Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    phibid
    @phibid
    Ah, I in fact already opened an issue for this exact same issue last year, and could not remember it ... uyuni-project/uyuni#2115
    zeromq v4.1 is offered by EPEL during the build of the bootstrap repo while Uyuni client needs zeromq v4.0
    ngo3
    @ngo3:matrix.org
    [m]
    Hello ! No one for my question .. :(
    ppc6446
    @ppc6446
    @lhw-sa , on the /var/lib/pgsql size, you may want to check if there are two copies from a postgresql12 to 13 migration, for example. If so, you could remove the old data
    speaking of which, I deleted the old 12 data on mine a while back and just finished with a package cleanup: # zypper remove postgresql12-12.7-8.20.1.x86_64 postgresql12-server-12.7-8.20.1.x86_64 postgresql12-contrib-12.7-8.20.1.x86_64

    For the python-zmq issues, I have documented EL7 and EL8 exclusion examples in this issue: uyuni-project/uyuni#3899

    That was a request to add an exclusion by default when adding EPEL using the uyuni scripts since presumably every user using EL7 or 8 would trip over this

    phibid
    @phibid
    @ppc6446 Ooooooh
    uyuni:/var/lib/pgsql # pwd
    /var/lib/pgsql
    uyuni:/var/lib/pgsql # du -sh *
    4.0K    analyze_new_cluster.sh
    36G     data
    38G     data-pg10
    3.0G    data-pg12
    4.0K    delete_old_cluster.sh
    4.0K    initlog
    !!!
    phibid
    @phibid
    # zypper remove postgresql10-10.17-8.35.1.x86_64 postgresql10-server-10.17-8.35.1.x86_64 postgresql10-contrib-10.17-8.35.1.x86_64
    # zypper remove postgresql12-llvmjit-12.7-8.20.1.x86_64 postgresql12-12.7-8.20.1.x86_64 postgresql12-server-12.7-8.20.1.x86_64 postgresql12-contrib-12.7-8.20.1.x86_64
    uyuni:/var/lib/pgsql # rm -rf data-pg10 data-pg12
    Better now :-)
    ppc6446
    @ppc6446
    nice!
    digdilem
    @digdilem
    ^ turns out I had an old data-pg12 taking up 13G that wasn't being used, as well. Surprised the upgrade process doesn't automatically tidy those up?
    phibid
    @phibid

    @digdilem There is this script:

    uyuni:~ # ls -lrt /var/lib/pgsql/delete_old_cluster.sh
    -rwx------ 1 postgres postgres 45 Jul 20 08:18 /var/lib/pgsql/delete_old_cluster.sh
    uyuni:~ # cat /var/lib/pgsql/delete_old_cluster.sh
    #!/bin/sh
    
    rm -rf '/var/lib/pgsql/data-pg12'

    But not sure when it is executed. Was not in my case though

    ppc6446
    @ppc6446
    It's intentionally left there as a precaution as part of the normal postgresql upgrade procedure from upstream of that project in case there's a problem with the new postgresql version to allow rollback
    Uyuni actually makes it easier than most to handle the postgresql version upgrades. I have to manually perform the postgresql upgrade for Icinga2
    I seem to remember an in-place upgrade option with SUSE Manager a while back, though
    digdilem
    @digdilem
    Yes, the upgrade process is smooth. And I kinda see the logic of leaving the old one behind - but I wonder if that needs to be forever? These things can be quite big.
    Although - I can't think of a good way of handling it if it is to be left. Scheduled delete is no safer. Prompting user in Uyuni itself might work.
    ppc6446
    @ppc6446
    Yeah, if most everyone in the channel unknowingly had old backups eating up space, it's probably something worth addressing. I kind of like the in-place upgrade, personally, but I tend to have backups before starting
    phibid
    @phibid
    Perhaps as a start, add in the upgrade documentation procedure that the old pgsql datadir can be further removed by hand to free space. If I am not wrong it does not appears anywhere.
    digdilem
    @digdilem
    That's a good idea. I don't recall reading about this before, and I was unaware of it. I don't use postgres often so didn't recognise the significance of the extra dir.
    phibid
    @phibid

    After a salt-minion registration of a SLE 15 server, I have the following behaviour while trying to install a package:

    Looking for gpg key ID 0D20833E in cache /var/cache/zypp/pubkeys.
    Repository Uyuni Client Tools for SLES15 SP3 x86_64 does not define additional 'gpgkey=' URLs.
    python3-pyvmomi-6.7.3-2.1.uyuni.noarch (Uyuni Client Tools for SLES15 SP3 x86_64): Signature verification failed [4-Signatures public key is not available]
    Abort, retry, ignore? [a/r/i] (a):

    Shouldn't the public key be automatically imported to the client while bootstrapping ?

