This is the mail archive of the mailing list for the Cygwin project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

1.5.5-1 trouble with mounting nfs shares on test RH9 client


Main test client is an RH9 box, almost fully patched (trying to keep the test box the same until I get this working.)
test client kernel: 2.4.20-24.9
nfs: nfs-utils-1.0.1-3.9

Main server is w2k sp4 + current as of last week, running Cygwin 1.5.5-1 (installed via setup.exe as normal).

cygcheck -s -v -r output is attached.



Share a read-write mount point out to a number of clients via nfs.  No authentication, and all_squash-ing to the guest user on NFS.


A subset of /etc/exports configurations cause the RH9 test client to fail with the error:

[root@netfiz108 tmp]# mount -t nfs netfix08:/testshare /mnt/temp
mount: wrong fs type, bad option, bad superblock on netfix08:/testshare,
       or too many mounted file systems


I got cygwin installed ok, and ran the nfs-server-config script.  The only unusual event was that the nfs-server-config script complained that I had installed in "user" mode (I installed in "system" mode), and printed the "mount (1) command did not return SYSTEM mounts" error.  The example given in the error message appeared to match my existing <mount -m> exactly (with all mounts listed as "system" except cygdrives), but I ran the commands anyway and the error did not reoccur on running the script again.

Currently, the nfs server functions properly with this /etc/exports configuration:

/testshare ip_address(rw)

This one also works:

/testshare ip_address(ro)

It appears that any other configuration leads to the error message on the client.  No errors are logged on the server side in the Event Viewer during this event.  For example:

/testshare ipaddress(rw,all_squash)

and also:

/testshare (rw)

results in the client immediately rejecting the connection with the "wrong fs type, bad option, or bad superblock" error.  Also, attempts to export another mounted filesystem fail with the same error, even if the configuration is the "safe" one:

mount -f -s -b "F:\share" "/realshare"

/realshare ip_address(rw)

The problem is reproduceable, and I can switch back and forth at will by changing the configuration and restarting the services in the mmc.  Since no one else has seen this, I'm convinced that I'm missing something basic, but am unable to determine what it is.  Any thoughts would be appreciated.  Thanks.


Attachment: cygcheck.out
Description: Binary data

Unsubscribe info:
Problem reports:

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]