On Mon, 07 May 2007 22:41:05 +0200, Mallikarjun <[EMAIL PROTECTED]> wrote:
> A bigfile is created to fill the filesystem (cat /dev/zero > /bigfile) and > crash is seen when the file is deleted. I am seeing this issue with Solaris > 11/06 on a HP server. > > Just wanted to know if this is a known issue or if there is a patch/pointers > which can help in addressing the issue better. > > mbd snapshots of the crash: > > pseudo-device: devinfo0 > devinfo0 is /pseudo/[EMAIL PROTECTED] > kernel memory allocator: > bufctl corrupted > buffer=ffffffff988e2008 bufctl=0 cache: lufs_mapentry_cache > > panic[cpu1]/thread=fffffe8000af8c80: > kernel heap corruption detected > > > fffffe8000af8ae0 fffffffffb9876f8 () > fffffe8000af8b20 genunix:kmem_cache_free_debug+78 () > fffffe8000af8b60 genunix:kmem_cache_free+43 () > fffffe8000af8bd0 ufs:logmap_remove_roll+12f () > fffffe8000af8c60 ufs:trans_roll+21a () > fffffe8000af8c70 unix:thread_start+8 () > > syncing file systems... > 31 > 3 > done > dumping to /dev/dsk/c0t1d0s1, offset 110231552, content: all > > Issue is seen with all the UFS filesystems > > -Mallik > I gave this a first shot on a test box running latest patched s10 system on a 30GB file system, 2 CPU box and immediately deleting the bigfile after file system became full with not hitting the panic described. could you give more details on the system (cpu count) and file system size used and when did you issue the delete ? thanks --- frankB _______________________________________________ ufs-discuss mailing list [email protected]
