Ben, I tried running ls repeatedly with no problems other than the missing files. This worked for a few minutes until it hit the error again. I did not see anything on the console as the system rebooted upon hitting the error. The kern.log file shows this just before the system rebooted:
CIFS VFS: RFC1001 size 35 bigger than SMB for Mid=749 Sep 16 09:48:31 webdata kernel: [ 475.319384] Bad SMB: : dump of 48 bytes of data at 0xf6a10980 Sep 16 09:48:31 webdata kernel: [ 475.319394] 00000023 424d53ff 00000074 00018800 # . . . � S M B t . . . . . . . Sep 16 09:48:31 webdata kernel: [ 475.319402] 00000000 00000000 00000000 0cec0000 . . . . . . . . . . . . . . � . Sep 16 09:48:31 webdata kernel: [ 475.319410] 02ed0000 6a00ff00 02000000 00003800 . . � . . � . j . . . . . 8 . . Sep 16 09:48:54 webdata kernel: [ 501.027724] CIFS VFS: server not responding Sep 16 09:48:54 webdata kernel: [ 501.027724] CIFS VFS: No response to cmd 116 mid 749 I have found the following bug report which seems to describe the same problem. http://www.linux-archive.org/debian-kernel/216752-bug-509428-bug-509428-libsmbclient-system-freeze-when-accessing-samba-cifs-share-network-appliances-storage.html -Kevin Ben Hutchings wrote: > On Wed, 2009-09-16 at 07:05 +0200, Christian Perrier wrote: > >> Quoting Steve Langasek (vor...@debian.org): >> >>> reassign 546809 linux-2.6 >>> thanks >>> >>> On Tue, Sep 15, 2009 at 01:49:47PM -0700, Kevin Crain wrote: >>> >>> >>>> After mounting a remote directory, I 'cd' to a directory contained >>>> therein and do an 'ls'. This directory contains over 200 files but I >>>> can see only 131. If I do an 'ls some_file' on a specific file that is >>>> not listed it will list it. Browsing the remote directories using >>>> Konqueror shows all files. If I do an 'ls|less' on the directory with >>>> over 131 files the kernel panics and the system either locks up or >>>> reboots. >>>> >>> Then you have a kernel bug. Reassigning. >>> >> I agree this has nothing to do with samba, but isn't it weird that >> this is the first time we get such report? Is the kernel the stock >> lenny kernel on the client? >> > > This could conceivably be a bug in a specific network driver, unlikely > as it seems. > > Kevin: > > I need to see the full panic messages (a photo will be fine). Please > switch to a text console to run ls. You may need to add 'vga=6' to the > kernel command line to avoid messages scrolling off the screen. > > Which network driver are you using? Can you trace the network traffic > for the directory listing by running tcpdump or Wireshark on the server? > > Ben. > > -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
