Posts by openletter

    On one of the subdirectories, I deleted the contents with rm -rf then ran the rsnapshot job through the GUI, and no problems reported.


    On another subdirectory, I set monthly to 1, then ran the job through the GUI and nothing was deleted. I then ran the above command for monthly and nothing was deleted. I then deleted the contents with rm -rf but I'm going to let the job run on schedule, see what happens (it should be fine, but just to test).

    Dumb guy question here:


    I want to start over with some rsnapshot jobs just because I accidentally backed up a lot of data that doesn't need backing up, so my simple thinking is just delete everything within the target directory (e.g., rm -rf targetdirectory/*)and the next time the job runs it will effectively be starting from scratch, which is fine.


    However, I'm not familiar enough with rsnapshot to know if this ends up causing a problem, so I thought I'd post here, first.

    I just completed a migration from OMV 5 to OMV 7 as a fresh install, and it's great.


    However, an rsync job I've been running for ages is failing, and it's been so long since I configured it, I don't know if I've somehow skipped a step and as best I can tell, OMV can't verify the Host key and I don't know where this is configured in the dashboard.


    Here is the error message when manually running from the dashboard:

    Code
    Please wait, syncing <remoteusername@example.com:/backup/source/directory/path/> to </srv/dev-disk-by-uuid-XXX/backup/destination/directory/path/>
    
    Host key verification failed.
    rsync: connection unexpectedly closed (0 bytes received so far) [Receiver]
    rsync error: unexplained error (code 255) at io.c(232) [Receiver=3.2.7]
    ERROR: The synchronisation failed.

    Of course, the first thing I noticed is the British English misspelling of synchronization.


    The source server auth log has only the following message:

    Code
    Aug 27 05:46:42 example sshd[1234657]: Connection closed by 203.0.113.38 port 54321 [preauth]

    I fixed the problem by logging into OMV, creating a root session with sudo -i, adding .ssh/id_rsa.pub to .ssh/authorized_keys for user on the source server, then opening a session to the source server with ssh username@example.com, which included confirming the host fingerprint.


    After this, I realized there was no need to add the public key to the remote server, because the host key would be added to OMV before attempting to log in, and the host key is the only thing I needed.


    It would be nice if something, somewhere in the dashboard alerted me to this. Optionally, the source server happens to use the rare SSHFP DNS record, but probably more work to implement that than a note in the GUI.

    This post showed at the top of my search engine results, hence the necro.


    When all else seems something wrong, try this, with appropriate changes based on system and CPU:


    Code
    cat /sys/class/thermal/thermal_zone*/type
    acpitz
    acpitz
    x86_pkg_temp
    ll /sys/class/thermal/thermal_zone*/type
    -r--r--r-- 1 root root 4.0K Aug 17 08:40 /sys/class/thermal/thermal_zone0/type
    -r--r--r-- 1 root root 4.0K Aug 17 08:40 /sys/class/thermal/thermal_zone1/type
    -r--r--r-- 1 root root 4.0K Aug 17 08:40 /sys/class/thermal/thermal_zone2/type
    cat /sys/class/thermal/thermal_zone2/temp 
    43000

    The number printed is in mC, so the temperature is 43C.

    I was getting similar errors to above post and it took some time to realize I forgot to close out a tmux session that was in a subdirectory of the drive from the last time I had it mounted, so apparently that can keep the drive from being mounted again but not keep the drive from being unmounted. Closing the tmux session allowed for immediate mounting of the drive.

    Although this doesn't help with your immediate problem, a way to deal with this in the future is to create an empty file for deletion should the partition become filled.


    Code
    fallocate -l 250M deleteifpartitionfull

    Note that if some errant process is rapidly filling your partition, this basically starts a countdown for you.

    I looked at Borg, and it isn't clear to me how to create local encrypted backups that can easily be transferred to external storage device.


    I mostly ask about Duplicity because I have at least some experience with it. Hadn't looked at Duplicati before.

    I'm wanting to create encrypted external storage backups of OMV, but I'm not seeing a way to do this conveniently with the available tools in OMV or OMV-Extras.


    Duplicity can create encrypted incremental backups and save them locally, so I'm looking at that and seeing it is in the Debian repos, but it's an older version, so I was wondering if anyone has experience using Duplicity on OMV?