Daily Log for #alfresco IRC Channel

Alfresco discussion and collaboration. Stick around a few hours after asking a question.

Official support for Enterprise subscribers: support.alfresco.com.

Joining the Channel:

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.

Getting Help

More help is available in this list of resources.

Daily Log for #alfresco

2017-03-09 05:26:39 GMT <Sandy> Hai evaryone,Need to get documents metadata from alfresco database, can anyone help me please.

2017-03-09 08:35:58 GMT <Sandy> bhagyas: Yes bhagyas

2017-03-09 08:37:08 GMT <Sandy> Tichodroma: i need to show document properties in alfresco database(i.e postgres)

2017-03-09 08:39:44 GMT <Tichodroma> Sandy: that does not make sense. *Where* do you want to "show" them?

2017-03-09 08:43:19 GMT <oussama> is it possible to login in alfresco using only a ticket?

2017-03-09 08:43:44 GMT <angelborroy> oussama it is

2017-03-09 08:44:00 GMT <angelborroy> oussama by using REST API

2017-03-09 08:44:19 GMT <oussama> how can i achieve it please? i don't mean to call webscripts using a ticket that i can do

2017-03-09 08:44:45 GMT <oussama> i want to actually get a session

2017-03-09 08:44:45 GMT <angelborroy> so what do you mean?

2017-03-09 08:44:57 GMT <angelborroy> you can use CMIS API for that

2017-03-09 08:45:06 GMT <oussama> like i used username/password to login

2017-03-09 08:45:28 GMT <Sandy> Tichodroma: i need to show document properties in database using sql query.

2017-03-09 08:45:44 GMT <angelborroy> oussama which API are you using?

2017-03-09 08:46:26 GMT <Tichodroma> Sandy: Do you want to access the database directly? Do you have problems understanding the table structure?

2017-03-09 08:47:19 GMT <oussama> <angelborroy>: i'm not sure i understood your question sorry

2017-03-09 08:47:39 GMT <angelborroy> oussama you want to use a ticket to obtain access to Alfresco, right?

2017-03-09 08:47:53 GMT <angelborroy> oussama so you need to use REST API

2017-03-09 08:48:29 GMT <Sandy> Tichodroma: Actually my aim is need to upload bulk documents into alfresco and need to update documents properties with excel sheet. is there any solution. suggest me friend

2017-03-09 08:48:46 GMT <oussama> yes, i'm trying to implement a delegation mecanism where a user can have access to another user's account for a certain period of time

2017-03-09 08:48:55 GMT <Tichodroma> Sandy: That's a completely different question.

2017-03-09 08:49:03 GMT <oussama> so i need to log him in without a password

2017-03-09 08:49:27 GMT <angelborroy> oussama look at user-account-switcher at github

2017-03-09 08:49:30 GMT <Tichodroma> Sandy: you could prepare the metadata XML files for the documents before you execute the bulk upload

2017-03-09 08:49:50 GMT <Tichodroma> Sandy: https://community.alfresco.com/docs/DOC-6079-bulk-importer

2017-03-09 08:49:53 GMT <alfbot> Title: Bulk Importer | Alfresco Community (at community.alfresco.com)

2017-03-09 08:50:24 GMT <oussama> <angelborroy> thank you so much i'll take a look at that

2017-03-09 08:52:03 GMT <Sandy> Tichodroma: ok thank you friend. :)-

2017-03-09 09:16:38 GMT * yregaieg wishes everyone on the channel a good morning !

2017-03-09 10:15:09 GMT <fwu> hi all!

2017-03-09 10:37:24 GMT <fwu> angelborroy, did you ever made an imap configuration for Outlook 2013 or 2016?

2017-03-09 10:39:33 GMT <angelborroy> nope, sorry

2017-03-09 10:53:49 GMT <IanW1> fwu: I did - was fairly straightforward IIRC

2017-03-09 11:02:17 GMT <fwu> IanW1, I have a strange problem

2017-03-09 11:02:39 GMT <fwu> the imap configuration in Alfresco seems to be working. I can login with telnet

2017-03-09 11:03:06 GMT <fwu> but in outlook 2013, outlook will atempt to send a test email, that it is failling.

2017-03-09 11:03:17 GMT * AFaust wonders if anyone noticed that Alfresco LockStore is not transactionally safe...

2017-03-09 11:03:50 GMT <fwu> I read somewhere that outlook will send the email in a wrong way, so it will get an error.

2017-03-09 11:04:12 GMT <fwu> the problem is that if outlook cant send the email, Im not able to complete the configuration....

