Re: New kernel, new problems...

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



I should also mention that even with server inode numbers disabled,
although this in theory could indicate a server problem, few client functions
are affected (cifs.ko autogenerates inode numbers for files when it detects
the server had problems) e.g.  hardlinks don't make much sense if we
can't tell if two files
are really the same so aren't supported in that case

On Sat, Mar 4, 2017 at 9:40 AM, Steve French <smfrench@xxxxxxxxx> wrote:
>> Autodisabling the use of server inode numbers on \\Athenae\C.
>> This server doesn't seem to support them properly.
>> Hardlinks will not be recognized on this mount. Consider
>> mounting with the "noserverino" option to silence this message.
>
> cifs.ko, as a linux file system, needs a unique number in order to identify a
> file (inode number).  It is particularly important when you see two files with
> different names but with the same inode number (which indicates that
> they are hardlinked to each other).   There are three cases where the cifs
> client can detect that the server gave out invalid inode numbers (or was
> unable to give out inode numbers). Note that inode numbers are called
> unique ids in the SMB2/SMB3 protocol).
> 1) the server returned 0 for the unique id on a file when listing
> directory contents
> or when querying info on a file
> 2) the server rejected the SMB2/SMB3 query that asks for the unique id
> (inode number)
> 3) we detect a collision where the server gave out the same unique id
> for two files of
> different types (e.g. a file and directory on the same mount with the
> same unique id)
>
> Not all file system types in Windows support unique ids (although NTFS
> does).  Could this
> be a FAT filesystem on the server?
>
> On Sat, Mar 4, 2017 at 12:31 AM, L A Walsh <cifs@xxxxxxxxx> wrote:
>> Brought up 4.10.1 w/options to help cifs info-dumping
>> (with CONFIG_CIFS_DEBUG, CONFIG_CIFS_DEBUG2, CONFIG_DYNAMIC_DEBUG enabled.)
>>
>> I probably should have tried the old options 1st, but added a
>> vers=2.1 to my mount line in /etc/fstab:
>>
>> //Athenae/C/    /athenae    cifs
>> user,noauto,rw,uid=law,gid=Administrators,nocase,serverino,vers=2.1,credentials=/home/law/.ssh/athenae,setuids,noauto
>> 0 0
>>
>> which resulted in a mount line:
>>
>> //Athenae/C/ on /athenae type cifs
>> (rw,nosuid,nodev,noexec,relatime,vers=2.1,sec=ntlmssp,cache=strict,username=law,\
>> domain=BLISS,uid=5013,forceuid,gid=544,forcegid,addr=192.168.3.12,file_mode=0755,\
>> dir_mode=0755,nocase,nounix,setuids,mapposix,rsize=1048576,wsize=1048576,\
>> echo_interval=60,actimeo=1,user)
>>
>>
>> Haven't seen the looping yet, but got some odd messages when I mounted in
>> my syslog:
>>
>> Mar  3 22:12:06 Ishtar kernel: [21527.100384] FS-Cache: Loaded
>> Mar  3 22:12:06 Ishtar kernel: [21527.120908] FS-Cache: Netfs 'cifs'
>> registered for caching
>> Mar  3 22:12:06 Ishtar kernel: [21527.121049] Key type cifs.idmap registered
>> Mar  3 22:12:06 Ishtar smbd[43429]: [2017/03/03 22:12:06,  0, class=rpc_srv]
>> Mar  3 22:12:06 Ishtar smbd[43429]:   pipe_schannel_auth_bind: Attempt to
>> bind using schannel without successful serverauth2
>> Mar  3 22:12:06 Ishtar smbd[43429]: [2017/03/03 22:12:06,  0, class=rpc_srv]
>> Mar  3 22:12:06 Ishtar smbd[43429]:   _netr_ServerAuthenticate3:
>> netlogon_creds_server_check failed. Rejecting auth request from client
>> ATHENAE machine account ATHENAE$
>> Mar  3 22:12:13 Ishtar kernel: [21533.978843] CIFS VFS: Autodisabling the
>> use of server inode numbers on \\Athenae\C. This server doesn't seem to
>> support them properly. Hardlinks will not be recognized on this mount.
>> Consider mounting with the "noserverino" option to silence this message.
>>
>> ----
>> Have never seen issues w/the hardlinks -- what does it mean that "this
>> server"
>> (windows 7SP1 doesn't support them?)
>>
>> Anyway, I wanted to try other options, but now I can't unmount what I just
>> mounted:
>>
>> # umount /athenae
>> This utility only unmounts cifs filesystems.
>>
>> ??
>>
>> It **used** to unmount things when I didn't specify vers=2.1...
>>
>> So how do I unmount this short of rebooting?
>>
>> Thanks,
>> -linda
>>
>>
>>
>>
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-cifs" in
>> the body of a message to majordomo@xxxxxxxxxxxxxxx
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
>
>
> --
> Thanks,
>
> Steve



-- 
Thanks,

Steve
--
To unsubscribe from this list: send the line "unsubscribe linux-cifs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html



[Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux