Beiträge von auanasgheps

  • SMART alerts are not sent

    Hello,

    I recently discovered that I have a drive with bad sectors, but not via OMV SMART monitoring email.

    Whilst I deal with the hard drive issue, I'd like to know why I'm not receiving a daily notification for this drive.


    As you can see, the drive is recognized as bad.




    This has been going on for about a week, but I still have to receive an email about this.

    Email notifications are working great for other sections like Login alert, cron jobs and so on.

    You can find attached my SMART config. Have I done anything wrong?

    In System > Notifications > Notifications, SMART is enabled.

    I also tried disabling and re-enabling SMART for the drive, no difference. I don't see any mail error in the logs, meaning the event is not triggered or some sort.


    I've also checked syslog events and I don't see errors for this drive. Here's the log from a couple of days ago.

  • Wakealarm plugin does not work

    We found the fix!


    It's super simple: Debian 11 does not have anymore python2 wich this file is trying to use.


    Will create a PR once I finish testing.


    EDIT

    Claimed victory too soon.

    By changing python to python3 the script runs fine but the system wakes up just after few seconds, even if the time in sys/class/rtc/rtc0/wakealarm is now populated and looks correct. What else could be wrong?


    EDIT2: Spontaneous wake up seems unrelated.
    Will come back after further testing.

  • Wakealarm plugin does not work

    Tried to run the commands on the system

    and here we go with a wonderful error!


    Code
    root@nas:~# test -x /usr/sbin/wakealarm && /usr/sbin/wakealarm
/usr/bin/env: 'python': No such file or directory

    On OMV this command works fine with no output.

  • Wakealarm plugin does not work

    The UI is set to English, because translations are usually misleading.


    I see the entry in the logs every minute. It's


    Code
    CRON[21653]: (root) CMD ( test -x /usr/sbin/wakealarm && /usr/sbin/wakealarm > /dev/null 2>&1)

  • Wakealarm plugin does not work

    Hello!


    I installed OMV6 on a friend's computer and we noticed the wakealarm plugin does not work.


    Versions

    OMV 6.0.7-1

    Wakealarm plugin 6.0


    I'm sure it's related to the plugin because native wakealarms command work. In fact, as a workaround we have set this scheduled job rtcwake -m off --date 7:30 which turns off the system when executed and wakes it up at 7:30.


    I have noticed that /sys/class/rtc/rtc0/wakealarm is empty, whilst on my OMV5 system is populated with the epoch time of the next wakeup.


    In /etc/wakealarm.conf we can see the test wakealarm that was configured via the webgui.


    However, in/proc/driver/rtc, the next wakealarm is the one set with the manual command, not the plugin.


    Something is definitely not working on this system with this plugin.


    Is there anything else I could check?


  • Nextcloud download of files larger than 1GB fails

    I know it's not related to your specific issue because you're using another container, but I'll also add how to fix this problem when using letsencrypt/nextcloud in conjunction with letsencrypt/swag as reverse proxy.


    Nextcloud documentations says

    Code
    By default, downloads will be limited to 1GB due to proxy_buffering and proxy_max_temp_file_size on the frontend.

    Therefore, simply add this configuration to the nextcloud.conf file from swag. You can add it just before the end.

    proxy_buffering off;


    And the magic will happen!

    I had this issue for a very long time and only few months ago I found the way to solve it.


    Nextcloud is a great piece of software but IMHO is very hard to master and to run properly. I invested more hours than I'd like to admit to make NC running "fine".

  • Unable to spin-down my disks, who is reading them?

    Hello there!

    If you followed my guide, double check what version got installed. The author's repo doesn't still have hd-idle for Debian 11 (requested yesterday) so you'll have to install it manually, otherwise the old and buggy hd-idle got installed from Debian's repo.


    1. Where is docker binary installed? The answer should be: a separate SSD, otherwise you have a problem.

    2. Where applications/services/dockers are installed? The answer should be: a separate SSD, otherwise you have a problem.

    3. Do these applications access your HDDs? The answer is: probably yes.

    3a. If the above answer is yes, do these applications access your HDD based on a schedule? E.g. Plex and Syncthing do this, but they can be adjusted not to do so.


    Another approach:

    Install fatrace

    Run in shell this command

    nohup fatrace -t -o/path/to/ssd/fatrace.log &


    This command will track systemwide all I-O access with process name, date and time. /path/to/ssd is going to be a path on your system or ssd drive (NOT HDD) to save the log file.

    After running the command, you can close the SSH session, it will survive in the background until you kill the process or turn off the server.

    Let it work for a couple of hours and go back to the logs. You will find what has accessed your drive, but there will be a lot of things so be patient and filter using your Hard Drives UUID or mergerfs UUID.

    Also note that logging in the WebGUI (at least OMV5) wakes instantly all of your drives.

  • Not receiving SMART alerts

    I can see another issue: smartd stops logging after some time.

    My system turns on at 11AM, goes to sleep at 1:30PM; wakes up again at 6PM and goes to sleep just after midnight.

    Looks like smartd is not "recovering" after the sleep.

  • Not receiving SMART alerts

    Hello,

    I recently discovered that I have a drive with bad sectors, but not via OMV SMART monitoring email.

    Whilst I deal with the hard drive issue, I'd like to know why I'm not receiving a daily notification for this drive.


    As you can see, the drive is recognized as bad.




    This has been going on for about a week, but I still have to receive an email about this.

    Email notifications are working great for other sections like Login alert, cron jobs and so on.

    You can find attached my SMART config. Have I done anything wrong?

    In System > Notifications > Notifications, SMART is enabled.


    I've also checked syslog events and I don't see errors for this drive. Here's the log from a couple of days ago.

  • Finally: my custom background image back

    This procedure unfortunately doesn't work with OMV6.

    I thought so because the whole UI has changed, but I still tried to apply the procedure.


    On OMV6, simply browse to /var/www/openmediavault/assets/images and replace benjamin-lehman-GNyjCePVRs8-unsplash.jpg with another picture using the same name, since the config is pointing to that file. I still have to find the config file so any name can be used, but will not change the end result.