Posts by clickwir

  • Disabling IPv6 causes webserver to not start

    Pretty simple, I installed OMV 6.0.12-1 in a VM today. Don't want or need IPv6 on any of them right now, so it's pretty standard practice to disable IPv6 in grub (/etc/default/grub and add 'ipv6.disable=1' to the linux default line). But on reboot I noticed a console message about something failing to start.


    Turns out it's the webserver. Can't get to the web admin interface when IPv6 is disabled. I re-enabled, rebooted and it was fine. Disabled, rebooted and couldn't access it. I would like to be able to disable IPv6 and still use the system.


    This was in syslog, not sure if it's an important part.


    Feb 13 15:30:11 nas monit[434]: 'nas' Monit 5.27.2 started

    Feb 13 15:30:11 nas monit[434]: 'nginx' process is not running

    Feb 13 15:30:11 nas monit[434]: 'nginx' trying to restart

    Feb 13 15:30:11 nas monit[434]: 'nginx' start: '/bin/systemctl start nginx'

    Feb 13 15:30:11 nas systemd[1]: Starting A high performance web server and a reverse proxy server...

    Feb 13 15:30:11 nas nginx[733]: nginx: [emerg] socket() [::]:80 failed (97: Address family not supported by protocol

    )

    Feb 13 15:30:11 nas nginx[733]: nginx: configuration file /etc/nginx/nginx.conf test failed

    Feb 13 15:30:11 nas systemd[1]: nginx.service: Control process exited, code=exited, status=1/FAILURE

    Feb 13 15:30:11 nas systemd[1]: nginx.service: Failed with result 'exit-code'.

    Feb 13 15:30:11 nas systemd[1]: Failed to start A high performance web server and a reverse proxy server.


  • 6.0.5-2 to 6.0.5-3 - Salt errors during update

    Just did a run of the mill 'apt update && apt dist-upgrade' and saw this error. I don't understand what it means, if there's actually a problem and if I need to do anything.


    Other than that ERROR, the update seemed to go ok. Can login WebUI, NFS seems fine.

  • Noticed web crash trying to view filesystem.

    Tried again, from a different system with Firefox same error.


    Tried from mobile, same error.


    I have never accessed OMV from either of those systems.


    Restarted the Proxy after clearing cache, watched for cache_hits and they were all cache_miss.


    It's coming from somewhere, I don't know where.

  • Login screen animation uses fair amount of CPU

    Reasonable to one may be unreasonable to another. Since it's basically always using CPU time, it doesn't allow the CPU to clock down and use less power as often.


    It's not nothing, that's my point. The old screen used nothing, now it's using something.


    Granted, it's not --much--, but it's more than zero.


    Again, I like it, just think it should stop at some point.

  • Noticed web crash trying to view filesystem.

    First, I know my setup is a bit out of the norm. OMV is a VM on Proxmox and I'm doing single disk passthrough, not HBA. So OMV sees QEMU disks.


    On this I have BTRFS, then OMV is sharing it out to things like NFS. Fairly simple, but not normal.


    Doing to <Storage>, <File Systems>, I see 1 entry. Which is correct, that's my BTRFS RAID. I'm used to it not showing up correctly and OMV not knowing how to handle it.


    However, if I click on the row with the filesystem and click <Edit>, I get a crash.


    Code
    Software Failure.   Press left mouse button to continue. 
The requested page was not found.


    Again, I know it's out of the ordinary so this might be a non-issue. But I thought it best to bring it up and maybe a more graceful situation can be thought out.

  • Login screen animation uses fair amount of CPU

    Just updated to OMV 6 RC1, everything went easy. After poking around for a bit, I logged out and went about my business. At some point I noticed Firefox using 8% CPU, even with very few tabs open and doing nearly nothing. Turns out it's the animation on the login screen for OMV.


    It's beautiful and I like it, but can we have it stop after a few zooms? Or an option to disable it?


    I prefer to always keep a tab open of OMV sitting at the login screen, but I don't want to have it using up CPU time like that.

  • btrfs (and zfs) slow?

    That status page isn't the best, granted.


    However, many people have tried to get failures with raid56 and have not. It just works.


    I've been using single drive, RAID1 and RAID10 for many years and it's never has any issues.

  • btrfs (and zfs) slow?

    Keep in mind those are specific stress tests. They are looking for limits.


    I've used the car analogy many times. Car A has a top speed of 250mph. Car B, 240. Car C 180.


    The vast majority of the time and and vast majority of people will never hit those limits. They will never see a performance difference because they aren't running simulated workloads back to back and graphing the results.


    It's not an exact comparison; it's easier to hit limitations on computer hardware than a super fast car. But it helps to bring the thoughts out.


    I've been using btrfs for years under many different hardware and software setups. It's never been a concern of speed for me. Btrfs has always performed great, usually if I'm hitting some limit it's not because of btrfs.


    That said, btrfs has many more modern features that really should make it the default filesystem used for most systems.

  • Convert NTFS to EXT4

    FWIW, how I would do it would be like this: copy data off to another location. Create new ext4 (btrfs is what I would use) file system. Copy data back.


    You could probably even use the same path names after as well so you wouldn't have to update all the things pointing to it.

  • SAS Write Cache not enabling

    Have OMV 5.3.2 installed as a VM in Proxmox. 4 SAS drives are passed through (with scsiblock=1).


    Running 'sdparm -g WCE /dev/sdb' shows that Write Cache is disabled.


    Code
    WCE           0  [cha: y, def:  1, sav:  0]


    Go into the WebUI and enable Write Cache for the same drive (save and apply), sdparm still shows it disabled.


    So I used 'sdparm -s WCE --save /dev/sdb', then check what sdparm shows and yes it reports that it is enabled. WebUI shows it's disabled.

    Code
    WCE           1  [cha: y, def:  1, sav:  1]


    Should OMV work with SAS drives and Write Cache through the WebUI?

  • Can mount btrfs file system, cannot unmount

    So poking around in there I see 3 things referencing the filesystem in question.


    Not sure what I should do with them though.


  • Raid1 in Bios OMV detects 2 single drives

    Quote from ferry58

    I like the thought of having a auto backup in case one drive is faulty. If for what reason a file is deleted by accident it still is available in the waste bin.

    RAID is not backup.


    Backup is not RAID.


    Trash/Waste/Recycle bin, is not backup.


    This short video covers some of the high points.

    External Content www.youtube.com
    Content embedded from external sources will not be displayed without your consent.
    Through the activation of external content, you agree that personal data may be transferred to third party platforms. We have provided more information on this in our privacy policy.


    To your original question, some motherboard (fake) RAID solutions can still pass through the drives connected without adding an abstraction layer. This can get confusing because it could be doing a RAID1, but you would expect to only see one drive. It may in fact be doing a RAID1 but not hiding the other drive from you. This could cause you to be able to use both drives for different things. This would be bad.


    The suggestion is to remove the motherboards RAID settings, make sure it's not doing anything. Delete the RAID set it created, if it actually created one. Then use OMV to RAID the drives together.


    Remember, RAID1 is protecting uptime and availability. Backups protect data. That's a BIG difference and understanding that is important.