Alfresco discussion and collaboration. Stick around a few hours after asking a question.
Official support for Enterprise subscribers: support.alfresco.com.
Join in the conversation by getting an IRC client and connecting to #alfresco at Freenode. Our you can use the IRC web chat.
More information about the channel is in the wiki.
More help is available in this list of resources.
2019-03-28 07:11:33 GMT <alfresco-discord> <MorganP> How do you usually manage the Content Store for the big repositories with several TB of documents?
2019-03-28 07:11:55 GMT <alfresco-discord> <MorganP> 1 big Content Store? Several small Content Store more or less of the same size?
2019-03-28 07:30:37 GMT <alfresco-discord> <MorganP> Maybe using the "org.alfresco.repo.content.filestore.VolumeAwareContentUrlProvider" to have a good distribution as mentioned here: http://docs.alfresco.com/6.1/concepts/cs-manage.html. Someone already tried that? I don't see what these "volume1" and "volume2" are, they aren't defined anywhere
2019-03-28 08:29:34 GMT <AFaust> MorganP: Or simply just using an AggregatingContentStore, especially when your content has grown over years and the typical solution was to incrementally add volumes when the existing volumes became full.
2019-03-28 08:31:09 GMT <AFaust> The VolumeAwareContentUrlProvider is only really useful when you already have multiple volumes mounted into the contentstore directory as top-level folders from the start, and all are still empty / not close to being full.
2019-03-28 08:31:55 GMT <AFaust> Since the distribution is random without checking for available space, you may even run into "disk space full" issues when the volumes are of different sizes or you get a significantly skewed file size distribution.
2019-03-28 08:33:23 GMT <AFaust> The "volume1" and "volume2" would have to be the top-level folders representing the volume mounts by name.
2019-03-28 08:34:40 GMT <alfresco-discord> <MorganP> The problem with the aggregatingContentStore in this case is that the customer doesn't have any control over disk speed
2019-03-28 08:34:44 GMT <AFaust> For a proper, non-random, near-to-equal distribution of content among many volumes, I would probably create my own routing content store instead of relying on a content URL provider
2019-03-28 08:34:47 GMT <alfresco-discord> <MorganP> so they will all have random speeds
2019-03-28 08:35:05 GMT <alfresco-discord> <MorganP> so it kinds of loose the advantage of this from my point of view
2019-03-28 08:40:24 GMT <AFaust> Since you never know how read/writes are distributed, speeds are always going to be somewhat "random" from the user's perspective...
2019-03-28 08:42:11 GMT <AFaust> And I mentioned the aggregating content store primarily in the context of how big content stores typically evolve. I have never seen a customer start with a 10 TiB storage spread across multiple volumes from the beginning, when they are growing by 500 or 750 GiB per year.
2019-03-28 08:44:05 GMT <AFaust> So they may start with a 2 TiB volume, then add another after 1.5 years... and use the aggregating content store to write to the new (empty) volume will serving old content from the previous one. Sometimes they may do some content re-balancing on the low level file system...
2019-03-28 08:44:59 GMT <AFaust> Important to note: With VolumeAwareContentUrlProvider you CANNOT do any rebalancing on the storage tier, as the URL contains the name of the volume.
2019-03-28 08:46:17 GMT <alfresco-discord> <MorganP> With the aggregating, you can move the files around in the FS, right?
2019-03-28 08:46:23 GMT <AFaust> Yes
2019-03-28 08:47:36 GMT <AFaust> We actually do at one customer of mine where we currently have 3 backing storages. Since deletions typically occur on the older ones, they grow in "free space" over time, so we periodically move some year/month folders from the newer volumes over to the old to fill those "holes"
2019-03-28 08:48:06 GMT <alfresco-discord> <MorganP> Yeah
2019-03-28 08:49:17 GMT <AFaust> Of course, the ideal scenario for large content stores is always to have a content-addresses storage system (S3, Caringo Swarm or the like) where you simply add storage transparently and let the storage system handle distribution + performance optimisations....