    Stefan Bogner
    @bochi
    hmm, under Software -> Channel List I can see all channels for sles15,sp1,sp2,sp3
    vailen
    @vailenwallaces

    We got an error of our salt-minions on SLES12 SP5:

    [salt.loader :1675][ERROR ][10477] Failed to import grains virt, this is due most likely to a syntax error:
    Traceback (most recent call last):
    File "/usr/lib/python2.7/site-packages/salt/loader.py", line 1658, in _load_module
    mod = imp.load_module(modnamespace, fn, fpath, desc)
    File "/var/cache/salt/minion/extmods/grains/virt.py", line 12, in <module>
    except ModuleNotFoundError:
    NameError: name 'ModuleNotFoundError' is not defined

    5 replies
    Stefan Bogner
    @bochi
    when I click on a sles15 system and go to alter channel subscriptions, I do not see any SUSE channels to change to
    i do see all my custom base channels though, centos etc
    also, i can change the basechannel using spacecmd to sles15 sp3
    why aren't they shown in the UI though? am I missing anything?
    phibid
    @phibid
    @bochi Had the same issue after updating Uyuni to 2021.08. I had to clear the browser cache, this solved the issue
    Stefan Bogner
    @bochi
    nope, tried that
    also product migration tells me "no successor could be found"
    even though everything is there for sure
    no successor to basesystem, legacy, server app + suma client tools
    Cédric Bosdonnat
    @cbosdonnat:matrix.org
    [m]
    just need to catch ImportError as well
    Cédric Bosdonnat
    @cbosdonnat:matrix.org
    [m]
    tanks
    Stefan Bogner
    @bochi
    narf, after upgrading to 2021.08 i do not have the navigation pane anymore
    its just blue and empty
    any idea?
    Stefan Bogner
    @bochi
    browser cache
    for whatever reason :)
    digdilem
    @digdilem
    Yah, CTRL+F5 after an upgrade. Caught me out before too!
    ngo3
    @ngo3:matrix.org
    [m]
    Hello, someone can tell me how i can stop a repo sync on CLI or on web GUI ?
    ngo3
    @ngo3:matrix.org
    [m]
    Its ok, i kill it
    StefanSa
    @StefanSa
    Hi there,
    is it possible to replace the Apache certificate without major problems, or is there a problem with the clients and an existing proxy ?
    Cédric Bosdonnat
    @cbosdonnat:matrix.org
    [m]
    @StefanSa: if you replace the server certificate with one signed by the same CA there shouldn't be any problem. However if the CA changed, then you need to propagate it to all clients and proxies
    heiwu
    @heiwu
    Hi all! i was asked to patch our SLES15 SP2 systems to SP3. Currently they are connected to SUSE RMT. Do you think that is a good opportunity to also migrate them into our uyuni?
    • Pros:
      • I don't have to touch every system twice
      • am able to do any post-upgrade task faster and more easy as i have "the power of salt"
    • Con: Double risk for incidents
      Or would you separate that change? -> why?
      I think the biggest risk lies within SP Upgrade, am i right? Is there any "compatibility check" / dry run i can do on target systems to see how they would be affected by the Upgrade before i run it?
    1 reply
    heiwu
    @heiwu
    ... and another (shorter) question: With SUSE RMT a SLES System is registered and controls its channels by using SUSEConnect. In Uyuni and SUSE Manager this is not the case, everything is controlled from within Uyuni/SuMa, right?
    1 reply
    ngo3
    @ngo3:matrix.org
    [m]

    Hello, when i try to add rhel tools channel i have an error. Any idea ?

    spacewalk-common-channels rhel7-pool-x86_64 res7-suse-manager-tools-x86_64
    SUSE Manager username: admin
    SUSE Manager password:
    No channels matching your selection.

    1 reply
    ngo3
    @ngo3:matrix.org
    [m]

    :point_up: Edit: Hello, when i try to add rhel tools channel i have "No channels matching your selection". Any idea ?

    spacewalk-common-channels rhel7-pool-x86_64 res7-suse-manager-tools-x86_64
    SUSE Manager username: admin
    SUSE Manager password:
    No channels matching your selection.