OS:                       v5.0-beta7

CPU:                     AMD Athlon 64 Processor 3200+

Motherboard:        MSI K9n Platinum

RAM:                     2 Gig of Corsair

Case:                    AZZA Helios 910 CSAZ-910 Mid-Tower Case (Black)

Drive Cage(s):       None

Power Supply:       CORSAIR Builder Series CX500 V2 500W  (DOA) -  

                            Corsair CMPSU-850TX 850W ATX12V v2.2

USB Flash Drive:    Lexar JumpDrive FireFly 4GB

Parity Drive:           HITACHI Deskstar 5K3000 2TB

Data Drive 1:         HITACHI Deskstar 5K3000 2TB

Data Drive 2:         HITACHI Deskstar 5K3000 2TB

 

 

This is my first build, and I was hoping someone could go through my system log to see if anything sticks out.  Currently all three drives are doing their pre-clear at 107 to 110 MB/s @ 30 to 31C.

syslog.txt

Looks like I may have an issue with one of my new drives. The write speed on SBD is much slower than the other two. They are all plugged directly on the board via the SATA headers.

 

The time stamp is when the progress email came in.

 

Zeroing Disk /dev/sdb in progress: 25% complete.    Calculated Write Speed: 65 MB/s    5:53 AM

Zeroing Disk /dev/sdc in progress: 25% complete.    Calculated Write Speed: 116 MB/s  5:11 AM

Zeroing Disk /dev/sdd in progress: 25% complete.    Calculated Write Speed: 118 MB/s  5:11 AM

 

Zeroing Disk /dev/sdb in progress: 50% complete.    Calculated Write Speed: 68 MB/s    7:48 AM

Zeroing Disk /dev/sdc in progress: 50% complete.    Calculated Write Speed: 111 MB/s  6:29 AM

Zeroing Disk /dev/sdd in progress: 50% complete.    Calculated Write Speed: 113 MB/s  6:28 AM

 

SDB (No email as of 8:35 AM)

Zeroing Disk /dev/sdc in progress: 75% complete.    Calculated Write Speed: 103 MB/s  8:00 AM

Zeroing Disk /dev/sdd in progress: 75% complete.    Calculated Write Speed: 105 MB/s  7:57 AM

Your syslog looks normal.  One drive was mounted as SATAI instead of SATAII, so presumably that is SDB.  It could be that SDB is bad, or it could be that the SATA port or cable are bad.  First step is to let the preclears finish, then post the output reports here.  Second, I would simply reseat the connection to SDB, it could be a loose cable.  Also try another cable and another SATA port if you have that option.  Now run another pass of preclear on SDB (you can start using the other two drives in your array if you are in a hurry).  If SDB still acts up after all that, then it is likely to be a bad drive.

Well after testing, I found my Sata1 header will only do 1.5 Gig for some reason. No luck with a bios reset or hard coding anything. It isn't that big of a deal since I will need SATA cards in it anyway.

 

I also did some testing to make sure I knew how to use it before I put production data on it.

- Copied data over to it without parity, and was getting 96 MB to 117 MB.

- Built the parity drive.

- Simulated a drive a failure, and rebuilt a new drive.

 

I still have one thing I have not figured out. No matter what I put on the user share, I only get RO on the user share, and the disk share is RW to everyone. I also can't figure out how to make the disk shares either not show up or hidden.

 

SMB Settings

Export - yes

Security - secure

 

User Access

root - RO

Uriel - RW

 

Edit:  I definitely have something wrong... I tried creating a new account with no luck too... Please note this is a fresh install of 5.0 b7, and not an upgrade...

