Hello, I have a mounted a IBM AS400 integrated file server (IFS) in etc/rc.d/rc.local using: mount /mnt/AS400/CVS and in fstab: \\as400\cvs /mnt/AS400/CVS smbfs user,password=password,uid=501,gid=501,username=as400,noauto 0 0 Every thing works ok until I try to edit/save a file on the server in kwrite for example. The network drive then hangs, and I cannot reconnect. The error in system log is : Sep 6 10:00:41 ISD-COMP1 kernel: smb_setup_bcc: Packet too large 3996 > 14 Sep 6 10:00:41 ISD-COMP1 mount.smbfs[1930]: [2004/09/06 10:00:41, 0] tdb/tdbutil.c:tdb_log(725) Sep 6 10:00:41 ISD-COMP1 mount.smbfs[1930]: tdb(/var/cache/samba/gencache.tdb): tdb_lock failed on list 30 ltype=0 (Bad file descriptor) Sep 6 10:00:41 ISD-COMP1 mount.smbfs[1930]: [2004/09/06 10:00:41, 0] tdb/tdbutil.c:tdb_log(725) Sep 6 10:00:41 ISD-COMP1 mount.smbfs[1930]: tdb(/var/cache/samba/gencache.tdb): tdb_lock failed on list 30 ltype=1 (Bad file descriptor) Sep 6 10:00:41 ISD-COMP1 mount.smbfs[1930]: [2004/09/06 10:00:41, 0] tdb/tdbutil.c:tdb_log(725) Sep 6 10:00:41 ISD-COMP1 mount.smbfs[1930]: tdb(/var/cache/samba/gencache.tdb): tdb_lock failed on list 30 ltype=1 (Bad file descriptor) Sep 6 10:24:48 ISD-COMP1 kernel: smb_lookup: find //date1.sh~ failed, error=-512 Sep 6 10:25:18 ISD-COMP1 kernel: smb_add_request: request [08c77e00, mid=400] timed out! Sep 6 10:25:48 ISD-COMP1 kernel: smb_add_request: request [08c77e00, mid=401] timed out! Sep 6 10:25:48 ISD-COMP1 kernel: smb_lookup: find //date1.sh failed, error=-5 Sep 6 10:26:18 ISD-COMP1 kernel: smb_add_request: request [08c77e00, mid=402] timed out! Sep 6 10:26:18 ISD-COMP1 kernel: smb_lookup: find //date1.sh failed, error=-5 Sep 6 10:26:48 ISD-COMP1 kernel: smb_add_request: request [08c77e00, mid=403] timed out! Sep 6 10:26:48 ISD-COMP1 kernel: smb_lookup: find //date1.sh failed, error=-5 Deleting/coping files is no problem. Things go wrong using CVS, Ant, OpenOffice etc. The problem started when I upgraded from 2.4 kernel. I have tried a Common Internet Filesystem (cifs), but this seems not to be supported by the server. Any ideas? Cheers Greg
I have noticed the same problem since I upgraded from the stock kernel on SuSE 9.2 (linux-2.6.8-24) to the SuSE updated kernel 2.6.8-24.10. On the stock kernel Samba worked fine but after the upgrade I got the same errors in the message log: Jan 5 15:13:02 starmax kernel: smb_trans2: invalid data, disp=0, cnt=0, tot=0, ofs=0 Jan 5 15:13:32 starmax kernel: smb_add_request: request [d6f7ee60, mid=3859] timed out! and the same behavior. From the command line I could copy, create and delete files without problems. Once I had to save files from a program (such as kwrite) onto a smb share the save would timeout and the file would not be saved and its length would be set to zero. I also noticed that this happens only on Linux to Linux connections. I have a windows share mounted on my Linux machine at home and have no problems like these (with the updated kernel), so it seems that the problem might be on the server side of Samba.
I just used SuSE's online update tool to update to 2.6.8-24.11 and I do not have this problem any more. It would be nice to know if this is a kernel or samba issue though.
- patches.fixes/smbfs-security-fix: merged 2.6.10-ac8 smbfs fixes and old stefan esser fixes (46204). From the kernel-default chnagelog. rpm -qp --changelog kernel-default-2.6.8-24.11.i586.rpm | head So it was a kernel bug.
I am running 2.6.10-1.741_FC3 and it still does not allow saving etc. This should be open still?
No sure as this looks like a Fedora and not Samba.org bug. Might be worth to inform the Fedora guys.
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=129124