Beiträge von hakarune

    In their defense, you are using a version of OMV that was EOL'd over two years ago. And they know from experience that the 4 to 5 then 5 to 6 upgrade can be challenging even with 15 years of Linux experience and that a reinstall is faster. If all of your data is on data drives like OMV designed, then the reinstall is faster. Has nothing to do with being a throwaway system. I reinstall RHEL8 systems in the enterprise instead of upgrading from RHEL7 because the upgrade is a pain in the ass.

    I'm on 5, it was a fresh install on new SSD that I did because I was told upgrading to 5 from 4 caused issues and no one would really be able to help troubleshoot issues from it. So I took time to manually track everything I had installed (all programs like python, my docker config and ymls, custom vi themes, aliases, nuts ups, custom simlinks, smarthome bluetooth presence program, etc) and wrote it all down and then also upgraded at the time to an AMD quad core ITX system with 16GB ram.


    Upgraded my system last year again without issue.


    Now my system has a lot more custom setups and programs installed alongside omv so wiping out my entire system for the mounting and formatting gui isn't an option at this time. If it's not fixable I can just mkfs/mkfs.xfs and mount with fstab editing manually.


    Recently (a few months ago) added a backup drives through the gui when my raid card was faulting to backup my 7TB of stuff to switch raid controllers and then put everything back. Then removed it without issue.


    Now my 3TB secondary drive is starting to go, so I'm trying to add 2 drives to take on the work of the 3TB until I can buy a new and that is when this issue started happening.


    I will do this when I'm back on my main workstation and home to manually reset after it locks up.

    Oh! Don’t be mean! He might do it. =O

    I expect the insults from the guy who can't read the reason for the post, but I expect more from a moderator on this forum. I didn't come here for keyboard assholes I came here because an obvious error in OMV system that needs a solution that isn't, just wipe and reinstall like it's some throw away system and I'm done moron that hasn't been using Linux for 15 years.

    Fix your signature:

    Can you post your /etc/fstab and output of omv-showkey mntent

    Does it happten as well if you try to make other changes?

    I can't even see that I have a signature on mobile. Didn't know if was so or if date.




    OMV4, the underlying Debian version and OMV5 are end of life. Current version is OMV6. I would do a fresh install.


    BTW you can attach pictures using the Attachments function below the text box of a new reply.

    OMV 5.6.20-1

    Debian Buster


    sde2 & sdd1 I can't mount or remove

    Any time I try add/mount or format a new drive from the gui everything crashes and has to be rebooted. Same when I try to delete a drive via the gui. I have screenshots but there's no upload button here...


    Here's the errors I get:


    Just want to say this was a big D move on the devs part. I have used for years when I was on OMV 2.x. When I redid the entire server around 3.5ish time and got in to using docker and other programs I had everything linked through the sharedfolders area and have continued to do so with each new update/upgrade. I recently update and rebooted and it wiped the sharesfolders binding and I spent over 2 hours trying to fix the issue thinking one of 8TB raid drives died at the same time as my backup pool or if some how the update/upgrade borked my entire system. I finally found this thread detailing the change you guys did without any notice to end users at all. I'm rarely on the forums and don't see these tiny threads with discussions that wipe out my entire setup and cause a panic attack and stress.


    The fact that such a substantial change was made and will effects end users without notice is a very messed up way to go about this.

    I've just tried the olbat docker on my VM and it installs and works as expected, whilst my printer is wireless anyway it actually detects it.

    I can add printers to the docker, but they are not broadcasted or accessible by any device on the network. I read somewhere that there might need to be SMB configuration for the printers to appear, but I can't find the options to configure that.

    try create dockers in host mode.

    Tried that, tried host with privileged mode as well.


    So far now after 3 hours more tonight working with it, I finally got the server itself to see the printers, but no computer (Mac, Windows, Linux) on the network or Android/iOS device can find the printers; I even tried forcing the print search to the server IP/IP:PORT and got nothing....

    I have tried multiple Docker CUPs images and after starting it up and adding printers, no device on the network can see the shared printers. Also, there is no persistency on the docker images and I loose all my settings and added printers on reboots, recreates, etc.


    How can I get CUPs working on OMV?

    Yeah, I have no idea what's going on and everything is set to 777 on the folders even and still nothing. That's why I can only assume it's a docker omv issue. I'm giving up on trying to use OMV and going to just convert my pi zero W to a LAMP server for testing PHP.



    As for the deleted, it wasn't working through the GUI either and took a bit, but eventually fixed it (though I posted after trying for a while unsuccessfully) then found a way and deleted the post.

    Okay, installing it now.



