bwbug: RE: Linux Cluster Problems

[Frank Summers]

Your problem sounds reminiscent of the ext3 bug in the
stock Red Hat 7.3 kernel (2.4.18-3). I believe that this
bug was for the SMP kernel only, though.

Red Hat provided an update less than a week after the
release of 7.3 to kernel 2.4.18-4. Make sure you are
using the newer kernel.

I have a 17 node cluster (master + 16 slaves) running
RH7.3 here at Space Telescope. I have seen some minor
performance delays due to NFS traffic, but nothing major.
Of course, I'm probably using my cluster in distinctly
different ways than you. My jobs are CPU hogs, but not
bandwidth intensive.

Frank Summers


Space Telescope Science Institute           410-338-4749
3700 San Martin Drive                            410-338-4767 (FAX)
Baltimore, MD  21218                             summers at stsci.edu
             http://terpsichore.stsci.edu/~summers/