Hi Peter,
On Thu, Nov 24, 2005 at 10:06:12PM +0100, Peter Welter wrote:
I use Hobbit 4.1.2p1 on SLES9.1 and unfortunately two strange things happened.
-1- the hobbitd_client crashed: #11 0x0804af6d in unix_disk_report (hostname=0x20333220 <Address 0x20333220 out of bounds>, hinfo=0x323a3232, fromline=0x36353a34 <Address 0x36353a34 out of bounds>, timestr=0x54454320 <Address 0x54454320 out of bounds>, capahdr=0x30303220 <Address 0x30303220 out of bounds>, mnthdr=0x202d2035 <Address 0x202d2035 out of bounds>, dfstr=0x656c6946 <Address 0x656c6946 out of bounds>) at hobbitd_client.c:385 Previous frame inner to this frame (corrupt stack?)
There's clearly some buffer overrun going on here, and it seems to be triggered by the handling of a disk status.
-2- Something strange happened: in the rrd-directory the following 'disks' appeared?! Only root is a valid disk. I checked the clientoutput but all seems fine.
linux:/var/lib/hobbit/rrd/linux # ls -ltr *disk* -rw-r--r-- 1 hobbit hobbit 38532 Nov 20 10:13 disk,status.rrd -rw-r--r-- 1 hobbit hobbit 38532 Nov 20 12:48 disk.rrd -rw-r--r-- 1 hobbit hobbit 38532 Nov 20 13:53 disk,etc,X11,xinit,xinitrc.rrd -rw-r--r-- 1 hobbit hobbit 38532 Nov 20 14:03 disk,nt.rrd -rw-r--r-- 1 hobbit hobbit 38532 Nov 20 15:03 disk,,alert.chk.rrd -rw-r--r-- 1 hobbit hobbit 38532 Nov 20 16:24 disk,-,+.rrd -rw-r--r-- 1 hobbit hobbit 38532 Nov 24 21:23 disk,root.rrd
However, I dont know if these two things are connected.
They might be - at least, both of them relate to a disk status and that feels like more than a coincidence.
Could you mail me a copy of the ~hobbit/client/tmp/msg.txt file from the host called "linux" on your network ? I'd like to see if the df output is any different.
Regards, Henrik