Search Results

Search results 1-20 of 1,000. There are more results available, please enhance your search parameters.

  • Access the web ui of traefik 8080 see if any errors show for routers and services. Or at least if they are registered.

  • Yes should be possible. Need to setup docker as a provider in traefik. Also for each container running you need to supply them with labels that traefik will read so it can route the front end requests properly. For non docker services you need to maintain a separate configuration file to point where the servers are. As long as the domains are resolving to the host running docker should work. Please read traefik docs, to see examples and configuration options.

  • You will have to use the Debian installer for this. From there you can choose luks+lvm or any other combination. After that install omv from the guides here in the forum or the omv extras install script.

  • yes. But for omv you need to move the port outside of 80. Traefik will be listening on there and depending on the hostname domain it will fwd the request accordingly. For this is recommended you set up traefik in docker.

  • I use plenty of sub domains. For this task i use traefik which acts as a reverse proxy very easy to setup. If you use letsencrypt it can automatically renew the certificates. for this to work in the local network a static dns entry that points to a domain using wildcard. I case this is not possible you have to edit the host file on the computer you’re using to access the server.

  • but again I see no necessity of using sudo in scheduled task.

  • Why sudo and not just plain root ? I can’t see the scheduled task sending you an email and waiting for you to enter password to run the task. Is the command explicitly allowed no password prompt in sudoers?

  • Quote from kekec888: “You should port forward 445 /SMB port in you router ” SMB is a local sharing protocol, doing this is a very bad idea. Use SFTP(wincsp or filezilla can do the job) with public key authentication or use SMB but over a vpn transport link, you will need at your home a very low latency uplink with decent upload speed to work as smooth as if you are on the local network

  • Quote from odinb: “So you are saying to hash out/remove: "UUID=f7e3bc30-c184-4f4d-aa81-b87ffeb6d844 /media/f7e3bc30-c184-4f4d-aa81-b87ffeb6d844 ext4 defaults,nofail,user_xattr,noexec,usrjquota=aquota.user,,jqfmt=vfsv0,acl 0 2" and then reboot, and OMV should pick it and re-add it to fstab? ” yes hash or remove, then reboot . After boot instead of going shared folder go to the filesystem section, the md0 device should be available there but not mounted. Proceed to mount via …

  • There is the base config.xml stored in /usr/share/openmediavault/unittests/data/config.xml Then you have to clear all omv tagged fstab entries, reboot no disks mounted and start again. The raid should be picked automatically at boot and be available to mount in the ui. From there start doing sharedfolders

  • Quote from odinb: “and I have started adding my configuration back ” Quote from odinb: “Think it has to do with my "/etc/openmediavault/config.xml" file that was re-installed? ” In any omv versions (except i think really early ones) this is still not feasible. Keep it as reference, but configuration has to be done manually. Only people with utter knowledge of how omv works internally might be able to perform a "restore" using config.xml, most of the time unsuccessful. Again don't waste your time…

  • Don’t do this. Any change in the interface though the web ui will revert this. Use the solution I recommended you above to prevent nic name change. Omv doesn’t handle the link files.

  • Quote from hp2: “Long story short: without a GPU the systems boots fine, but onboard LAN is not working. ” Do you still have the GPU card? if so stick it in, boot to the OS. As I mentioned before use udev rules, but reading docs seems like is deprecated. So you have to use systemd .link unit to set a fix name. read custom schemes using .link section As i said before nic name shouldn't change but my guess…

  • I would try a reinstall with static ip configuration on the server not the router. Also check the manual of the board in regards of the gpu. My guess is that when you remove the card the nic changes name since is also Pcie based(it shouldn’t, like systemd introduced predictable hw name but rarely works). Reinstall with gpu, use a udev rule to set a fixed name for the nic, then try again without the gpu.

  • How do you know the ip? Did you set static or dhcp?

  • SMART Error, but where?

    subzero79 - - General


    Change the disk as soon as possible.

  • Well you need to fix the rootfs permissions at top level. chmod 755 /

  • What is this ? Arm board?

  • Drop down to terminal as root an run blkid command. It should list al the partitions or disk with filesystem signature. Encrypted format should display something like CRYPTO_luks if is there means the disk is encrypted and in general it doesn’t take long to encrypt. Sure you can’t unlock the disk? The deselection is normal, but you should be able to pick it. Otherwise just unlock it in terminal. cryptsetup luksOpen /dev/sdX sdX-crypt Replace there the x for the drive letter assigned.

  • You need to unlock it after you encrypt it. Try that.