2019-03-28 08:52:37 GMT <alfresco-discord> <MorganP> Thanks for the feedback
2019-03-28 08:55:13 GMT <alfresco-discord> <MorganP> Maybe one last question... Do you know how the search happen when the aggregating is trying to find for a file? It must add a overhead in performance, especially if there are huge stores with millions of nodes
2019-03-28 09:02:31 GMT <alfresco-discord> <yreg> @MorganP AFAIK, that would depend hugely on the aggregated stores...
2019-03-28 09:06:13 GMT <alfresco-discord> <yreg> and on where the content will be found ...
2019-03-28 09:06:35 GMT <alfresco-discord> <yreg> Write will always go to the primary/current store, while read will try to check if the file exists on the primary store, then iterates over the secondary stores one by one ...
2019-03-28 09:06:46 GMT <alfresco-discord> <yreg> https://github.com/Alfresco/alfresco-repository/blob/4ce23217bddf2cf1ccddf943a2106c6fa368fec5/src/main/java/org/alfresco/repo/content/replication/AggregatingContentStore.java#L130
2019-03-28 09:06:47 GMT <alfbot> Title:alfresco-repository/AggregatingContentStore.java at 4ce23217bddf2cf1ccddf943a2106c6fa368fec5 · Alfresco/alfresco-repository · GitHub (at github.com)
2019-03-28 09:08:21 GMT <alfresco-discord> <MorganP> Yes absolutely
2019-03-28 09:28:02 GMT <hi-ko> MorganP, AFaust: using a HSM aware storage you could define speeds and location by rules. You don't need S3 or other API/Ticket based storage provider to implement storage rules independant from alfresco
2019-03-28 09:28:47 GMT <angelborroy> Hardware Security Module?
2019-03-28 09:29:02 GMT <hi-ko> https://en.wikipedia.org/wiki/Hierarchical_storage_management
2019-03-28 09:29:03 GMT <alfbot> Title:Hierarchical storage management - Wikipedia (at en.wikipedia.org)
2019-03-28 09:29:07 GMT <AFaust> Sure... I wasn't going to list all options and just focused on what alternatives I have seen so far
2019-03-28 09:29:08 GMT <angelborroy> ah :D
2019-03-28 09:29:52 GMT <hi-ko> I'm not a fan of reinventing the wheel in Alfresco if there is mature technology available
2019-03-28 09:39:31 GMT <hi-ko> Could someone enlighten me for the activity feed mechanism, please? How does Alfresco find out which entries have to be created in table alf_activity_feed?
2019-03-28 09:40:49 GMT <hi-ko> The mechanism seems not to operate reliable but I didn't find a pattern which documents are covered by the activity post/feed jobs and which not.
2019-03-28 09:47:57 GMT <AFaust> alf_actiovity_feed is based off of alf_activity_post
2019-03-28 09:58:49 GMT <hi-ko> thanks Axel - so I would like rephrase: how works the logic which inserts the entries in alf_activity_post to find out why for some documents they occur and for some not.
2019-03-28 10:00:26 GMT <hi-ko> AFaust: My suspicion is that the entries are dependant from the transform transaction
2019-03-28 10:01:08 GMT <AFaust> hi-ko WIll answer after my telco is done
2019-03-28 10:01:25 GMT <hi-ko> AFaust: thanks!
2019-03-28 10:20:29 GMT <AFaust> Ok... alf_activity_post entries are actively triggered by various components, e.g. specific Share actions via a "poster" web script, WebDAV / CMIS via special segments in their integration layer, or any custom code wanting to create a specific record about an "activity"
2019-03-28 10:21:49 GMT <AFaust> So e.g. concerning preview / download activity post events, it absolutely depends on which interface / API the user has used. E.g. an access via the IMAP service should not trigger an activity post (last I checked it had no code for activity posting), but using Share would - for the very same document
2019-03-28 10:22:49 GMT <AFaust> Adding someone to SiteManager group of a site using Share admin console tools will not trigger an activity post, but doing so via the Site Members Share page will...
