Beiträge von Skello

I think I'm not doing something right:

Even if I set Host Type as name-based in the Nginx plug-in, set the subdomain name, then I choose enable SSL, port 443 and "only use SSL" and save and apply settings, then I get an error and OMV UI dies. That's because the plug-in creates both listen directives creating conflict with the OMV UI which also runs on 443. Here is the config the plug-in generates (changed real domain name):

server {
    listen 443 ssl;
    listen [::]:443 ssl;
    ssl_certificate     /etc/ssl/certs/openmediavault-63538c91-1c8a-45b1-b6b4-01aa85fd605b.crt;
    ssl_certificate_key /etc/ssl/private/openmediavault-63538c91-1c8a-45b1-b6b4-01aa85fd605b.key;
    server_name torrent.lan.example.com;
    set $root_path "/srv/dev-disk-by-label-store/public_html";
    root $root_path;
    index index.html;
    access_log /var/log/nginx/668865ad-30c8-4452-b53b-713f7030ce62-access.log;
    error_log  /var/log/nginx/668865ad-30c8-4452-b53b-713f7030ce62-error.log;
    large_client_header_buffers 4 32k;
    location ^~ / {
proxy_pass http://127.0.0.1:9091;
proxy_http_version 1.1;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $remote_addr;
proxy_read_timeout 1d;
}
}

If I manually edit the generated zzz-omv-nginx config file and delete the first IPv4 listen directive and leave only listen [::]:443 ssl; and restart nginx then it works. However, when I go back into OMV UI it will say I have unsaved settings (probably because of the manual edit) and if I save the settings it adds the listen 443 ssl; directive back, crashing the UI again. Not sure what I'm doing wrong.

Hmm. There appears to be a problem. If I apply some changes in the OMV UI, the Nginx plug-in adds the listen 443 ssl; directive back to the zzz-omv-nginx configuration file. This crashes the OMV UI. Maybe the plug-in doesn't have support for multiple services on the same port? Not sure. Do I even need the Nginx plug-in installed to achieve what I want?

Nice. Thank you!

Changing to listen [::]:443 ssl; worked!

Hi,

Since Let's Encrypt has recently launched wildcard certificates I've decided to set up a subdomain with wildcard certificate for all HTTP services on my LAN. Essentially I have an acme.sh client running as a cron job on a box, configured to perform DNS-based validation for the subdomain via Cloudflare API, renew the certificate and automatically deploy it to all services via SSH, this includes my router for the web-based admin interface, Proxmox and OMV. This allows me not to expose any services externally for domain validation, which is nice. So let's say the subdomain is lan.example.com and the certificate is for *.lan.example.com. My local DNS server is configured to resolve all lan.example.com hosts inside the LAN, OMV being omv.lan.example.com for example.

Now, I figured I could also have separate hosts for all web services running on OMV aside from the admin UI, for example Transmission, so I installed the Nginx plug-in and tried to configure as reverse proxy with name-based server blocks. However, is there a way to have all these hostsreachable on port 443? For example my nginx configuration for transmission looks like this:

server {
    listen 80;
    listen 443 ssl;
    ssl_certificate     /etc/ssl/certs/openmediavault-63538c91-1c8a-45b1-b6b4-01aa85fd605b.crt;
    ssl_certificate_key /etc/ssl/private/openmediavault-63538c91-1c8a-45b1-b6b4-01aa85fd605b.key;
    server_name torrent.lan.example.com;
    root /var/www/torrent.lan.example.com;
    index index.html;
    access_log /var/log/nginx/10a54529-eea8-4ebf-b3ad-514a736893e0-access.log;
    error_log  /var/log/nginx/10a54529-eea8-4ebf-b3ad-514a736893e0-error.log;
    large_client_header_buffers 4 32k;
    location ^~ / {
proxy_pass http://127.0.0.1:9091;
proxy_http_version 1.1;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $remote_addr;
proxy_read_timeout 1d;
}
}

This fails. Either the OMV UI stops responding or nginx fails to restart with multiple errors:
Mar 18 21:32:16 omv nginx[17833]: nginx: [emerg] bind() to 0.0.0.0:80 failed (98: Address already in use)
Mar 18 21:32:16 omv nginx[17833]: nginx: [emerg] bind() to 0.0.0.0:443 failed (98: Address already in use)

Is there any way to make this work with different hosts/websites on the same port? If I change the ports to something else, it works, but I'd really like to have clean URLs. For example I'd like https://omv.lan.example.com to point to the OMV UI, while https://torrent.lan.example.com to point to Transmission interface, even though they are both on the same machine/IP address.

