From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.2) Gecko/20040809 Epiphany/1.3.8 Description of problem: Plugging a massstorage device in, pulling it out and then plugging it in again causes hal to stop working. Version-Release number of selected component (if applicable): hal-0.2.97.cvs20040901-1 How reproducible: Always Steps to Reproduce: 1. Plug in a USB mass storage device 2. take it out 3. plug it back in again Actual Results: Device works the first time then hal hangs and the device is not mounted the second time. Expected Results: Device is mounted Additional info:
Created attachment 103841 [details] Hal daemon output
Created attachment 103842 [details] /var/log/messages output
Created attachment 103843 [details] lshsl output
Hi, is it possible to get an strace from hald from just before the process stops responding? Also, when hald hangs is it possible to kill the process? Even with -9?
Can't get an strace or kill it. Trying to attach to the process with gdb hangs gdb and I have to kill it with -9.
Please try to get the strace with the device in question (Ed's Cruzer IIRC).
Reassigning to kernel. It seems the 256mb version of the cruzer mini causes a kernel oops. I also linked to an strace of hal while plugging in and out the USB drive. It is over 2 megs so I decided to just post it on my peoples page. It looks like hal blocks on a read of the device. http://people.redhat.com/johnp/hal.strace
Interesting. hald locks up on a read on /proc/bus/usb/devices. We shouldn't really be reading that file anymore now that 2.6.6 and later contains my patch for the sys files we need. I suppose the oops causes this behaviour?
Similar symptoms reported in bug #132572 but without the kernel oops.
I've been using a 256MB cruzer on a daily basis for months without problems, so I guess this is safe to close. Please reopen if its still causing problems for you.