Mapping N5200 as network drive - not in my network places

[jbl]

It appears that most of the files in raid/data/tmp/var are messed up:

root@127.0.0.1:/raid/data/tmp/var# ls
ls: ./messages.tdb: Input/output error
ls: ./connections.tdb: Input/output error
ls: ./brlock.tdb: Input/output error
ls: ./locking.tdb: Input/output error
ls: ./gencache.tdb: Input/output error
ls: ./registry.tdb: Input/output error
ls: ./group_mapping.tdb: Input/output error
ls: ./account_policy.tdb: Input/output error
ls: ./ntdrivers.tdb: Input/output error
ls: ./ntprinters.tdb: Input/output error
ls: ./ntforms.tdb: Input/output error
ls: ./unexpected.tdb: Input/output error
browse.dat     sessionid.tdb

I'm not sure what this means. Maybe the files are there but cannot be read? I managed to delete messages.tdb and restarted smbd manually so now I get:

root@127.0.0.1:/raid/data/tmp/var# ls
ls: ./connections.tdb: Input/output error
ls: ./brlock.tdb: Input/output error
ls: ./locking.tdb: Input/output error
ls: ./gencache.tdb: Input/output error
ls: ./registry.tdb: Input/output error
ls: ./group_mapping.tdb: Input/output error
ls: ./account_policy.tdb: Input/output error
ls: ./ntdrivers.tdb: Input/output error
ls: ./ntprinters.tdb: Input/output error
ls: ./ntforms.tdb: Input/output error
ls: ./unexpected.tdb: Input/output error
browse.dat     messages.tdb   sessionid.tdb
root@127.0.0.1:/raid/data/tmp/var#

However, trying to delete the rest of the files with input/output errors fails:

root@127.0.0.1:/raid/data/tmp/var# rm connections.tdb
rm: unable to stat `connections.tdb': Input/output error
root@127.0.0.1:/raid/data/tmp/var#

These other files seem to be causing problems (still cannot see N5200 on the network). The file /var/log/log.nmdb shows this is the case:

[2008/05/08 05:20:14, 0] nmbd/nmbd.c:main(697)
  Netbios nameserver version 3.0.26a started.
  Copyright Andrew Tridgell and the Samba Team 1992-2007
[2008/05/08 05:20:15, 0] nmbd/nmbd.c:main(716)
  standard input is not a socket, assuming -D option
[2008/05/08 05:22:58, 0] libsmb/unexpected.c:unexpected_packet(53)
  Failed to open unexpected.tdb

The file /var/log/samba/samba.smbd doesn't have the error about messages.tdb anymore:

root@127.0.0.1:/var/log/samba# vi samba.smbd
[2008/05/08 05:20:14, 0] profile/profile.c:init_clock_gettime(175)
  WARNING: __profile_clock != CLOCK_REALTIME: forced to use a slow profiling clo
[2008/05/08 05:20:14, 0] smbd/server.c:main(986)
  standard input is not a socket, assuming -D option
[2008/05/08 05:25:40, 0] profile/profile.c:init_clock_gettime(175)
  WARNING: __profile_clock != CLOCK_REALTIME: forced to use a slow profiling clo
[2008/05/08 05:25:40, 0] smbd/server.c:main(986)
  standard input is not a socket, assuming -D option
[2008/05/08 05:30:44, 0] profile/profile.c:init_clock_gettime(175)
  WARNING: __profile_clock != CLOCK_REALTIME: forced to use a slow profiling clo
[2008/05/08 05:30:44, 0] smbd/server.c:main(986)
  standard input is not a socket, assuming -D option
[2008/05/08 05:36:27, 6] param/loadparm.c:lp_file_list_changed(3077)
  lp_file_list_changed()
  file /etc/samba/smb.conf -> /etc/samba/smb.conf  last mod_time: Thu May  8 05:

[2008/05/08 05:36:27, 2] lib/interface.c:add_interface(81)
  added interface ip=192.168.1.100 bcast=192.168.1.255 nmask=255.255.255.0
[2008/05/08 05:36:27, 2] lib/interface.c:add_interface(81)
  added interface ip=192.168.2.254 bcast=192.168.2.255 nmask=255.255.255.0
[2008/05/08 05:36:27, 3] lib/util.c:interpret_addr(1309)
  sys_gethostbyname: Unknown host. wlan*
[2008/05/08 05:36:27, 2] lib/interface.c:interpret_interface(130)
- samba.smbd 1/225 0%

I'm going to work on deleting those other files in the raid/data/tmp/var folder...
Edit - "rm -f connections.tdb" does not work either.

[jbl]

I think I found a solution. Running a Filesystem check through the webui fixes the problem of the input/output errors in raid/data/tmp/var, and also fixes the problem of the Users, Groups and Batch Management screens being inaccessible. Then I can map a public (no ACL yet) network drive.

However, immediately after doing so, I rebooted the N5200, and the RAID became degraded. I checked, and the files in raid/data/tmp/var were still fine. However, after rebuilding the RAID, all the files in that folder had input/output errors again. So rebuilding a degraded RAID is what introduces those errors. This forced me to run another Filesystem Check, and I'm afraid to try rebooting for fear of degrading the RAID again.

I'll probably start a separate thread for being unable to map folders with ACL (i.e. non-public folders) as network drives. However, I've seen other postings for this error on the Thecus user group with no solutions, and I've had this problem since I bought the unit.

Many thanks to omega and pit234a and everyone else who replied. I was in way over my head and would never have been able to diagnose that needed services were not running.

[Thecus - Yvon]

Dear Sir,

After check your post and found the Windows paltform is HOME edition, due to HOME edition with much limiation for networking accounts, thus would having some issue with ACLed folders.
The Hexus forum is a platform for users'experience sharing and provide suggestion for Thecus, if need the Thecus Thechnical Support please submit a ticket from http://esupport.thecus.com/support/ since you sumit a ticket and Thecus staff willing reply your issue and under department supervise.
Also during sumit ticket process we will get the infromation about your environment, firmware version and etc, it is and plus for us to undertanding your environment and assist to reslove your issue.


Yvon.

[jbl]

Sorry, I have WinXP Pro, the first post was mistaken. I'll submit a ticket.

[omega]

@jbl,

based on what I can read from your postings, the filesystem on /raid/data is obviously damaged (indicated by these I/O errors).

Normally these errors should/could be fixed by doing a filesystem check (and repair). If this doesn't help another possibility would be to recreate the raid array.

Andreas