Posts by sonofwatt

    1. What kind of hardware do you have?
    2. Do you run Nextcloud or Plex on it?

    1. I have a Asus server grade 1U enclosure.

    2. No to Nextcloud, although I'm thinking of deploying it. Yes to Plex, but it's running in a docker container.


    This is continuing to happen about once a week, but at random. Nothing for 2 weeks, then twice within the last 16 hours.
    As with last time the last time time the final messages were 2 emails about the 'Monitoring restart' failing. But when I check the service status service nginx status it shows as up and running. Albeit as of the time of the last 'Monitoring alert -- Connection succeeded' email.

    Hey guys, got this error today after the weekly scrub. WARNING! Unexpected file errors!


    The subject of the email was.

    Cron <root@Odin> /var/lib/openmediavault/cron.d/userdefined-db82bbec-3456-4784-ba42-04ed2326a436 | mail -E -s "Cron - Weekly Scrub" -a "From: Cron Daemon <root>" root >/dev/null 2>&1


    I checked in /var/log/snapraid.log and syslog. The log for snapraid was old. Possibly from the last restart. In syslog, I could find the event from the email's subject, but no mention of an error.


    Any ideas?

    Last night before midnight I got 10 email alerts from OMV. They were all about the service nginx failing and succeeding.

    Searched around but couldn't find this issue elsewhere.
    Date: Mon, 23 Mar 2020 23:26:00

    Service: nginx

    Event: Connection succeeded

    Description: connection succeeded to [127.0.0.1]:80 [TCP/IP]


    Date: Mon, 23 Mar 2020 23:25:13

    Service: nginx

    Event: Connection failed

    Description: failed protocol test [HTTP] at [127.0.0.1]:443 [TCP/IP TLS] -- HTTP: Error receiving data -- Connection timed out


    7RQ1CXz.png

    I get the same thing.

    if I keep trying to reconnect and refresh my browser, I can occsionally get connected.

    Sorry I dont have an answer, though

    Yeah that's exactly it.
    Have you checked your logs to see if you're getting the same error?

    Just an update for any future users with this issue.
    I was never able to fix this. The install was fresh, so I just reinstalled from scratch. This time since those old entries in the USG's hosts file were removed, I was able to get the hostname I wanted, and the emails show the names correctly.

    I'm having an issue with Cockpit. 9 times out of 10, I can't log in. Either the Cockpit login screen won't load and I get a Connection Failed screen telling me to check journalctl -u cockpit, or after clicking Log In, nothing happens. I'm using a password manager so I can be certain the credentials are fine.


    When I checked journalctl -u cockpit I see the following output.

    When I can get the interface to load, it usually works fine although can take 10-20 seconds to load. No other services on OMV have this issue, and the server is on a current gen high-end cpu.

    Any ideas?

    Update: I did some digging and my Unifi USG did have an old entry in it's /etc/hosts file with this old hostname linked to the OMV server's IP.
    I deleted the entry, restarted the router service, and the OMV server, but the notification emails from OMV still show this old hostname in the email subject line. There must be somewhere in OMV that's storing this info.



    Anyone with the same issue, this is the thread I followed.

    https://community.ui.com/quest…f2-4268-8078-aba32ff21aee

    Look in /etc/hostname

    In that file it's the new hostname.


    What you have done is correct, but the name (hostname) is still stored in your router, ChristiesiPhone IP is now being used for your OMV, either wait for the router to play catch up or login to the router and change the name.

    That's so weird. That device hasn't been on the network for over 2 years, so I'm not sure waiting will work.
    I've logged into the router, and it's reporting this IP as the new Hostname.


    What's still confusing is that in the email notification it shows both hostnames in the email subject. "[ChristiesiPhone.localdomain] Test message from newhost.localdomain". Is the notification script pulling this hostname from the router, or is the initial host name stuck in Debian somewhere?

    This is a weird one. I noticed when I was installing OMV v5.3.4 that when it got the screen where you choose the hostname and it had already entered "ChristiesiPhone". I'd also noticed a previous time when instilling the v5 beta that it named itself Ubuntu. Anyway, I chose a different hostname for the server and finished the install.
    Just now I finished setting up the SMTP server settings to enable notifications, and the test email read as "[ChristiesiPhone.localdomain] Test message from blah.blah"

    I should mention that I believe this device name was for an iphone on my network years ago, but no longer.

    How do I fully change the system name? Also why would the installer pull that random old name?

    If the OS drive is in read only mode, then something may have happened during the bootup to put it there.

    The weird part is its not happening on boot up. It's at a random time up to days or weeks after the a reboot.


    The kernel does put a filesystem into read-only mode with reason, so you have to check your hardware. What hardware are you using? How is the device containing the root filesystem connected? USB?

    I'm starting to think it must be hardware related. There are a lot of posts by people with similar issues, but no indication if it's the same underlying cause.
    The hardware is an HP ProLiant DL360 G7, and a RaidAge external 4 bay SAS enclosure. The boot drive is in the server, attached to a backplane thats connected to the internal port of the PCIe SAS card.

    Searching on here, there are others that have had this same or similar issue. This is so weird.
    Locked out of OMV Web GUI
    Admin password stops working
    Locked out of webGUI



    In one of the threads they talk about how the OS drive has been put into read-only mode. That's also what's happening to me. And one guess was that the drive data cable could be damaged. I'll take a look at that, but seems like a long shot. At this point I'm out of ideas.


    Does anyone offer paid support for OMV?

    Are you sure that is the OS drive?

    I think so. Here's an image of what I see after running the df -h command. https://i.imgur.com/YarUTg0.jpg


    One thing I wonder about is if the logs are filling up with that error message about the monitor "no|invalid EDID".
    SDA1 has 79GB free, so can't see logs filling that up in a week, but what about one of the tmpfs mounts? Where are the logs stored?

    Hi, I have a fresh install of OMV v4.1.22. Installed it from USB, and verified the download's hash prior to using balenaEtcher. It's been about a week since the install and I can't login to my server via the web GUI. Still have access via CLI.
    Also network shares stop working, so it seems the issue isn't just too many incorrect password attempts.
    Although to be certain, I launched the omv-firstaid utility and selected the option to reset the login counters, it reported no accounts were listed. I then tried to do a web interface password reset, but I got: ERROR: Failed to execute command 'blahblahblabh' with exit code '1': usermod: cannot lock /etc/passwd; try again later.


    This is around the 4th or so install I've done of OMV 4, and all of them have had this issue where after some amount of time, 2 days, a week, 2 weeks, I can't login. I'm nearly at my wits end with this. Any help is welcome and greatly appreciated.