2019-03-28 10:24:49 GMT <AFaust> The background activity feed generator will process activity post entries, check which user(s) should get it in their feeds, and generated that many entries in the feed table...
2019-03-28 10:32:27 GMT <hi-ko> AFaust: thanks. I've seen that activityPoster is included explicitly in varous operations with no or only rare debug. So I will start with the FileFolderActivityPosterImpl to debug why some docs occur in the posts and some not ...
2019-03-28 10:33:16 GMT <AFaust> Keep in mind that feed generation will also perform permission checking to filter out the recipients for feed entries
2019-03-28 10:34:12 GMT <hi-ko> sure but post entries should be created anyway
2019-03-28 10:34:21 GMT <AFaust> And the nodeRef has to be provided in a very specific way for that to work in the generic manner that it does. I know from experience that some code may not do that correctly (I have patched blog-related web scripts at one customer for a couple of years to fix that)
2019-03-28 10:34:31 GMT <AFaust> Yes, post entries should be created anyway
2019-03-28 10:38:13 GMT <hi-ko> so we should throw it into the trash bin and go for an event driven aproach similar to jeffs alfresco-kafka showcase ;-)
2019-03-28 10:40:57 GMT <AFaust> I bet that is going to be the way of the future anyway with their new event gateway in 6.1+
2019-03-28 10:41:27 GMT <AFaust> Though they will have to create special kinds of events to separate those from low-level behaviour / policy events
2019-03-28 10:42:44 GMT <AFaust> Alternatively, I think there is also a chance that activities will be dropped entirely when Share is axed in 2+x years, since there is no indication that activities have a future in ADF / ACA
2019-03-28 10:43:16 GMT <AFaust> Activities are part of the "collaboration platform" approach that Alfresco all but abandoned strategically
2019-03-28 10:45:43 GMT <hi-ko> and their their new event gateway in 6.1+ is EE only
2019-03-28 10:47:50 GMT <hi-ko> so if we want to still use an community approach in future we should evaluate if this is something we sould work on (community event mechanism) or to leave (love it or leave it)
2019-03-28 10:51:23 GMT <hi-ko> i'm afraid Alfresco will become meaningless if all these changes will not be part of the CE and many EE customers will leave if there is no migration path for "collaboration platform" use cases out of the box
2019-03-28 10:53:09 GMT <alfresco-discord> <digcat> well the 3 year PE flip is still to come, so what the strategy will be after that who knows
2019-03-28 10:54:50 GMT <hi-ko> digcat: sure - but people decide now when to switch horses if there are alternatives for the use cases which may work in 3 years somehow
2019-03-28 10:55:40 GMT <alfresco-discord> <digcat> yes too true
2019-03-28 10:59:46 GMT <AFaust> hi-ko: well, the message queue for events is part of CE
2019-03-28 11:01:04 GMT <AFaust> The transformation queue is currently EE-only. We should not confuse the various queues Alfresco is introducing...
2019-03-28 11:06:58 GMT <alfresco-discord> <digcat> adding to that, on 6.1.2-ga, we also have a rendition queue, doing something?
2019-03-28 11:07:00 GMT <alfresco-discord> <digcat> https://cdn.discordapp.com/attachments/451644531323174914/560781885069066260/unknown.png
2019-03-28 11:09:31 GMT <alfresco-discord> <drazen04> Hi guys, anyone knows how to properly set activiti debug?
2019-03-28 11:18:30 GMT <hi-ko> AFaust: it is confusing. Where can I find the message queue for events?
2019-03-28 11:19:36 GMT <AFaust> See https://github.com/Alfresco/acs-community-deployment/blob/master/docker-compose/docker-compose.yml
2019-03-28 11:19:37 GMT <alfbot> Title:acs-community-deployment/docker-compose.yml at master · Alfresco/acs-community-deployment · GitHub (at github.com)
2019-03-28 11:21:10 GMT <angelborroy> I’m not sure that ActiveMQ is used now in Community
2019-03-28 11:21:29 GMT <angelborroy> But it must be used in the future (imo)
2019-03-28 11:21:58 GMT <hi-ko> angelborroy: used for what?