Hi,

I'm posting here because the topic of the thread matches what I'm currently doing.

In the process of migrating from OMV2 to OMV3 I've decided to go virtual with Proxmox. I'm going to document my steps with as much detail as possible in case someone wants to point out something I'm doing wrong or a better option.

I'm using PVE 5.1-36 with kernel 4.13.8-2-pve. I've installed OMV3 in a VM with a 32GB virtual drive (SCSI-VirtIO single with iothread -- it's my understanding this is the most performant option). The default OMV kernel installed is 4.9.0-0.bpo.4-amd64, but after installing OMV-Extras I have the option to switch to an older kernel compiled for PVE: pve-kernel-4.4.95-1-pve. Which kernel should I use? 4.9.0 bpo or 4.4.95 pve?

I finally solved it!

I figured I need to solve the inability to unmount the primary /dev/sdc1 partition first. I stopped NFS, Transmission and anything else that was using that partition, but could not delete the shared folder configured on it. It simply didn't have the active delete button. So I figured "OK then, I'll edit the shared folder and assign it to a different partition temporarily." Once I did that, the Unmount button appeared for /dev/sdc1 and I could unmount the partition without a problem.

Next I fired up local gparted through SSH with X forwarding and didn't get any error anymore about the partitions on that drive being locked. I created the second EXT4 partition, with a label and it worked without a hitch. After a few seconds it appeared in the OMV interface and I could mount it without any problem.

Thanks for answering subzero79.

I have tried the backports kernel, but no luck. Still can't mount it. I used GParted Live to remove the partition and then create it as unformatted, but for some reason it still formats it as ext4 and assigns the previous label. I then selected "cleared" instead of "unformatted" as file system. OMV doesn't see it under "File Systems," but another problem is that if I hit "Create" I can't select any HDD from the "select a device" dropdown list. For a brief second it shows like the dropdown list is loading, then it doesn't even open anymore.

Hello,

So here's the background and my problem. I had a 3TB USB3.0 hard drive with a single EXT4 partition that was attached to a Linux-based embedded device (router). I attached this hard disk to OMV some time ago and the partition mounted. The were some sort of warning in the interface, but I can't remember what it was. Anyway, the partition was mounted and got mounted at each reboot since. For all intents and purposes it functioned properly, the only problem being that when I select it in the interface the "Unmount" button is not active so I can't seem to unmount it. The umount command doesn't work either. But that's not the main issue.

I wanted a new partition on this drive, so I booted off a GParted Live USB stick and resized the old partition and created a new one of 1TB. All went well, or so I thought. First time when I booted into OMV it went OK, but once in the interface I attempted to mount the new /dev/sdc2 partition. It failed with this error:

Error #6000:
exception 'OMVException' with message 'Failed to mount 'febf09f0-2618-47fc-a80d-cdaf6b6d4f89': mount: wrong fs type, bad option, bad superblock on /dev/sdc2,
missing codepage or helper program, or other error
In some cases useful info is found in syslog - try
dmesg | tail or so
' in /usr/share/openmediavault/engined/module/fstab.inc:90

dmesg | tail shows:

[ 337.648757] EXT4-fs (sdc2): couldn't mount RDWR because of unsupported optional features (400)
[ 1141.372193] EXT4-fs (sdc2): couldn't mount RDWR because of unsupported optional features (400)

The next time I rebooted OMV it got stuck at this:
/dev/sdc2 has unsupported feature(s): metadata_csum
e2fsck: Get a newer version of e2fsck!
fsck died with exit status 8

Then something about repairing the file system manually and entering an administrator shell or pressing CONTROL-D to resume boot. Pressing CTRL-D resumes the boot process and OMV boots normally, but the same issue remains: cannot mount /dev/sdc2 from the interface.

I booted from GParted Live again and performed a file system check. It finished OK, but had the same problem at OMV boot. I then went into GParted Live and deleted the partition. OMV booted without errors. I then used SSH with X forwarding to run gparted directly from OMV with the HDD attached, thinking that maybe GParted Live uses a newer version with unsupported EXT4 features.