    Here are the 3 folders I'm testing with the docker on 3 separate drives (an external raid 2x2.5" HDDs, External Drive - 3.5" HDD, external m.2 SSD)

    Code
    root@omvnas:/srv/dev-disk-by-label-extended/web/apache/www# stat html/
    File: html/
    Size: 43              Blocks: 0          IO Block: 4096   directory
    Device: 801h/2049d      Inode: 51216076    Links: 2
    Access: (2777/drwxrwsrwx)  Uid: ( 1000/hakarune)   Gid: (  100/   users)
    Access: 2020-08-18 23:39:59.333785209 -0700
    Modify: 2020-08-18 23:39:11.664686720 -0700
    Change: 2020-08-18 23:39:59.309784655 -0700
    Birth: -
    Code
    root@omvnas:/srv/dev-disk-by-label-External/apache/www# stat html
    File: html
    Size: 31              Blocks: 8          IO Block: 4096   directory
    Device: 811h/2065d      Inode: 4472780167  Links: 2
    Access: (2777/drwxrwsrwx)  Uid: (   33/www-data)   Gid: (  100/   users)
    Access: 2020-08-18 04:51:14.463048068 -0700
    Modify: 2020-08-13 02:43:03.343514542 -0700
    Change: 2020-08-13 02:45:03.770344591 -0700
    Birth: -
    Code
    root@omvnas:/srv/dev-disk-by-label-Raid/Programs/apache# stat public/
    File: public/
    Size: 0               Blocks: 0          IO Block: 4096   directory
    Device: 822h/2082d      Inode: 1105453     Links: 1
    Access: (0755/drwxr-xr-x)  Uid: ( 1000/hakarune)   Gid: (  100/   users)
    Access: 2020-08-13 01:44:01.813786700 -0700
    Modify: 2020-08-09 00:18:58.046778900 -0700
    Change: 2020-08-13 01:44:01.217174800 -0700
    Birth: -



    I've also attached 3 images of how the docker is setup in the GUI (it's the default from the build in this forum).

    Hi,


    I’m curious what services are you guys use Apache for ?

    ? I'm trying to use it for specific software, in this instance Sentrifugo, which does not work well with Ngnix and the hacks for it require a newer PHP on Ngnix to use, but it is still a bit of a hack job workaround that isn't clean or smooth. There are also some other web software and programing code that doesn't work well with Ngnix but work with no issue on apache.

    Use the resetperms Plugin. That also resets ACLs.


    Greetings

    David

    will this reset anything beyond ACLs? I have a bunch of chmod foldes and setup with my internal server for things like calibreweb, emby, kodi, etc. Don't want to destroy all the work I've put in to this thing.


    Also, I created a new directory on the root of all 3 HDD and created a docker with this built image and set the /var/www/html to a fresh public folder and all 3 failed the exact same way. All 3 have something public viewing web interface already and work fine, it's just apache docker that seems to be failing non-stop.

    • HDD1 is a the default nginx web server with many internal websites and testing sites
    • HDD2 is the CalibreWeb portal and a couple other portals
    • HDD3 is the Emby, Kodi, Mylar holder

    Can you think of reason why it would fail on all 3 the same way? I created new folders, named public, via cli: chmod 777; chown hakarune:users; chmod a+x. And the error is the same for all. I even checked ACL/perms via the gui and they were greyed/blank for all but the normal default ngnix one that you suggested I wipe of ACL.


    Bash
    AH00558: apache2: Could not reliably determine the server's fully qualified domain name, using 172.17.0.4. Set the 'ServerName' directive globally to suppress this message
    AH00558: apache2: Could not reliably determine the server's fully qualified domain name, using 172.17.0.4. Set the 'ServerName' directive globally to suppress this message
    [Thu Aug 13 01:52:26.478670 2020] [mpm_prefork:notice] [pid 1] AH00163: Apache/2.4.25 (Debian) PHP/7.3.21 configured -- resuming normal operations
    [Thu Aug 13 01:52:26.478815 2020] [core:notice] [pid 1] AH00094: Command line: 'apache2 -D FOREGROUND'
    [Thu Aug 13 01:52:40.626552 2020] [core:error] [pid 16] (13)Permission denied: [client 10.1.1.2:4371] AH00035: access to / denied (filesystem path '/var') because search permissions are missing on a component of the path
    10.1.1.2 - - [13/Aug/2020:01:52:40 -0700] "GET / HTTP/1.1" 403 493 "-" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.125 Safari/537.36"
    10.1.1.2 - - [13/Aug/2020:01:52:40 -0700] "GET /favicon.ico HTTP/1.1" 403 492 "http://10.1.1.10:7080/" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.125 Safari/537.36"
    [Thu Aug 13 01:52:41.026613 2020] [core:error] [pid 16] (13)Permission denied: [client 10.1.1.2:4371] AH00035: access to /favicon.ico denied (filesystem path '/var') because search permissions are missing on a component of the path, referer: http://10.1.1.10:7080/
    [Thu Aug 13 01:52:41.602089 2020] [core:error] [pid 16] (13)Permission denied: [client 10.1.1.2:4371] AH00035: access to / denied (filesystem path '/var') because search permissions are missing on a component of the path
    10.1.1.2 - - [13/Aug/2020:01:52:41 -0700] "GET / HTTP/1.1" 403 492 "-" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.125 Safari/537.36"

    Yes it's running as PGID 1000 (users) and PUID 100 (my default account, hakarune). I also tried running with elevated perms, run as default and run as root; still nothing.


    The folders are set as 777 (you can see the html folder is listed as such too). I've also run chmod -R a+x on the root folder and again on html folder. I didn't use ACL to set permissions for folders/volumes, attached the image so you could see everyone can rw. If there's a way to wipe ACL settings let me know because I don't recall ever setting them up and always use SSH for setting perms.


    I went through many stack Qs as well before posting on here because I'm stumped.