2019-03-28 11:22:03 GMT <alfresco-discord> <digcat> hi @angel.borroy the deploy has it, and its doing something on the latest 6.1.2-ga release, logging into MQ you can see messages being enqueued/dequeued
2019-03-28 11:22:41 GMT <angelborroy> at least for Transformer Service
2019-03-28 11:22:51 GMT <hi-ko> haha
2019-03-28 11:22:54 GMT <angelborroy> @digcat acs-community-deployment
2019-03-28 11:22:55 GMT <angelborroy> ?
2019-03-28 11:23:02 GMT <alfresco-discord> <digcat> yes
2019-03-28 11:23:22 GMT <hi-ko> digcat: so maybe deployed by accident ;-)
2019-03-28 11:24:07 GMT <angelborroy> I see
2019-03-28 11:24:30 GMT <angelborroy> it looks like it’s configured for some service in the repo
2019-03-28 11:24:31 GMT <angelborroy> https://github.com/Alfresco/acs-community-deployment/blob/master/docker-compose/docker-compose.yml#L29
2019-03-28 11:24:32 GMT <alfbot> Title:acs-community-deployment/docker-compose.yml at master · Alfresco/acs-community-deployment · GitHub (at github.com)
2019-03-28 11:26:35 GMT <AFaust> angelborroy The messaging subsystem is used for the event-based out-of-process behaviour / policies
2019-03-28 11:26:59 GMT <angelborroy> Wow, I didn’t know that
2019-03-28 11:27:33 GMT <AFaust> See https://github.com/Alfresco/alfresco-repository/blob/cda91cf7606b8b0487beab6681557f0424ca9ebb/src/main/java/org/alfresco/repo/events/EventGenerationBehaviours.java
2019-03-28 11:27:34 GMT <alfbot> Title:alfresco-repository/EventGenerationBehaviours.java at cda91cf7606b8b0487beab6681557f0424ca9ebb · Alfresco/alfresco-repository · GitHub (at github.com)
2019-03-28 11:27:46 GMT <AFaust> and https://github.com/Alfresco/alfresco-repository/blob/cda91cf7606b8b0487beab6681557f0424ca9ebb/src/main/resources/alfresco/subsystems/Events/default/events-context.xml#L39
2019-03-28 11:27:47 GMT <alfbot> Title:alfresco-repository/events-context.xml at cda91cf7606b8b0487beab6681557f0424ca9ebb · Alfresco/alfresco-repository · GitHub (at github.com)
2019-03-28 11:28:36 GMT <AFaust> And they go the queue via the eventsDispatcher bean (CamelMessageProducer)
2019-03-28 11:28:39 GMT <angelborroy> Nice
2019-03-28 11:29:20 GMT <alfresco-discord> <dgradecak> it is becoming an "enterprise" grade product 😃
2019-03-28 11:30:29 GMT <AFaust> angelborroy: So I take it the orientation at Alfresco does not really give new people an overview of the product? That is what I thought after realising how most of the people that started on ADF a few years ago don't know half of what Alfresco is typically used for...
2019-03-28 11:31:18 GMT <angelborroy> You are right
2019-03-28 11:31:42 GMT <AFaust> And you really need the ActiveMQ in the default configuration or Alfresco 6.1 will hang during startup (unless you disable autoStart for the subsystem)
2019-03-28 11:32:35 GMT <AFaust> I am still waiting on some negative side-effects of constantly running my 6.1 instances with messaging disabled
2019-03-28 11:32:42 GMT <alfresco-discord> <dgradecak> who ever needs it is : messaging.subsystem.autoStart=false events.subsystem.autoStart=false
2019-03-28 11:32:51 GMT <AFaust> ...and without an ActiveMQ to connect to if some component tries to activate messaging
2019-03-28 11:32:56 GMT <alfresco-discord> <dgradecak> I do the same AFaust
2019-03-28 11:33:12 GMT <AFaust> Ah - haven't used the events.subsystem.autoStart=false yet - should probably do that
2019-03-28 11:33:18 GMT <alfresco-discord> <dgradecak> even an enterprise client does not want to enable it
2019-03-28 11:33:45 GMT <alfresco-discord> <dgradecak> so they do not need to care about activemq by having admins for that
2019-03-28 11:34:05 GMT <AFaust> I mean, for most use case purposes it is completely impractical to do out-of-process behaviours with the non-transactional, low-granularity ReST API we have.
