Not sure if this is a bug, or by design, and if by design then I guess I have to live with it.
I have NS authenticate to a Zentyal AD domain controller. Every time there is an update to WebTop, the domain disappears from WebTop, and I have to add the domain again. The settings are not that simple, so I have the settings saved and cut and paste them in, but still it is a bit of a pain, as every time, I can’t login to webtop, then remember that the domain is gone, and the re-add.
NS access to Zentyal via SSSD or via LDAP?
LDAP. Happens automatically when I connect NS using Active Directory in the ‘Accounts provider’ configuration in the web frontend.
@lucag the command
signal-event nethserver-webtop5-update could be useful in this kind of evenience?
Launching this event re-provisioning the account provider on WebTop 5.
It is not clear to me @wbilger why the configuration is lost at each update …
You didn’t edit the configuration directly, did you?
Any edits should be made in /etc/e-smith/template-custom/etc/… or you will experience what you report.
(I habitually forget that to this day)
@planet_jeroen No, I did not edit any configuration directly.
@lucag This has been consistent the last 3 updates, but never before that. I used to be able to do the udpates (but don’t remember how many webtop updates there were), but the last 3 times there was a Webtop update, I no longer had a domain in webtop.
Is there anything I can do to test, I have a test machine that does not have the latest update yet?
In the meantime, if I run
signal-event nethserver-webtop5-update, after the webtop update (when the domain is gone), it will re-add the domain automatically?
I think that should worth the try, @wbilger. Anyway, it’s quite strange that WebTop loose the domain configuration after updates, because it’s quite “normal” update a software…
Yes @wbilger, with this event, the provider account configuration is re-provisioned.
I confirm that it is strange what reported @wbilger
I could not replicate the problem, updating never lost the configuration.
I just tested this on my 2 test servers and could not replicate as well.
Something is obviously corrupt with my production server then. At some point I am going to do a fresh install on it and just port over all the data and hopefully that fixes it.