I noticed our PowerEdge R710 server running CentOS 5.4 had a glowing orange monitor, so I checked the logs in the DRAC:
Mem ECC Warning: Memory sensor, transition to critical from less severe ( DIMM_B9 ) was asserted
Mem ECC Warning: Memory sensor, transition to non-critical from OK ( DIMM_B3 ) was asserted Mem ECC Warning: Memory sensor, transition to non-critical from OK ( DIMM_B9 ) was asserted
Oh wow, two bad DIMMs. Before I call up Dell support, I know from experience they're going to expect a dset report first. I've done so successfully on Windows servers before, but this is my first Linux server with a complaint.
I uploaded delldset_v2.2.125_x64_A01.bin to my home directory and ran
sudo ./delldset_v2.2.125_x64_A01.bin --extract. It responded
delldset_v2.2.125_x64_A01.bin: Command not found. I dropped the extract flag and even renamed it to just delldset.bin, but the error won't go away. Unless I'm overlooking some painfully obvious syntax goof, I now believe I'm not using Linux dset properly, despite online instructions suggesting I am. What am I missing?
I observed this question but I'm wary of bringing down an important DB to make a huge BIOS version jump whose compatibility with CentOS 5.4 is uncertain for a potentially non-applicable problem.
UPDATE: mdpc nailed it. I'm facing a new error message, encountered during report generation: I receive a stack trace and the message 'NOTE: Not found one or more packages required for chassis and storage data collection.'