2019-03-28 11:34:32 GMT <AFaust> low-granularity => low-level
2019-03-28 11:34:39 GMT <alfresco-discord> <dgradecak> completely agree, for BI processing it does make sense but that is another story
2019-03-28 11:35:09 GMT <alfresco-discord> <dgradecak> however an out of transaction behaviour I (for now) do not see the use case
2019-03-28 11:35:26 GMT <AFaust> Let me count the number of use cases / requests I have received over the years.... wonder if I'll have to use my second hand.
2019-03-28 11:35:45 GMT <AFaust> ^^ regarding BI I mean
2019-03-28 11:35:54 GMT <alfresco-discord> <dgradecak> if you hid the first hand than you might need the second 😄
2019-03-28 11:36:02 GMT <hi-ko> but for BI it is overengeneered since we still need to dequeue and store in another structure
2019-03-28 11:36:15 GMT <alfresco-discord> <dgradecak> indeed
2019-03-28 11:36:52 GMT <hi-ko> strange - I need a concepts guide ;-)
2019-03-28 11:37:04 GMT <AFaust> Well... someone is probably bound to use it out there, and I hope they feel privileged to have a feature mostly to themselves...
2019-03-28 11:37:25 GMT <alfresco-discord> <dgradecak> hi-ko for the activity feed (I just read previous messages), there are some feeds happening out of treansaction
2019-03-28 11:37:32 GMT <alfresco-discord> <dgradecak> but you probably know all that
2019-03-28 11:38:38 GMT <alfresco-discord> <dgradecak> btw, anyone knows what will happen to the "auditing" inside of alfresco?
2019-03-28 11:39:04 GMT <alfresco-discord> <dgradecak> are they going to drop it or leave it without using it later?
2019-03-28 11:39:30 GMT <hi-ko> dgradecak: the system I'm inspecting has not a problem by some feeds happening out of treansaction but missing feeds which does not have a pattern I can see (same methods, differnt users, documents)
2019-03-28 11:40:00 GMT <alfresco-discord> <dgradecak> do you know what feeds exactly?
2019-03-28 11:40:37 GMT <hi-ko> simple once like create, modified content
2019-03-28 11:41:11 GMT <hi-ko> which should be captured by the filefolderservice
2019-03-28 11:41:36 GMT <hi-ko> I think the auditing makes no longer sense the way as today if we have messaging
2019-03-28 11:42:32 GMT <alfresco-discord> <dgradecak> well I think it does, as messaging is "out of transaction" and I wonder what would happen to Records management/AGS without auditing
2019-03-28 11:43:00 GMT <AFaust> Well, it still makes sense, since you cannot extract "current" context-information from the aync event queue
2019-03-28 11:44:22 GMT <AFaust> The main thing about audit that no longer makes sense though is storing the data in the same DB, with nested write transactions in read-only transactions. That is where the queue should be used.
2019-03-28 11:46:52 GMT <alfresco-discord> <dgradecak> I agree that for readonly transactions it is not a good practice, but on the other side how to do that if "read" events need to be audited. Although XA tranactions could be used with the queue but htan processing different systems to get some data for a node ...
