[Coco] DriveWire 4.3.3o hanging NitrOS-9 on VCC 2.01...Nope

William Carlin whcarlinjr at gmail.com
Wed Sep 14 01:19:07 EDT 2016


NitrOS-9 Enthusiasts and CoCo Lovers

I have been experiencing what appeared to be at first random locks of
NitrOS-9 shortly after bootup when accessing DriveWire 4.3.3o

After boot, even accessing the 'dw' command by itself would cause my Telnet
session to lock.  Then I tried right at the console and it would lock.
Pressing F5 would reboot NitrOS-9 and it would hang right after displaying
the NitrOS-9 URL.  Using F9 to "power down' VCC and then boot NitrOS-9
again would do the same thing.  At first I wanted to blame VCC (sorry Bill)
and even when as far a stripping down the NitrOS-9 drivers to the bare
minimum and removing SuperIDE support (that I have noticed to be a little
flaky after heavy reads and writes) and the same thing would happen again
and again.  Then I realized that this is the same behavior that would
happen if I started VCC without first starting DriveWire.  Sure enough that
was the cause.  However, there is nothing wrong with DriveWire 4.3.3o.  It
has not been updated since 6/21/2013 and my configuration is simple and
does not change except for what disks I have mounted.  What HAS changed is
the version of Java VM.  A recent update to 1.8.0_101-b13 seems to have an
unexpected "feature".  I use a custom heap size to give DriveWire more
memory to work with in case I decide to create a new disk on the fly and
keep it in memory until I do a write to file or just kill it.  I looked at
my config in the Java Control Panel applet and sure enough my custom heap
size parameters are gone.  I added back my changes '-Xms128m -Xmx512m" and
no more lockups! This never happened when Sun was running the show.  Just
though I would share my findings in case someone else was ripping their
hair out over this.  Us older CoCo guys need to keep what little hair we
have left.

CoCo4Life

William Carlin


More information about the Coco mailing list