Several problems running OMV 3

    This site uses cookies. By continuing to browse this site, you are agreeing to our Cookie Policy.

    • Several problems running OMV 3

      Hi there,

      OMV was running very stable for about half a year now.
      Now I am afraid of some things I would like to change:

      The SMART monitoring is showing a red status for the attribute name "RAW_READ_ERROR_RATE" -> 1 Raw_Read_Error_Rate POSR-K 001 001 051 NOW 102366
      While booting the microserver is showing the message: "PORT3: NOT SMART-capable hard disk". Running OMV, SMART is activated and working for all hard disks. It is the same vendor and hard disk like the others.

      Sometimes the system is showing this:[IMG:https://thumb.ibb.co/k2kMzS/IMG_2838.jpg]


      My solution for using 4 disks and running the system on a different disk:
      A SD-Card is pointing to the internal System-HDD connected to CDROM-SATA-Port. Since 2 weeks the system does not come up. In this situation it seems like the system is not able to find the system HDD, because the SD-card is not being found to point to the system HDD. After rebooting 2-3 times and nothing nothing different, the system is working as it is meant to.

      So, some help would be very great... is the SMART-error a warning to change the HDD on port 3 ? And, why is the system sometimes not able to find the internal SD card ?
    • It seems like you may be dealing with 2 problems:

      - An SD card that may be marginal or going bad.
      - A hard drive that is getting older.

      On the hard drive:
      The following are the SMART attributes that predict failure with decent accuracy.

      SMART 5 – Reallocated_Sector_Count.
      SMART 187 – Reported_Uncorrectable_Errors.
      SMART 188 – Command_Timeout.
      SMART 197 – Current_Pending_Sector_Count.
      SMART 198 – Offline_Uncorrectable.

      If the hard drive is beginning to fail, there's nothing to be done. Replace it.
      ________________________________________________

      On the SD card:
      Did you install the flashmemory plugin? If not, after 6 months, your SD-card is probably getting ready to fail.

      In this -> User Guide, the processes required to install the flashmemory plugin fully are detailed, with the needed manual edits to /etc/fstab. Also, the process of cloning an SD-card is covered so you'll have backup for your boot drive - SD-card.

      Unfortunately, going on what you've described (and assuming the flashmemory plugin was not installed), I wouldn't trust the SD-card you're now using. I'd rebuild from scratch with a new card, mount the existing data drives, reconfigure everything, then clone the cleanly built SD-card. (I would throw the old one away.)

      Video Guides :!: New User Guide :!: Docker Guides :!: Pi-hole in Docker
      Good backup takes the "drama" out of computing.
      ____________________________________
      Primary: OMV 3.0.99, ThinkServer TS140, 12GB ECC, 32GB USB boot, 4TB+4TB zmirror, 3TB client backup.
      OMV 4.1.13, Intel Server SC5650HCBRP, 32GB ECC, 16GB USB boot, UnionFS+SNAPRAID
      Backup: OMV 4.1.9, Acer RC-111, 4GB, 32GB USB boot, 3TB+3TB zmirror, 4TB Rsync'ed disk
    • Thanks for your answer... do you recommend cloning (image restore) the existing sd card to a new one without the whole new building process?

      So, hope, I understood right according to the SMART message:
      Your list does not contain the raw_read_error_rate my hdd is telling me. Does this mean it is not important to change the hard drive immediately?
    • (Sorry - I was out of town for a few days.)

      I wouldn't trust the existing SD-card. And since the process of fully testing the card is destructive, I'd rebuild it from scratch, configure it up the way I wanted it, then clone it.

      The user guide describes the build process and how to test and clone flash media.
      _________________________________

      The SMART list above is geared toward predicting drive failure. For example, if you had 1 or 2 reallocated sectors (SMART 5), it might not be a big deal. But, if SMART 5 started incrementing, future drive failure can be expected.

      The raw read error rate attribute is certainly not a good thing, 0 is preferred, but it must be noted that hard drives are mechanical devices so preventative measures are designed into them. The raw read error rate is the drive detecting that it read bad data and (hopefully) corrected it with ECC code.
      Unfortunately, these raw counts and what they mean vary from OEM to OEM. Seagate may assign a logarithmic value to it where WD may assign a discrete linear value. Unless you research the drive's OEM pages, the raw values of misc. attributes can't be interpreted.
      _________________________________

      What I'd do is test the file system of the hard drive. What filesystem do you have on it?

      Video Guides :!: New User Guide :!: Docker Guides :!: Pi-hole in Docker
      Good backup takes the "drama" out of computing.
      ____________________________________
      Primary: OMV 3.0.99, ThinkServer TS140, 12GB ECC, 32GB USB boot, 4TB+4TB zmirror, 3TB client backup.
      OMV 4.1.13, Intel Server SC5650HCBRP, 32GB ECC, 16GB USB boot, UnionFS+SNAPRAID
      Backup: OMV 4.1.9, Acer RC-111, 4GB, 32GB USB boot, 3TB+3TB zmirror, 4TB Rsync'ed disk
    • Ok, re-biult the sd card... now, omv is booting without problems...

