Extremely slow access and read/write times on a nearly full drive


chooch

Recommended Posts

I have 3 5TB Toshiba drives and they are all nearly full (30-200 GB left empty on each of them).  Disk 3 has around 150 GB left and is very slow now when I access the drive.  I understand that drive performance slows down as they get close to full but none of my other 5TB drives are acting this slow.  Disk 2 actually has less free space then disk 3 and I'm not really seeing any performance issues with that drive.  I don't think it's network related because I used the mv -v function to move a 20GB movie onto disk 3 last night and it took a few hours.  Normally this would a few minutes so a few hours is a major change.  Also, when you click on disk 3 form windows or unraid it can take a very long time to access it and open the folder.  My current setup is:

 

Parity - 5TB TOSHIBA PH3500U-1I72

disk1- 5TB TOSHIBA PH3500U-1I72

disk2- 5TB TOSHIBA PH3500U-1I72

disk3- 5TB TOSHIBA PH3500U-1I72

disk4 - 2TB Hitatchi Deskstar 5K3000

disk5 - 2TB Hitatchi Deskstar 5K3000

 

Would removing the drive and rebuilding possibly help?  I'm not an expert when it comes to this and wanted to see if anyone had any suggestions.  Is there an optimal % that a drive should be left empty to avoid performance issues?  I do have a brand new 5TB drive I plan on adding soon to replace one of the 2TB drives so I could always put that in the slot for disk 3 for now and remove the current one for off-line testing.  I'm only running with 1GB of memory (1GB recently died) however unraid is showing that my memory is only being used at 70-75% and CPU 30-35% so I didn't think that is the issue.  Any help would be appreciated because while I am a long time unraid user I am far from an expert and just have a basic setup just for backing up my media.

Link to comment

What filesystem are the running? RieserFS is known to be really slow when nearly full.

 

Attached is a syslog from a few days ago and it appears I am running RieserFS.  What other file systems are there and is it too late to switch?  Or should I just leave a certain % empty if I am stuck with RieserFS.  I have since upgraded to 6.2.4 however even when I was on 5.0.6 disk 3 still behaved the same way.  I will update with a new syslog tonight but really the only thing that changed was upgrading from 5.0.6 to 6.2.4. 

 

Dec 12 23:17:38 Tower emhttp: shcmd (20): mkdir /mnt/disk1
Dec 12 23:17:38 Tower emhttp: shcmd (21): set -o pipefail ; mount -t reiserfs -o user_xattr,acl,noatime,nodiratime /dev/md1 /mnt/disk1 |& logger
Dec 12 23:17:38 Tower kernel: REISERFS (device md1): found reiserfs format "3.6" with standard journal
Dec 12 23:17:38 Tower kernel: REISERFS (device md1): using ordered data mode
Dec 12 23:17:38 Tower kernel: reiserfs: using flush barriers
Dec 12 23:17:38 Tower kernel: REISERFS (device md1): journal params: device md1, size 8192, journal first block 18, max trans len 1024, max batch 900, max commit age 30, max trans age 30
Dec 12 23:17:38 Tower kernel: REISERFS (device md1): checking transaction log (md1)
Dec 12 23:17:38 Tower kernel: REISERFS (device md1): Using r5 hash to sort names
Dec 12 23:17:38 Tower emhttp: shcmd (22): mkdir /mnt/disk2
Dec 12 23:17:38 Tower emhttp: shcmd (23): set -o pipefail ; mount -t reiserfs -o user_xattr,acl,noatime,nodiratime /dev/md2 /mnt/disk2 |& logger
Dec 12 23:17:38 Tower kernel: REISERFS (device md2): found reiserfs format "3.6" with standard journal
Dec 12 23:17:38 Tower kernel: REISERFS (device md2): using ordered data mode
Dec 12 23:17:38 Tower kernel: reiserfs: using flush barriers
Dec 12 23:17:38 Tower kernel: REISERFS (device md2): journal params: device md2, size 8192, journal first block 18, max trans len 1024, max batch 900, max commit age 30, max trans age 30
Dec 12 23:17:38 Tower kernel: REISERFS (device md2): checking transaction log (md2)
Dec 12 23:17:38 Tower kernel: REISERFS (device md2): Using r5 hash to sort names
Dec 12 23:17:38 Tower emhttp: shcmd (24): mkdir /mnt/disk3
Dec 12 23:17:38 Tower emhttp: shcmd (25): set -o pipefail ; mount -t reiserfs -o user_xattr,acl,noatime,nodiratime /dev/md3 /mnt/disk3 |& logger
Dec 12 23:17:38 Tower kernel: REISERFS (device md3): found reiserfs format "3.6" with standard journal
Dec 12 23:17:38 Tower kernel: REISERFS (device md3): using ordered data mode
Dec 12 23:17:38 Tower kernel: reiserfs: using flush barriers
Dec 12 23:17:38 Tower kernel: REISERFS (device md3): journal params: device md3, size 8192, journal first block 18, max trans len 1024, max batch 900, max commit age 30, max trans age 30
Dec 12 23:17:38 Tower kernel: REISERFS (device md3): checking transaction log (md3)
Dec 12 23:17:38 Tower kernel: REISERFS (device md3): Using r5 hash to sort names
Dec 12 23:17:38 Tower emhttp: shcmd (26): mkdir /mnt/disk4
Dec 12 23:17:38 Tower emhttp: shcmd (27): set -o pipefail ; mount -t reiserfs -o user_xattr,acl,noatime,nodiratime /dev/md4 /mnt/disk4 |& logger
Dec 12 23:17:38 Tower kernel: REISERFS (device md4): found reiserfs format "3.6" with standard journal
Dec 12 23:17:38 Tower kernel: REISERFS (device md4): using ordered data mode
Dec 12 23:17:38 Tower kernel: reiserfs: using flush barriers
Dec 12 23:17:38 Tower kernel: REISERFS (device md4): journal params: device md4, size 8192, journal first block 18, max trans len 1024, max batch 900, max commit age 30, max trans age 30
Dec 12 23:17:38 Tower kernel: REISERFS (device md4): checking transaction log (md4)
Dec 12 23:17:38 Tower kernel: REISERFS (device md4): Using r5 hash to sort names
Dec 12 23:17:38 Tower emhttp: shcmd (28): mkdir /mnt/disk5
Dec 12 23:17:38 Tower emhttp: shcmd (29): set -o pipefail ; mount -t reiserfs -o user_xattr,acl,noatime,nodiratime /dev/md5 /mnt/disk5 |& logger
Dec 12 23:17:39 Tower kernel: REISERFS (device md5): found reiserfs format "3.6" with standard journal
Dec 12 23:17:39 Tower kernel: REISERFS (device md5): using ordered data mode
Dec 12 23:17:39 Tower kernel: reiserfs: using flush barriers
Dec 12 23:17:39 Tower kernel: REISERFS (device md5): journal params: device md5, size 8192, journal first block 18, max trans len 1024, max batch 900, max commit age 30, max trans age 30
Dec 12 23:17:39 Tower kernel: REISERFS (device md5): checking transaction log (md5)
Dec 12 23:17:39 Tower kernel: REISERFS (device md5): Using r5 hash to sort names

syslog.txt

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.