Sat Sep 29 18:41:45 PDT 2007

New Release (3.1) of hmcclinic LaTeX Class

While helping one of the teams with their Statement of Work, I discovered a minor formatting issue with the previous release of the hmcclinic class. We updated the class to correct some issues with formatting Statements of Work (which are based on the article class) by making some command definitions conditional on the type of document. One of the commands I moved was the redefinition of \cleardoublepage; as a result, LaTeX puts headers on what are meant to be blank pages in Statements of Work. I have now corrected that problem in the class file, and made a new release, 3.1. of the hmcclinic class.

Version 3.1 is now the default version on the math cluster machines; if you're typesetting your documents on one of our Linux machines, you do not have to do anything special to pick up the changes, just retypeset your document.

If you are typesetting on your own machine, you will need to get the latest version of the class file and supporting files from http://www.math.hmc.edu/computing/support/tex/classes/hmcclinic/.


Posted by Claire Connelly | Permalink

Mon Sep 10 16:12:25 PDT 2007

Mirrors Back

The second run of fsck found no additional errors, so I'm leaning toward believing that the system is working again.

There may be some speed issues for a bit while the mirrors sync back up against upstream sources, but I hope they will resolve themselves within a day or so.


Posted by Claire Connelly | Permalink

Mon Sep 10 15:46:22 PDT 2007

Mirrors Temporarily Offline

The mirrors we run at yum.math.hmc.edu are temporarily offline while I verify the drives. I expect them to be available again by the end of the day (2007 September 10).

I found some SCSI errors starting at noon on Saturday, 2007 September 8. The array utility claimed that the controller, drives, and array were all fine, but I noticed some performance issues that could be related to disk corruption and took the array offline to verify the file system. I have fixed problems found in the first run, and am doing another ``paranoia'' run for good measure.

Sorry for any inconvenience; the problems are also preventing me from moving forward with a number of my own projects, so fixing things is a very high priority.


Posted by Claire Connelly | Permalink