2006-02-15 (水) の 09:32 -0800 に Peter J. Stieber さんは書きました: > PS = Pete Stieber > PS>> I recently posted mentioning that I was getting out of memory > PS>> (oom) errors with FC4 kernels 1656 and 1831 on both > PS>> x86_64 and i686 version of the kernel. > PS>> > PS>> > http://www.redhat.com/archives/fedora-list/2006-February/msg01112.html > PS>> > PS>> I have simple Bash shell scripts (first line has #!/bin/bash) > PS>> that seem to start up many unicode_start processes, but > PS>> it doesn't occur every time I run the scripts. I haven't been > PS>> able to figure out what triggers this behavior. The script that > PS>> I believe caused the problem in my prior post is one that > PS>> cleans up a source directory hierarchy. > PS>> > PS>> #!/bin/bash > PS>> # Checking to see if your home directory exists. > PS>> TOP_DIR=$HOME > PS>> if test ! -d $TOP_DIR; then > PS>> echo ERROR: Directory $TOP_DIR does not exist! > PS>> exit > PS>> fi > PS>> # Checking to see if a source directory exists. > PS>> SOURCE_DIR=$TOP_DIR/Source > PS>> if test ! -d $SOURCE_DIR; then > PS>> echo ERROR: Directory $SOURCE_DIR does not exist! > PS>> exit > PS>> fi > PS>> cd $SOURCE_DIR > PS>> find . -name aclocal.m4 -exec rm {} \; > PS>> find . -name configure -exec rm {} \; > PS>> find . -name Makefile.in -exec rm {} \; > PS>> find . -depth -type d -name autom4te-2.53.cache -exec rm -rf {} \; > PS>> find . -depth -type d -name autom4te.cache -exec rm -rf {} \; > PS>> > PS>> I have run this script many times since and haven't > PS>> seen the problem, but every once in a while my > PS>> Bash scipts start a ton of unicode_start processes. > PS>> I use the system provided .bashrc. > PS>> > PS>> Does anyone have a suggestion as to what might > PS>> cause all of the unicode start processes? > > JR = Joel Rees > JR> Got any non-US-ASCII characters in path names? > JR> Like file names with umlauts and such in them? > > No. If I did wouldn't the problem occur every time I ran the script? Depends on what's in the directory you run it on, I would think. Sometimes, you can't tell be looking whether a character (encoding point, really) is one of the 7-bit US-ASCII set. But I'm just guessing. > I have seen this behavior with other bash shell scripts I have written > as well, but the problem is intermittent. When I Google "unicode_start" > I can't find anyone complaining about a similar problem. It makes me > think I have a configuration problem, but I'm not sure what it is. > > $ set | grep LANG > LANG=en_US.UTF-8 UTF-8 is definitely intended to mean Unicode UTF-8. > Has anyone else ever experienced this? > > Pete > >