I created the new partition again and got an error about not being able to update the kernel and then the gparted interface got stuck. I had to kill it, but the new partition did appear in the OMV interface. However, the same problem as before: cannot mount it, same error. Rebooted OMV and it's stuck again at fsck with the same error.

The first partition on the drive /dev/sdc1 appears to work properly. It mounts automatically, but still has the old problem: can't unmount it from the interface.

Any ideas? At this point I'm tired of searching all sorts of Linux websites and trying various suggestions, so maybe one of you has encountered this. I just want a second EXT4 partition on this drive! Creating partitions and mounting them is not supposed to be so complicated!

Thanks. Rsync does sound like the easiest option. Is it also the most performant of the three? If I want encryption with LUKS, I guess I'll have to encrypt the partitions separately.

Hi everyone,

I'm new to OMV. I'm using it on a rig that's not that powerful: a FUJITSU SIEMENS ESPRIMO E5730

Intel Core 2 Duo CPU E8500 @ 3.16GHz
2GB DDR2 800Mhz
1 x Gigabit LAN
1 x 80GB WD HDD -- OMV installation
1 x 1TB Samsung HDD -- two 500GB NTFS partitions
1 x 3TB WD HDD -- one ext4 partition

The 3TB WD HDD is an external USB3.0 HDD connected to a USB3-PCI Express card as the motherboard didn't have native USB3.0 chipset.

I'm looking for help/suggestions regarding the best options to achieve the storage configuration that I have in mind:

I want to set up a single 1TB partition on the 1TB Samsung HDD that will be replicated on a 1TB partition on the 3TB WD HDD for redundancy. This partition will hold important data, like some of my old files that I've backed up over the last few years and are currently on the 2 500GB partitions. There are many files, typically small, that won't change very often. I'd also like to use this new partition for backup from my laptop with crashplan and mobile devices with maybe OwnCloud or BTSync. Ideally, this partition would be encrypted. The 3TB hard drive will have a remaining 2TB partition for media, movies/TV episodes which I think will be enough for my needs. I usually delete them after watching.

What will be my best options to achieve this set-up with a replicated 1TB partition across two HDDs? Is RAID possible between one HDD and a partition from another HDD? Is SnapRAID a better option? Maybe just Rsync running during night time to keep the partitions in sync? Also what filesystem would be best?

I performed:

apt-get purge openmediavault-snapraid

and now the interface works again! Might be a conflict with a different plug-in?

Hi,

Thanks for answering. It is weird though that it happened right after the plug-in install process, when it said something like the page needs to reload. Here is the output:

ii openmediavault 2.2.4 all Open network attached storage solution
ii openmediavault-keyring 0.4 all GnuPG archive keys of the OpenMediaVault archive
ii openmediavault-letsencrypt 2.4 all Generate free and valid SSL certificates for OMV
ii openmediavault-omvextrasorg 2.13.2 all OMV-Extras.org Package Repositories for OpenMediaVault
ii openmediavault-plexmediaserver 1.0.14 all OpenMediaVault Plex Media Server plugin
ii openmediavault-snapraid 1.5 all snapraid plugin for OpenMediaVault.
ii openmediavault-transmissionbt 1.4 all OpenMediaVault Transmission (BitTorrent client) plugin.

Hello,

First off, I'm new to OMV. I have some experience with Linux and the command line, but I wouldn't consider myself an advanced user. My problem is as follows:

I'm running OMV 2.2.4 with the OMV-Extras repository installed. Today I installed the SnapRAID plug-in from the repository through the web interface. The process said Done, but when it attempted to reload the Web page, it came back empty. I though it might be a browser cache issue, so I tried in an Incognito tab. I get to the login screen, fill in the correct username and password, then the page stays blanks instead of showing the interface. I rebooted OMV from the command line. Same issue. It's like the SnapRAID plug-in broke something in the interface. I had no problems before installing it and I have other plug-ins that I've installed and they worked OK, like Plex, BitTorrent.

Any idea what might have caused this and how can I fix it? Is there a way to disable/uninstall this plug-in from the command line?

I hope that I'm posting in the right thread. Thanks.