Very slow directory fetching in both OSX 10.8 and 10.9


Recommended Posts

Hi all, new to unRAID as I've taken over IT duties from a previous user. Both of us have limited knowledge in IT setups so hopefully you can provide some assistance.

 

Current setup: 24 iMacs on Cat6 running into a GS728TXS Netgear smart switch. Switch is patched into the chassis via SFP+. Chassis is a 24-bay with 3TB WD Green drives running Limewire unRAID. One single Mac Pro has a direct Fibre line to the switch as well. All computers experience the same issues, which is:

 

Upon first connecting to the NAS, there is a 20-30 sec delay while the Finder window fetches the listing of Shares. From here, there is typically a 10-30 sec delay in loading EVERY directory as we move further into the shares and away from the root. Sometimes when we leave a Finder window open and idle for a few mins, the directory will display a 20-30 delay next time we click into that directory.

 

Also, I've noticed that if we attempt to search within a folder using Spotlight, it will never search any sub-directory within the current search folder. We also use a program called VisualDiffer to compare directory contents between local and NAS folders. The program will often list a folder as empty on the NAS, but when going into the directory, there are most certainly files inside it. Doing another compare will then show the contents of the directory.

 

Additionally, if Finder is trying to fetch info from any of the NAS shares and is experiencing the 10-30 delay I'm referring to, ALL local folders will also exhibit identical delays in fetching directory info, despite being local. It's as if the NAS slows down the Finders fetching process across the board.

 

From what I understand WD Greens are terrible NAS drives due to their desire to spin down after 8 seconds of inactivity. Within the unRAID setup, we've told the OS to never spin down discs, but that did not seem to have any affect on load times. Unfortunately this is what we have to work with and replacing the HDD is likely not an option at this point.

 

Any ideas of what the slow down could be caused from, or any suggestions for a fix? Thanks.

Link to comment
  • 4 weeks later...
  • 3 months later...

I'm experiencing the same problem, also a friend of mine as well (his Unraid is nearly identical to mine).

We're both using v5.0.5 accessed from Mac OS (10.10.1 on mine, not sure about Paul's)

 

Syslog is attached.

 

SysInfo is

System: Gigabyte Technology Co., Ltd. - F2A88XM-D3H

CPU: AMD A4-5300 APU with Radeon HD Graphics - 3.4 GHz

Cache: 96 kB, 1024 kB

Memory: 12288 MB (max. 8 GB)

Network: eth0: 1000Mb/s - Full Duplex

Connections: One

 

Any help would be appreciated. My friend is thinking of buying a Drobo.

Unraid_syslog.txt

Link to comment

I've had a crack at using

smartctl  -a  -d  ata  /dev/md6

 

But get this error

Read Device Identity failed: Inappropriate ioctl for device

 

Regardless, through the dashboard I can see disk 6 has 197 errors.

 

It's a 3TB drive, so I'm quite happy to just pull it out and replace with a new 4TB drive rather than risk having it fail while in use.

Link to comment

Password:

Linux 3.9.11p-unRAID.

root@Tower:~# smartctl  -a  /dev/md6

smartctl 6.2 2013-07-26 r3841 [i686-linux-3.9.11p-unRAID] (local build)

Copyright © 2002-13, Bruce Allen, Christian Franke, www.smartmontools.org

 

/dev/md6: Unable to detect device type

Please specify device type with the -d option.

 

Use smartctl -h to get a usage summary

 

root@Tower:~#

 

Link to comment

Password:

Linux 3.9.11p-unRAID.

root@Tower:~# smartctl  -a  /dev/md6

smartctl 6.2 2013-07-26 r3841 [i686-linux-3.9.11p-unRAID] (local build)

Copyright © 2002-13, Bruce Allen, Christian Franke, www.smartmontools.org

 

/dev/md6: Unable to detect device type

Please specify device type with the -d option.

 

Use smartctl -h to get a usage summary

 

root@Tower:~#

You want to do smartctl -a /dev/sdx  (where x is the drive letter)  You can see it from the unRaid main screen

Link to comment

root@Tower:~# smartctl -a /dev/md6 

smartctl 6.2 2013-07-26 r3841 [i686-linux-3.9.11p-unRAID] (local build)

Copyright © 2002-13, Bruce Allen, Christian Franke, www.smartmontools.org

 

/dev/md6: Unable to detect device type

Please specify device type with the -d option.

 

Use smartctl -h to get a usage summary

 

root@Tower:~#

Link to comment

You want to do smartctl -a /dev/sdx  (where x is the drive letter)  You can see it from the unRaid main screen

root@Tower:~# smartctl -a /dev/md6

 

/dev/sd(?) is not /dev/md(?)

 

Look at the unraid main GUI screen. Find the line that has disk 6 in it. Find in that line the reference to /dev/sd?. Run the smartctl command using the appropriate letter with /dev/sd(?) instead of the ?.

Link to comment

Disk6 has a lot of reallocated sectors. I'd replace the drive and rebuild disk6. Then run preclear on the disk. I don't think it will pass.

 

How are you sitting down? The system was not shutdown correctly. Improper shutdowns cause parity checks on startup which will slow performance considerably. The numerous read errors on disk6 also slows performance.

 

Before rebuilding disk6 post SMART reports for the rest of the disks.

 

Link to comment

I've suffered several unexpected power cuts and with no UPS the server has suffered several unclean shutdowns.

It's also running a little too hot (we're in peak summer now), so I need to shut it down to move it to a cooler part of the house.

I'll fit a UPS at the same time.

The drive temps range from 35 deg C to 43 deg C (43 = 104 deg F)

 

SMART log is attached for all disks.

 

Thank you for your time looking at this.

SMART_log_-_all_disks.txt

Link to comment

I'm not certain if disk six is dead. It has a high reallocated sector count and this is a strong indication of impending failure. I would replace it with a pre-cleared spare and the run pre-clear of the disk. If the disk can survive 3 pre-clear passes then I'd use it but keep monitoring.

Link to comment
  • 1 month later...
  • 5 months later...

Just upgraded to V6.1, all went smoothly, parity check 106-155 MB/s, but I am have extremely slow Read Speeds and currently doing a back up of my OSX system drive and it very, very slow compared to V5.0.4... This is from OSX 6.8 so it is not just the newer OS's. I'm only 1/8 into the backup after 28 minutes and normally it would have been done around 20 minutes. I may have to go back to V5 if this can't be resolved. I'm now stuck with letting it finish, if I cancel I loss that backup and I can't work on this computer while this is going on. This is on my backups server.

studioserver-syslog-20150909-1100.zip

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.