2019-03-28 11:48:52 GMT <AFaust> In Alfresco we already have transaction listeners, so audit data in a txn could simply be accumulated and submitted before commit / rollback. Txn listeners also work for read-only txns
2019-03-28 11:49:35 GMT <AFaust> ^^ rather on commit/rollback
2019-03-28 11:50:10 GMT <alfresco-discord> <dgradecak> indeed, but having two resources in a transaction is always "problematic", the "data was written to the queue" but somehow rollbacked in the db
2019-03-28 11:50:23 GMT <alfresco-discord> <dgradecak> than need to remove the data from the queue etc. .. if you see what I mean
2019-03-28 11:50:54 GMT <AFaust> Yeah, but easily doable in a post-processing step correlating audit txns with the alf_transaction table
2019-03-28 11:51:02 GMT <alfresco-discord> <dgradecak> that is where a XA transaction comes in place, db + mq
2019-03-28 11:51:14 GMT <alfresco-discord> <dgradecak> indeed it is always "esay" to do that way 😉
2019-03-28 13:03:13 GMT <hi-ko> dgradecak, AFaust: we had this discussion also during lunch. Transaction aware messaging is really an issue and required (either by compensation or by encapsulation) Our conclusion so far: messages should be send _after_ transaction which is not easy but maybe doable if we get a concept of outer transaction finish. I think a solution is far away if we don't take a view on the use cases also.
2019-03-28 13:05:18 GMT <AFaust> I don't know if any of you checked the implementation yet, the EventsService in 6.1 already uses txn listeners to send behaviour events on commit and a special rollback event on, well, rollback
2019-03-28 13:07:03 GMT <AFaust> Minor correction - the behaviour events are sent beforeCommit, but onCommit there will be a special commit event
2019-03-28 13:07:28 GMT <AFaust> So in the queue you can already check which events have been cancelled by a rollback that occurs during commit
2019-03-28 13:08:11 GMT <hi-ko> hmm. what if I only want see see really committed events?
2019-03-28 13:10:51 GMT <hi-ko> It's not my expectation of messaging if I have to still replay all transactions to have a consistent view
2019-03-28 13:15:11 GMT <AFaust> For "really committed" events you drain all the events from the queue into a temporary structure, and once you get he "commit" event for the same txn ID you can process them. If you get a rollback event for the txn ID you discard...
2019-03-28 13:19:52 GMT <hi-ko> That kind of service shouldn't be implemented outside. The implementation you described remembers me the dicussion we had at oracle years ago when they moved from replication (which requires to replay every event) to sync based on streams (which guarantees a consistent view based on net changes/events required for an external view)
2019-03-28 13:20:18 GMT <angelborroy> · Probably we would use the same approach for SOLR Tracker
2019-03-28 13:20:49 GMT <hi-ko> angelborroy: s/would/should/
2019-03-28 13:21:20 GMT <angelborroy> agree
2019-03-28 13:21:27 GMT <angelborroy> But it’s not my decision ;-)
2019-03-28 13:21:37 GMT <hi-ko> what a pitty!
2019-03-28 13:22:08 GMT <alfresco-discord> <yreg> @angel.borroy if you do you will reach (almost) transactional consistency in solr as opposed to the current state
2019-03-28 13:22:32 GMT <angelborroy> what I consider a nice feature, right?
2019-03-28 13:22:40 GMT <hi-ko> we implemented something like this in our smarttransformer queue to avoid unnecessary transformations but it should be seen on a generic CRUD level
2019-03-28 13:22:48 GMT <alfresco-discord> <yreg> where Alfresco recommends 2-10 minutes tracking interval on such big repos
2019-03-28 13:23:02 GMT <angelborroy> “Alfresco recommends”
2019-03-28 13:23:08 GMT <angelborroy> bah
2019-03-28 13:23:26 GMT <angelborroy> Who’s following Alfresco recommendations?
2019-03-28 13:23:28 GMT <angelborroy> :-P
2019-03-28 13:23:38 GMT <hi-ko> you should now ;-)
2019-03-28 13:24:26 GMT <alfresco-discord> <yreg> well, believe it or not, clients over here follow vendor recommendations over integrators most of the time ....