      2 more questions, please:

      1) I modified /etc/fstab as described for flash drives-plugin. But, do I need to activate the plugin itself afterwards ?
      2) HDD /dev/sdc is pre-failing smart-test (raw_read_error)... don´t really know how to test the file system. And, if I want to replace the hdd: currently the raid 5 consists of 4 x 2GB WD red. Replacing the third 2GB-WD-hdd (sdc) with a Seagate Iron Wolf 4GB, for example, does not lead to a problem ? Just shutting down the ProLiant, replacing the hdd, booting it into OMV 3 and the RAID 5 will rebuild all automatically ?

      Thanks in advance.
    • Blebbens wrote:

      I modified /etc/fstab as described for flash drives-plugin. But, do I need to activate the plugin itself afterwards ?
      Not anymore. Installing it means it is "activated".
      omv 4.1.14 arrakis | 64 bit | 4.15 proxmox kernel | omvextrasorg 4.1.13
      omv-extras.org plugins source code and issue tracker - github

      Please read this before posting a question and this and this for docker questions.
      Please don't PM for support... Too many PMs!
    • Blebbens wrote:

      1) I modified /etc/fstab as described for flash drives-plugin. But, do I need to activate the plugin itself afterwards ?2) HDD /dev/sdc is pre-failing smart-test (raw_read_error)...
      a. don´t really know how to test the file system.
      b. And, if I want to replace the hdd: currently the raid 5 consists of 4 x 2GB WD red. Replacing the third 2GB-WD-hdd (sdc) with a Seagate Iron Wolf 4GB, for example, does not lead to a problem ?
      c. Just shutting down the ProLiant, replacing the hdd, booting it into OMV 3 and the RAID 5 will rebuild all automatically ?

      Thanks in advance.
      1. the plugin is active. The "activate" thing is a "cling-on" from an earlier version of the plugin.

      2.
      a. I asked before :) - what file system are you using? (EXT4?)
      b. You'll only get the use of 2GB, of the 4GB drive, but there's no other consequence.
      c. The rebuild will not be automatic, unless you add the 4GB drive to the array as a spare before removing the 2GB drive. If you have the space in the ProLiant, I would add the 4GB drive to the array as a spare.

      Do you have Putty installed on a client? Can you get to the server's command line as root?

      Video Guides :!: New User Guide :!: Docker Guides :!: Pi-hole in Docker
      Good backup takes the "drama" out of computing.
      ____________________________________
      Primary: OMV 3.0.99, ThinkServer TS140, 12GB ECC, 32GB USB boot, 4TB+4TB zmirror, 3TB client backup.
      OMV 4.1.13, Intel Server SC5650HCBRP, 32GB ECC, 16GB USB boot, UnionFS+SNAPRAID
      Backup: OMV 4.1.9, Acer RC-111, 4GB, 32GB USB boot, 3TB+3TB zmirror, 4TB Rsync'ed disk
    • The RAID5 uses ext4, yes.
      Ordered a WD red 2TB.
      There is no space left in the ProLiant, it is a 4bay server, filled with 4 x 2TB hdds. I have to replace the defect hdd.

      Have access to mac and windows desktops, can use terminal or putty for ssh. Getting access to omv as root is no problem.
    • As a threshold warning:
      If you have irreplaceable data on the array, back it up now. (Maybe an external USB drive?) "Stuff" can go wrong in the following.
      __________________________________

      First, I'd replace the drive. Since you don't have space for a spare drive, what you'll need to do requires a few steps but the process can be done in the GUI.

      Confirm which drive you're replacing /dev/sd?. (This can done by comparing info under Physical Disks to what's in SMART, the Devices tab, in the Information button, and the Attributes tab.)

      Go into Storage, RAID Management:
      - Click on your array, then the - Remove button.
      - Select the appropriate drive and OK (apply the change)
      (The array State goes from active to degraded.)

      Shutdown and physically replace the drive.

      Boot up, quick wipe the new drive. (The new drive will likely show up in the same place as the old drive, in Physical Disks.)


      Go back into Storage, RAID Management:
      - Click on your array, then the + Recover button.
      - Select the new drive. (In your case, it should be the only drive available in the devices list.) Click OK. (apply the change)


      The State should be active, degraded, recovering. Give it awhile to finish sync'ing.
      __________________________________

      Second, if you want to check the file system, realize there's risk in this as well. The "EXT4" equivalent of Windows NTFS chkdsk is fsck. And just like chkdsk, fsck can "fix" you right out of a working filesystem. On the other hand, if you have file system issues, the risk must be taken to straighten out problems. Again, backing up irreplaceable data is strongly recommended. With full data backup, there's nothing to fear.

      SSH into the server as root.

      Unmount the filesystem/array. Note that unmounting is required for accurate results.
      (The following assumes your array device name is /dev/md0)

      umount /dev/md0

      fsck -n /dev/md0

      The -n switch will prevent changes. If it comes back clean, for example; RAID5EXT4: clean, 447/983040 files, 564610/3929088 blocks, everything is OK and you're done. If there are problems reconsider backing up, if you haven't done so, and then doing the following:

      fsck -p /dev/md0

      The above does automatic repair without prompting you with questions.

