removing piece by piece and seeing if the server booted again. I had to do this 2 time before on different blades as well.