2017-03-09 11:05:20 GMT <IanW1> I think we configured outlook to use the a different outgoing server (don't have one I can check at the moment) on that account i.e. Alfresco imap on incoming only

2017-03-09 11:06:27 GMT <fwu> but the configuration is for both. I only have a server to specify.

2017-03-09 11:07:09 GMT <fwu> so, if I set the alfresco server, it will be used or incoming and outcoming... or am I seing it wrong?

2017-03-09 11:14:27 GMT <IanW1> I'm pretty sure you can define separate incoming and outgoing servers - advanced options?

2017-03-09 11:22:30 GMT <fwu> IanW1, yes, you are right. I have two boxes

2017-03-09 11:23:32 GMT <Sumit> Hi, I have tried sign pdf of alfresco pdf toolkit. It gives me success message and signed icon but document not showing any signature on it while viewing. can someone knows about this?

2017-03-09 11:23:42 GMT <fwu> What I need is to have an Alfresco folder inside outlook, so that each email droped in that folder, will be seen in Alfresco

2017-03-09 11:24:00 GMT <fwu> to accomplish this I only need to set the incoming server?

2017-03-09 11:24:47 GMT <fwu> what will be the use for the outcoming server then?

2017-03-09 11:39:39 GMT <IanW1> There's no use for the outgoing server (other than keeping outlook happy)

2017-03-09 11:52:12 GMT <mrks_js> IanW1: have you encountered this bug anywhere? https://issues.alfresco.com/jira/browse/MNT-16931

2017-03-09 11:54:23 GMT <IanW1> mrks_js: Don't use Kerberos - hooray!

2017-03-09 11:54:50 GMT <mrks_js> :-(

2017-03-09 11:56:51 GMT <mrks_js> right! why am i mixing this up ... again

2017-03-09 12:07:57 GMT <AFaust> mrks_js: That issue you linked is not a bug IMHO - Alfresco has consciously never supported falling back to other authentication subsystems for SSO. Either the first succeeds or authentication fails.

2017-03-09 12:08:12 GMT <AFaust> This is more like an enhancement request...

2017-03-09 12:08:22 GMT <douglascrp> morning

2017-03-09 12:09:37 GMT <AFaust> I even believe that someone commented on a different JIRA ticket (that angelborroy reported) that they do not intend to allow fallbacks to less secure protocols (i.e. BASIC authentication) since "that would open back doors"

2017-03-09 12:11:53 GMT <angelborroy> AFaust are you referring this one? https://issues.alfresco.com/jira/browse/ALF-21521

2017-03-09 12:12:26 GMT <AFaust> Yeah...

2017-03-09 12:12:52 GMT <AFaust> Basically this could be simplified to "We believe that having a single authentication stack securing all [...] endpoints is more secure, even if it is less flexible."

2017-03-09 12:13:15 GMT <AFaust> Although in that ALF-21521 the scope was specific to ReST endpoints

2017-03-09 12:14:33 GMT <angelborroy> AFaust they were very clear at this point

2017-03-09 12:14:39 GMT <AFaust> But who knows - MNT-16931 was put "IN PROGRESS" two hours ago...

2017-03-09 12:15:17 GMT <angelborroy> AFaust they are not going to solve this issue becaues they think that this is not an issue

2017-03-09 12:15:31 GMT <angelborroy> AFaust I asked them last week in London

2017-03-09 12:16:06 GMT <angelborroy> mrks_js I published a home made patch for this at https://github.com/angelborroy/mnt-15899-share

2017-03-09 12:16:07 GMT <alfbot> Title: GitHub - angelborroy/mnt-15899-share (at github.com)

2017-03-09 12:16:14 GMT <angelborroy> mrks_js under your own risk

2017-03-09 12:25:01 GMT <angelborroy> someone knows if new AOS have the same problem like old VTI with large paths?

2017-03-09 12:28:05 GMT <AFaust> angelborroy: It has not - there is a UUID-based alternative path now and Share UI should switch to that if normal path is too long

2017-03-09 12:28:15 GMT <angelborroy> ok, thanks

2017-03-09 12:28:25 GMT <AFaust> angelborroy: Old VTI also had that capability in 4.2 or so, but they did not include the fallback in Share

2017-03-09 12:28:44 GMT <AFaust> I once ported this fallback over to VTI as well

2017-03-09 12:28:50 GMT <angelborroy> me too :)

2017-03-09 12:36:57 GMT <mrks_js> interesting!

2017-03-09 12:37:12 GMT <mrks_js> angelborroy: the patch is removing the api/* filter mapping?

