Using a Cisco Blade with 40 cores and 256 gb memory in a Cisco stack and using boot from SAN, I can't get a memory dump on BSOD RRS feed

  • Question

  • I've got a blade server with lots of memory and cores that I can cause to bluescreen at will (give me an hour). Normal work loads it is fine but when I crank IOMeter up with 8 workers with 25% random writes and 75% random reads 32k blocks on 32k boundary with a queue depth of 32, I get a bsod within an hour.

    But I can't capture the core dump.  I've set the CrashControl registry settings to auto reboot, DumpFile to both the c:\ drive (boot from SAN drive), a local hard drive, ... And still I've searched the entire host for the memory dump without success.

    Friday, April 25, 2014 10:28 PM

All replies