2019-03-28 13:24:47 GMT <angelborroy> Just don’t let me talk
2019-03-28 13:25:11 GMT <alfresco-discord> <yreg> and when you hit issues with solr's tracking crippling repo/db you have no other option than following the recommendation 😄
2019-03-28 13:26:50 GMT <angelborroy> We’re preparing a huge environment to measure perfomance and upgrading operations
2019-03-28 13:26:56 GMT <angelborroy> But it’s not ready yet
2019-03-28 13:38:34 GMT <alfresco-discord> <dgradecak> regarding the solr tracking, taht is what I did for my "BI" tool, and I am using it for all kind of integrations with alfresco now
2019-03-28 13:38:54 GMT <alfresco-discord> <dgradecak> track alfresco transactions and replay them in a different structure where it is needed
2019-03-28 13:39:36 GMT <alfresco-discord> <dgradecak> and the same approach for auditing but just using a different alfresco rest api, audit apis instead of solr alfresco api
2019-03-28 13:41:01 GMT <alfresco-discord> <dgradecak> btw, do you guys have a sample how to reuse a share "user selection" component for a custom model property (list of users)
2019-03-28 13:41:22 GMT <angelborroy> it’s in my blog
2019-03-28 13:41:25 GMT <angelborroy> let me see
2019-03-28 13:41:47 GMT <angelborroy> https://angelborroy.wordpress.com/2018/02/17/alfresco-share-5-using-authority-ftl-form-control/
2019-03-28 13:41:48 GMT <alfbot> Title:Alfresco Share 5, using “authority.ftl” form control | Programming and So (at angelborroy.wordpress.com)
2019-03-28 13:42:31 GMT <hi-ko> dgradecak: we started this way for our sync mechansim but it's on my wishlist to rework this to a consolidated view having only committed changes per node and a different table with all latest facts which are required for bi but to expensive to query
2019-03-28 13:43:01 GMT <alfresco-discord> <dgradecak> did you check my presentation at the devcon hi-ko? was a lightning talk
2019-03-28 13:43:28 GMT <alfresco-discord> <dgradecak> you can get it from dockerhub too and try the "app"
2019-03-28 13:43:34 GMT <hi-ko> dgradecak: not yet. but I will do!
2019-03-28 13:44:02 GMT <alfresco-discord> <dgradecak> the "worksapce data" is free, the audit data is not
2019-03-28 13:46:00 GMT <hi-ko> We should join forces and work on alternatives commercial or non commercial to not rely on the stuff not to come ...
2019-03-28 13:46:45 GMT <angelborroy> eh, eh, the new stuff is (will be) reliable
2019-03-28 13:47:13 GMT <alfresco-discord> <dgradecak> agree hi-ko, ping for a chat 😉
2019-03-28 13:47:36 GMT <alfresco-discord> <dgradecak> @angel thank you for the reference
2019-03-28 13:47:51 GMT <hi-ko> angelborroy: I'd love to believe
2019-03-28 13:48:05 GMT <angelborroy> I’m here because I’m a believer :D
2019-03-28 13:49:38 GMT <hi-ko> I'd take the role of a maker ;-)
2019-03-28 13:49:55 GMT <alfresco-discord> <dgradecak> any idea if the authority.ftl could use a "multiple" property instead of an association?
2019-03-28 13:50:09 GMT <angelborroy> no, requires an association
2019-03-28 13:50:34 GMT <angelborroy> with a cm:person target
2019-03-28 14:35:33 GMT <alfresco-discord> <dgradecak> the category.ftl works without associations and the authority.ftl with associations, I wish share was consistent 😄
2019-03-28 14:38:53 GMT <angelborroy> late!
2019-03-28 14:39:01 GMT <angelborroy> It’s EOL now :-P
2019-03-28 14:39:17 GMT <alfresco-discord> <dgradecak> I know and hopefully it is
2019-03-28 14:41:00 GMT <alfresco-discord> <dgradecak> maybe one day there will be a decent replacement 😉
The other logs are at http://esplins.org/hash_alfresco