2017-03-09 12:37:22 GMT <angelborroy> right

2017-03-09 12:37:36 GMT <angelborroy> so default user/pass auth is applied to API

2017-03-09 12:37:39 GMT <mrks_js> this does not work anymore with 201702GA

2017-03-09 12:37:49 GMT <mrks_js> i had this patch in 5.0.d working

2017-03-09 12:38:15 GMT <angelborroy> we need to find a new patch :)

2017-03-09 12:38:24 GMT <mrks_js> yes!

2017-03-09 12:39:45 GMT <mrks_js> maybe there will be an update in https://issues.alfresco.com/jira/browse/MNT-16931 soon

2017-03-09 12:39:48 GMT <yregaieg> At what level alfresco caches data ?

2017-03-09 12:41:36 GMT <yregaieg> I just found out from a memory dump that 30% of my heap was allocated to org.apache.ibatis.executor.SimpleExecutor -> localCache->cache->table

2017-03-09 12:43:07 GMT <yregaieg> 13% is reserved ti org.alfresco.repo.cache.DefaultSimpleCache->cache->localCache->segments (This makes way more sense to me than the former)

2017-03-09 12:44:48 GMT <yregaieg> and 13% in java.lang.Thread ->threadLocals->table .. (I assume this is for TransactionalCache, so it should be ok)

2017-03-09 12:46:16 GMT <yregaieg> AFaust: any idea how to limit the size of ibatis caches in alfresco ?

2017-03-09 12:46:34 GMT <AFaust> Via the ibatis config...

2017-03-09 12:47:06 GMT <AFaust> Should be configurable via the alfresco-SqlMapConfig.xml

2017-03-09 12:47:32 GMT <AFaust> unfortunately that means copy&pasting the original file and overriding via the /WEB-INF/classes/ directory

2017-03-09 12:48:21 GMT <yregaieg> Does it make sense for it to allocate 3Gb of caches ?

2017-03-09 12:49:57 GMT <yregaieg> I can not locate any cache config in the suggested file

2017-03-09 12:51:12 GMT <IanW1> I have a nasty workaround for my CAS SSO where I skip the CAS filters if a basic-auth challenge is done as part of the request - seems to work...

2017-03-09 12:51:35 GMT <AFaust> There is no cache config by default in the Alfresco file so it should use some internal defaults

2017-03-09 12:51:45 GMT <AFaust> Probably defaults that do not define a limit

2017-03-09 12:54:49 GMT <AFaust> I also think that concrete caches are to be configured in the actual mapper files (which would mean even more copy&override of default config files)

2017-03-09 12:57:18 GMT <yregaieg> I do not understand

2017-03-09 12:58:01 GMT <yregaieg> the job resulting in this OOM error used to work just fine on alf 4.2 with only 4 GB of RAM

2017-03-09 12:58:22 GMT <yregaieg> on the old test environment

2017-03-09 12:58:55 GMT <AFaust> You were talking about ibatis caches. ibatis caches are AFAIK not handled via regular Alfresco caches

2017-03-09 12:59:04 GMT <yregaieg> now on alf 5.1 I am giving 9Gb to the tomcat and it still throws OOME halfway

2017-03-09 12:59:14 GMT <AFaust> ibatis has its own cache interfaces / implementations

2017-03-09 13:00:04 GMT <yregaieg> Is it normal for ibatis caches in alfresco to retain more then actual alfresco caches ?

2017-03-09 13:00:11 GMT <yregaieg> combines

2017-03-09 13:00:18 GMT <yregaieg> combined**

2017-03-09 13:00:25 GMT <AFaust> threadLocals -> table is not necessarily related to just TransactionalCache - it is related to anything that is handled via thread locals, e.g. listeners, policy queue etc.

2017-03-09 13:00:51 GMT <AFaust> It is not normal, but it depends on the kind of queries you are forcing Alfresco to do and the structure of your data.

2017-03-09 13:01:46 GMT <AFaust> e.g. ibatis may cache the entire result set of a query, so if you query for a large set with many of columns the result sets may be quite large in memory footprint

2017-03-09 13:02:34 GMT <yregaieg> this job lists all documents from alfresco (path, name, size) in a ~1Gb csv file

2017-03-09 13:03:02 GMT <yregaieg> it uses batch processor with 2 threads

2017-03-09 13:03:18 GMT <yregaieg> for preparing the reports

2017-03-09 13:03:21 GMT <AFaust> So if you look at the SimpleExecutor + BaseExecutor classes you will see that "localCache" is a PerpetualCache with an unbounded map as "cache"

2017-03-09 13:03:52 GMT <AFaust> On a commit, the localCache should be cleared

2017-03-09 13:04:06 GMT <AFaust> same with rollback and any modifying operation

2017-03-09 13:04:27 GMT <AFaust> so it is only used to cache data for read-only results

2017-03-09 13:04:43 GMT <AFaust> So is the job running in a single thread?

