29 January 2014 – Wednesday


  • Three former students in last night’s update.  Dancing mode, engaged.
  • The update died, a new student with two consecutive hyphens in their last name shone a bright light on a bug in the sn attribute processing when a null value was attempted to be added.  The code was updated to not have this happen again, and the update was completed.


  • NW153818 has not claimed anymore tapes overnight.  EMC says they should have the hot fix ported to in a day or two.
  • No response on the SR60727670 (lgtoclnt- does not cause compat-libstdc++-33 to be installed when compat-libstdc++-296 is already installed) from EMC yet.
  • Tape clones were run and completed.
  • Even though the COM/IS group is still writing away (at a pitiful 15MB/s) I have started the COM/IS clone process – it will never complete if I don’t get it started.
  • Disk clones started too.
  • Reading the Red Hat documentation on NFS performance.  Which boils down to “don’t touch it until you have problems – then make huge guesses”.  UGH.
  • Ran test for EMC – may have duplicated the issue (NW153818) from yesterday – not completely sure I think it was a fair test.  Need to reload test server with 7.6 and re-run the test to see if it worked there.
  • Loaded on the test server and verified that the test is valid and does work.
  • EMC provided a patched nsrd at the end of the day, I downloaded it and installed it on the test server (after putting on and verifying that it still had the problem).  The patched nsrd does fix the problem.  I have updated the SR with that fact and have asked if they (EMC) want to do any cleanup before I put it into production (the binary is significantly larger than the original one was).


Leave a Reply

You must be logged in to post a comment.