No subject

hang i mean, no data can get transfered from or to the server and every
programm accessing the a cifs share (df,cp,ls) freezes. The shares are
from two different servers, from two different networks, so i guess its
not the server part that fails. Other machines are also able to use the
shares.

Its only the cifs mounts that fail, iscsi continue to work and ftp
connections do not get droped, so gues ist not the network ether.

The mounts can not be unmounted. If i do a unmount, even with force, the
unmount process hangs like any other and can not be killed, even with
-9.
A reboot fixes the problem, but only a reboot. I can not get rid of the
hanging mounts.

Any idea on how to debug the problem?


I have some entrys in dmesg

 CIFS VFS: Send error in FindClose =3D -9
 CIFS VFS: Send error in FindClose =3D -9
 CIFS VFS: Send error in FindClose =3D -9
 CIFS VFS: Send error in FindClose =3D -9
 CIFS VFS: Send error in FindClose =3D -9
 CIFS VFS: Send error in FindClose =3D -9
 CIFS VFS: Send error in FindClose =3D -9
 CIFS VFS: Send error in FindClose =3D -9
 CIFS VFS: Send error in FindClose =3D -9
 CIFS VFS: Send error in FindClose =3D -9
 CIFS VFS: Send error in FindClose =3D -9
 CIFS VFS: Send error in FindClose =3D -9
 CIFS VFS: Send error in FindClose =3D -9
 CIFS VFS: Send error in FindClose =3D -9
 CIFS VFS: server not responding
 CIFS VFS: No response for cmd 50 mid 19472
 CIFS VFS: server not responding
 CIFS VFS: No response for cmd 50 mid 30264
 CIFS VFS: No response to cmd 46 mid 34122
 CIFS VFS: Send error in read =3D -11
 CIFS VFS: No response for cmd 50 mid 34123
Status code returned 0xc0000008 NT_STATUS_INVALID_HANDLE
 CIFS VFS: Send error in read =3D -9

Which apear all the time. I tried to google the error msg but no luck so
far. Some have the same error msg, but under complete different
circumstances. Any idea on that codes?

All i get from  cat /proc/fs/cifs/DebugData is


---------------------------------------------------
CIFS Version 1.50cRH
Active VFS Requests: 0
Servers:
1) Name: 172.17.0.85  Domain: COLOR Mounts: 1 OS: Windows Server 2003
3790 Service Pack 2
        NOS: Windows Server 2003 5.2    Capability: 0x1f3fd
        SMB session status: 1   TCP status: 1
        Local Users To Server: 1 SecMode: 0x3 Req On Wire: 0
MIDs:

2) Name: 10.9.170.20  Domain: DMZ Mounts: 2 OS: Windows Server 2003 R2
3790 Service Pack 2
        NOS: Windows Server 2003 R2 5.2 Capability: 0x1f3fd
        SMB session status: 1   TCP status: 1
        Local Users To Server: 1 SecMode: 0x3 Req On Wire: 0
MIDs:

Shares:
1) \\172.17.0.85\img01 Uses: 1 Type: NTFS DevInfo: 0x20 Attributes:
0x700ff
PathComponentMax: 255 Status: 1 type: DISK
2) \\10.9.170.20\orders_09 Uses: 1 Type: NTFS DevInfo: 0x20 Attributes:
0x700ff
PathComponentMax: 255 Status: 1 type: DISK
3) \\10.9.170.20\BackupMo_09 Uses: 1 Type: NTFS DevInfo: 0x20
Attributes: 0x700ff
PathComponentMax: 255 Status: 1 type: DISK

It is a centos 5.2 system (kernel 2.6.18-92.el5 #1 SMP). It is not a
samba server, even samba is installed (samba-3.0.33-3.15), it is not
started.=20
It is an ftp server and writes files to a windows 2k3 server share,
where it gets collected from varios other servers.

When i check with wireshark/tcpdump there is no communication on the
wire when the cifs module hangs. As the connection has heavy load from
time to time, i can not sniff for weeks in advance. So there is no
tcpdump i could check.

There are no entrys in the server logs. The server thinks the client is
still connected.
I can not mount shres when the cifs module hangs, even i can ping the
server.

I tried to echo 3 > /proc/fs/cifs/cifsFYI but i get only more dmesg
entrys with no information for me. So i hope someone here will be able
to explain what my server does.




Regards,
eric