Hi, I am working with an IBM THinkpad R40 2722 using Debian unstable (Linux 2.6.8, compiled from Debians kernel-source package). JFS is compiled built-in and is the fs of the root partition. While on another machine jfs workes great (after xfs eat my data various times), there are some problems on my laptop
1) Access to the filesystem results in a fairly high cpu load. The usual updatedb cron-job, when find walks across the disk, gives a load of no less than 4 with the system getting really slow. Transfers via ftp oder scp across eth0 nearly stall when data is written onto the disk. xosview shows a 100% cpu load (sys) while gkrellm suggests, the processor is mostly idle. top shows about "85% wa". I am rather unsure whether this is enough to find the cause, so please let me know, which further information you need and how I can provide it. 2) I am using laptop mode which worked well with ext2. Some websites say jfs is supported while others do not. However, standalone LM does nothing while being used with the smartspindown script the disk actually spins down. However, few accesses disturb the procedure and I would like to find out, which file has been read/written to. dmesg gives e.g. "write block 7688"; so how do I find out, which file is associated with this block? 3) Is there an easy way to find the file which causes this log? non-latin1 character 0x393 found in JFS file name mount with iocharset=utf8 to access Or can I just ignore this? Thanks in advance! Best wishes Torsten _______________________________________________ Jfs-discussion mailing list [EMAIL PROTECTED] http://www-124.ibm.com/developerworks/oss/mailman/listinfo/jfs-discussion