 

Jun 26 11:40:00 Tower logger: useradd: invalid user name 'Uriel'

Jun 26 11:40:00 Tower emhttp: shcmd (95): chpasswd <<< Uriel:"*****"

Jun 26 11:40:00 Tower emhttp: _shcmd: shcmd (95): exit status: 1

Jun 26 11:40:00 Tower emhttp: shcmd (96): smbpasswd -L -s -a Uriel <<< "*****"$'\n'"*****"

Jun 26 11:40:00 Tower emhttp: _shcmd: shcmd (96): exit status: 1

Jun 26 11:40:00 Tower emhttp: shcmd (97): cp /etc/passwd /etc/shadow /etc/samba/private/smbpasswd /boot/config

Jun 26 11:40:46 Tower emhttp: shcmd (98): useradd -g users -d / -s /bin/false -c 'uriel' uriel 2>&1 |logger

Jun 26 11:40:46 Tower useradd[13199]: new user: name=uriel, UID=1001, GID=100, home=/, shell=/bin/false

Jun 26 11:40:46 Tower logger: /usr/sbin/nscd: relocation error: /usr/sbin/nscd: symbol __nss_services_lookup, version GLIBC_PRIVATE not defined in file libc.so.6 with link time reference

Jun 26 11:40:46 Tower last message repeated 3 times

Jun 26 11:40:46 Tower emhttp: shcmd (99): chpasswd <<< uriel:"*****"

Jun 26 11:40:46 Tower emhttp: shcmd (100): smbpasswd -L -s -a uriel <<< "*****"$'\n'"*****"

Jun 26 11:40:46 Tower emhttp: shcmd (101): cp /etc/passwd /etc/shadow /etc/samba/private/smbpasswd /boot/config

 

 

This is what I found...

 

 

root@Tower:~# ls -l /mnt/user/

total 56

drwxrwx--- 1 nobody users    72 Jun 25 09:25 MediaCenter/

-rw-rw---- 1 nobody users 54544 Jan 25  1998 ROBOCOPY.EXE

 

I manually added myself to the root group, but still no access.

root@Tower:~# id murphrjo

uid=1000(murphrjo) gid=100(users) groups=100(users),0(root)

 

I tried to remove myself from the users group.

root@Tower:~# usermod -G users murphrjo

/usr/sbin/nscd: relocation error: /usr/sbin/nscd: symbol __nss_services_lookup, version GLIBC_PRIVATE not defined in file libc.so.6 with link time reference

/usr/sbin/nscd: relocation error: /usr/sbin/nscd: symbol __nss_services_lookup, version GLIBC_PRIVATE not defined in file libc.so.6 with link time reference

/usr/sbin/nscd: relocation error: /usr/sbin/nscd: symbol __nss_services_lookup, version GLIBC_PRIVATE not defined in file libc.so.6 with link time reference

/usr/sbin/nscd: relocation error: /usr/sbin/nscd: symbol __nss_services_lookup, version GLIBC_PRIVATE not defined in file libc.so.6 with link time reference

 

 

I also noticed I get the same error above when trying to create a new user from the web interface.

 

To kill time I am going to reinstall the unRAID os to the flash drive. Maybe something got corrupt during the first copy or during a plugin install.

 

 

 

 

I don't use the betas, so I can't advise you on your permissions issues.  I know how to do what you want in 4.7, if you want to revert to that.

 

When you build parity, always run a parity check immediately afterwards.  The parity build writes to the entire parity disk, the check reads from it.  Until you have completely both without errors, you can't trust your parity data.

Did you ever find out what the problem was regarding to the error:

/usr/sbin/nscd: relocation error: /usr/sbin/nscd: symbol __nss_services_lookup, version GLIBC_PRIVATE not defined in file libc.so.6 with link time reference

 

I'm seeing this message 4 times now. Just set up with the b12 release and following the configuration tutorial. done nothing special yet. currently preclearing 4x disks.

Wow that was a ways back. I do remember I did copy down v5.0-beta7 to the flash drive again. I also found out I had to reboot after I made some changes through the web interface for it to take effect.

 

And one other thing. At the time I was testing, I was only using the 3 drive free version which only supports simple user shares. It does not support User Level Security. After I got my keys, everything started working as I wanted it to outside from the one SATA port that still only does 1.5 gig/s.

 

http://lime-technology.com/products/registration-keys

Seeing the date posted indeed Thanks though for replying. I saw them passing by on the physical console. I also rebooted and have not seen the messages since. I also use the free basic edition as I'm new to this all. I guess I'll wait it out Thanks again.