      Finally:

      mount /dev/md0

      and you're back in business.

      Video Guides :!: New User Guide :!: Docker Guides :!: Pi-hole in Docker
      Good backup takes the "drama" out of computing.
      ____________________________________
      Primary: OMV 3.0.99, ThinkServer TS140, 12GB ECC, 32GB USB boot, 4TB+4TB zmirror, 3TB client backup.
      OMV 4.1.13, Intel Server SC5650HCBRP, 32GB ECC, 16GB USB boot, UnionFS+SNAPRAID
      Backup: OMV 4.1.9, Acer RC-111, 4GB, 32GB USB boot, 3TB+3TB zmirror, 4TB Rsync'ed disk

      The post was edited 1 time, last by flmaxey: edit ().

    • While your SD-card is clean, clone it. In the guide I mentioned, there's guidance on when to consider updating and rotating in the cloned boot drive.

      Video Guides :!: New User Guide :!: Docker Guides :!: Pi-hole in Docker
      Good backup takes the "drama" out of computing.
      ____________________________________
      Primary: OMV 3.0.99, ThinkServer TS140, 12GB ECC, 32GB USB boot, 4TB+4TB zmirror, 3TB client backup.
      OMV 4.1.13, Intel Server SC5650HCBRP, 32GB ECC, 16GB USB boot, UnionFS+SNAPRAID
      Backup: OMV 4.1.9, Acer RC-111, 4GB, 32GB USB boot, 3TB+3TB zmirror, 4TB Rsync'ed disk
    • Hi flmaxey,

      Yesterday I removed the old hdd and switched to a new 3TB seagate iron wolf.
      The raid is working fine now and showing the status "clean".

      Now I wanted to do a fsck.
      After executing sudo umount /dev/md0, it tells me the target is busy.
      the command fsck is not found by the system.
    • Blebbens wrote:

      Now I wanted to do a fsck.After executing sudo umount /dev/md0, it tells me the target is busy.
      the command fsck is not found by the system.
      After using umount on my test step up, I got the busy message as well. It was necessary to stop my Dockers (which link to the array) and, while I didn't have to do it, it might be best if you log out of OMV.
      ____________________________________________________________________

      fsck is not found by the system? All versions of fsck (there are several) are located under /sbin

      Do the following, and copy and paste the output.

      e2fsck -n /dev/md0

      Video Guides :!: New User Guide :!: Docker Guides :!: Pi-hole in Docker
      Good backup takes the "drama" out of computing.
      ____________________________________
      Primary: OMV 3.0.99, ThinkServer TS140, 12GB ECC, 32GB USB boot, 4TB+4TB zmirror, 3TB client backup.
      OMV 4.1.13, Intel Server SC5650HCBRP, 32GB ECC, 16GB USB boot, UnionFS+SNAPRAID
      Backup: OMV 4.1.9, Acer RC-111, 4GB, 32GB USB boot, 3TB+3TB zmirror, 4TB Rsync'ed disk

      The post was edited 3 times, last by flmaxey: edit ().

    • Frankly, I'm not sure how to advise you from here. fsck is standard, among several other essential file system programs, under the /sbin folder. In my test setup, under /sbin, there are 4 symbolic links and 8 fsck files for the various file system types. If they're not there, there's no knowing what else may be missing as well. If it was me, with missing files, I'd rebuild.

      I have to ask, did you rebuild your existing SD-card or did you rebuild on a new card? (And)
      Do you have backup?

      Video Guides :!: New User Guide :!: Docker Guides :!: Pi-hole in Docker
      Good backup takes the "drama" out of computing.
      ____________________________________
      Primary: OMV 3.0.99, ThinkServer TS140, 12GB ECC, 32GB USB boot, 4TB+4TB zmirror, 3TB client backup.
      OMV 4.1.13, Intel Server SC5650HCBRP, 32GB ECC, 16GB USB boot, UnionFS+SNAPRAID
      Backup: OMV 4.1.9, Acer RC-111, 4GB, 32GB USB boot, 3TB+3TB zmirror, 4TB Rsync'ed disk
    • If it's clean (suber), your filesystem (Ext4) is OK.

      That's good because discovering what is keeping the array busy might have been an ordeal.

      Don't forget to clone your boot card.

      Video Guides :!: New User Guide :!: Docker Guides :!: Pi-hole in Docker
      Good backup takes the "drama" out of computing.
      ____________________________________
      Primary: OMV 3.0.99, ThinkServer TS140, 12GB ECC, 32GB USB boot, 4TB+4TB zmirror, 3TB client backup.
      OMV 4.1.13, Intel Server SC5650HCBRP, 32GB ECC, 16GB USB boot, UnionFS+SNAPRAID
      Backup: OMV 4.1.9, Acer RC-111, 4GB, 32GB USB boot, 3TB+3TB zmirror, 4TB Rsync'ed disk