You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When I delete a user from LDAP which isn't in LDAP anymore with occ user:delete <user> it should only take a couple of seconds.
Actual behaviour
When jsxc.nextcloud is active, deleting such a user takes several minutes.
These are the SQL queries that the SQL server is receiving during that time:
SELECT "appid", "configkey", "configvalue" FROM "oc_preferences" WHERE "userid" = $1
SELECT "ldap_dn" FROM "oc_ldap_user_mapping" WHERE "owncloud_name" = $1
INSERT INTO "oc_ojsxc_stanzas" ("to", "from", "stanza") VALUES($1,$2,$3)
@DanScharon thanks for reporting, it's a bit silly I didn't think of this earlier.
can we safely disable all hooks if the internal backend is not used?
Yes 😃
why is there a roster push for all users? It should be enough for users who are online.
Because when a user is offline (i.e. there is no browser window open) the roster is still cached in the localStorage of users, when they open JSXC again the roster is loaded from localStorage.
jsxc.nextcloud: 3.3.2
Expected behaviour
When I delete a user from LDAP which isn't in LDAP anymore with
occ user:delete <user>
it should only take a couple of seconds.Actual behaviour
When jsxc.nextcloud is active, deleting such a user takes several minutes.
These are the SQL queries that the SQL server is receiving during that time:
It looks like although I am using an external XMPP server, JSXC is trying to keep the rosters for its internal server up to date after a user removal.
As a result it took up to 9 minutes to delete a user from (an admittedly large) LDAP.
The text was updated successfully, but these errors were encountered: