Re: [rep:User] autosave when bulk creating users

2015-09-15 Thread Nicolas Peltier
Hi Angela, this mail just to say i was wrong at my first analysis to incriminate persistence, looks like i was using isMember before adding a user to a group which is #1 useless, and #2 more & more expensive as i was adding users to group. Nicolas > On 06 Aug 2015, at 10:56, Angela Schreiber

Re: [rep:User] autosave when bulk creating users

2015-08-06 Thread Nicolas Peltier
i am not too familiar with the sync mechanism... but looking just quickly at the code it seems that it persists the sync of each user/group. that looks reasonable to me (even if it comes with some Root.commit() overhead) as it allows to specifically retry or revert the changes made during

Re: [rep:User] autosave when bulk creating users

2015-08-05 Thread Nicolas Peltier
whenever you create or modify a user/group using regular user management API the changes are not autosaved by default and it's up to the caller to persist the changes by calling Session.save(). i didn’t know that, great. or do you specifically mean the creation of a user in the context of

[rep:User] autosave when bulk creating users

2015-07-31 Thread Nicolas Peltier
Hi, i’ve seen that UserManager.autosave was an unsupported operation, and thus using LDAP sync with many users generates a save at each user/group creation which makes the sync very slow. Installing a package of users is much quicker, so i’m wondering if there is a way to create user with an

Re: [defaultsynchandler] make default sync handler more customisable (Was: [ldap sync] prepopulate a set of users)

2015-06-15 Thread Nicolas Peltier
the DefaultSyncContext. let me know if you need more override points. regards, toby On Fri, Jun 12, 2015 at 11:48 AM, Tobias Bocanegra tri...@apache.org wrote: Hi, On Mon, Jun 8, 2015 at 9:06 AM, Nicolas Peltier npelt...@adobe.com wrote: Hi Tobias, as discussed off list, this defaultsynchandler

[defaultsynchandler] make default sync handler more customisable (Was: [ldap sync] prepopulate a set of users)

2015-06-08 Thread Nicolas Peltier
after and before each sync, and may be also after the initial configuration) wouldn’t do the trick, wdyt? e.g. [0] could be fixed by adding a component checking the user attribute Nicolas [0] https://issues.apache.org/jira/browse/OAK-2948 On 15 May 2015, at 16:16, Nicolas Peltier npelt

[osgi][auth-external] strange osgi wiring with felix bnd plugin

2015-05-15 Thread Nicolas Peltier
Hi, after trying to add ExternalIdentityProvider’s dependency [0] to my project, i get a strange import instruction generated in the built bundle [1] leading to bundle in Installed state, as [0]’s export is version 1.0.0 i fixed this issue by adding explicitally

[ldap] paged search in LdapIdentityProvider.listUsers

2015-05-15 Thread Nicolas Peltier
Hi, using LdapIdentityProvider.listUsers i get results limited to 1000 results, i guess this is a server limitation as the total number is 10k. Also default timeout is 60s, and i guess doing multiple paged requests would be better. Is this something i should handle at project level or would

Re: [ldap sync] prepopulate a set of users

2015-05-15 Thread Nicolas Peltier
from your code. Regards, Toby [1] https://github.com/apache/jackrabbit-oak/blob/jackrabbit-oak-1.2.2/oak-auth-external/src/main/java/org/apache/jackrabbit/oak/spi/security/authentication/external/impl/jmx/SyncMBeanImpl.java#L179 On Wed, May 6, 2015 at 2:13 AM, Nicolas Peltier npelt

Re: [ldap sync] prepopulate a set of users

2015-05-06 Thread Nicolas Peltier
To be clearer, said in a different way, i’d like to pull a given set of users (LDAP query) from a LDAP server in a scheduled manner, using at most the oak code :-) Nicolas On 06 May 2015, at 11:09, Nicolas Peltier npelt...@adobe.com wrote: Hi, i’m implementing an application where users

Re: reindex improvements

2014-08-26 Thread Nicolas Peltier
Hi Davide, this would be nice indeed! wouldn’t that be “indexPath”, not “re-indexPath” ? Nicolas On 26 Aug 2014, at 12:04, Davide Giannella dav...@apache.org wrote: Hello team, when we issue the reindex by changing the index definition with `reindex=true` the algorithm scan all the