Dapper test results on HP hardware

[LaMont Jones]

One of my co-workers at HP recently ran Dapper Drake through the workload
test that we use for evaluating the stability of a new Linux release.
The report from the test is below.  (CHO == Continuous Hours of Operation,
under the stress-load.)

lamont

=======================================================================
Here are the tests I was able to complete with Ubuntu last week. I was only
able to file a couple of the bug reports so far, but I plan on reporting
the rest of them in the next day or two.  Overall the DL385 systems did
quite well, but the ia64 release needs some work before it can be
considered release quality:

System #1:
   Tested software: Dapper Drake RC (amd64)
   System Type: DL385
   Memory: 16GB
   Processors: Opteron 275
   Storage: Smart Array 64xx (1x300GB JBOD, 1x900GB RAID5)
   Tested:
      Successfully completed 80 CHO using 900GB drive

System #2:
   Tested software: Dapper Drake RC (x86)
   System Type: DL385
   Memory: 4GB
   Processors: Opteron 275
   Storage: Smart Array 64xx (1x300GB JBOD, 1x900GB RAID5)
   Tested:
      Successfully completed 80 CHO using 900GB drive

System #3:
   Tested software: Dapper Drake 20060525 (ia64)
   System Type: rx2620
   Memory: 16GB
   Processors: 2 x Itanium 2 1.6GHz
   Storage: 53c1030 PCI-X Fusion-MPT Dual Ultra320 (2x73GB, 1x146GB)
   Tested:
      Successfully completed 12 CHO 
   Bugs:
      (critical) Drive letters change between installer and installed system
      (critical) No getty serial console entry in inittab after serial based install
      (major)    Lots of ACPI debug messages when booting install CD and installed system
      (minor)    The graphics for elilo interface are off
      (minor)    The graphics in the partitioner are off
      (minor)    The efivars module isn't loaded

=======================================================================