2017-03-09 13:04:49 GMT <AFaust> e.g. no batch processing etc.?

2017-03-09 13:04:58 GMT <yregaieg> 2 threads

2017-03-09 13:05:14 GMT <AFaust> And how many transactions?

2017-03-09 13:05:39 GMT <AFaust> e.g. is the data for the 1 GiB CSV file collected in few, large transactions or many small ones?

2017-03-09 13:07:10 GMT <yregaieg> AFaust: splitTxn with 200 nodes per batch

2017-03-09 13:07:47 GMT <yregaieg> http://pastebin.com/HCSRfDwE

2017-03-09 13:07:48 GMT <alfbot> Title: [Java] BatchProcessor<NodeRef> batchProcessor = new - Pastebin.com (at pastebin.com)

2017-03-09 13:08:17 GMT <AFaust> How does your work provider operate?

2017-03-09 13:09:02 GMT <AFaust> Is it avoiding loading to much data into the outer transaction?=

2017-03-09 13:10:30 GMT <yregaieg> hmmm it is loading pages 1000 per page on demand, storing them in a private ArrayList, and finally feeding them to the processor 200 at a time

2017-03-09 13:10:40 GMT <yregaieg> 1000 nodes

2017-03-09 13:10:58 GMT <yregaieg> nothing fancy

2017-03-09 13:11:16 GMT <AFaust> Ok - but HOW is it loading the nodes?

2017-03-09 13:11:56 GMT <AFaust> SearchService.query()?

2017-03-09 13:12:21 GMT <yregaieg> exactly

2017-03-09 13:12:48 GMT <AFaust> Ok - and did you use: searchParameters.setBulkFetchEnabled(false);

2017-03-09 13:13:40 GMT <yregaieg> ~ tell angelborroy is there an ETA for hotel/train deals for Beecon ? Should we make plans or wait for deals to be published ?

2017-03-09 13:13:40 GMT <alfbot> yregaieg: Error: I haven't seen angelborroy, I'll let you do the telling.

2017-03-09 13:14:00 GMT <yregaieg> ~later tell angelborroy is there an ETA for hotel/train deals for Beecon ? Should we make plans or wait for deals to be published ?

2017-03-09 13:14:00 GMT <alfbot> yregaieg: The operation succeeded.

2017-03-09 13:14:35 GMT <yregaieg> AFaust: no

2017-03-09 13:17:08 GMT <yregaieg> How would that affect caches in my case ?

2017-03-09 13:20:59 GMT <AFaust> When you don't set the bulk fetch to disabled, the query will result in all the nodes not yet cached to be loaded via ibatis

2017-03-09 13:21:38 GMT <AFaust> Since the work provider is running in the outer transaction which will not commit until the batch processor is done, you are keeping all the cache data associated with this pre-loading of nodes alive until the end

2017-03-09 13:21:59 GMT <AFaust> Better to disable pre-loading

2017-03-09 13:22:45 GMT <AFaust> It doesn't help with batch processing any way, because the pre-loaded data is not published to the shared cache until the transaction completes (after the batch is done), so each batch worker is loading the nodes that are not cached yet again any way...

2017-03-09 13:24:19 GMT <AFaust> so let's say you have a million nodes - now you are potentially caching the JDBC result set for 1000x1000 nodes, 1000x1000xNNN node properties, 1000x1000xMMM node aspects in the ibatis cache

2017-03-09 13:25:22 GMT <yregaieg> wow, that is too much indeed

2017-03-09 13:25:34 GMT <yregaieg> but was it any different in the past ?

2017-03-09 13:25:39 GMT <AFaust> Until the entire processing is done - if you don't do bulk loading, each transaction of a batch is at most holding 200 nodes, 200xNNN node properties and 200xMMM node aspects in the ibatis cache before it commits and the cache is cleared before starting the next batch

2017-03-09 13:25:45 GMT <yregaieg> we did not have issues with it with 4.2

2017-03-09 13:26:17 GMT <AFaust> I would have to check the internal ibatis implementation for that version, but I doubt it has changed substantially in that regard...

2017-03-09 13:27:58 GMT <yregaieg> I will disable batch loading and see if it makes it any better (but I strongly doubt it would matter

2017-03-09 13:28:29 GMT <AFaust> Is the query a DB AFTS query?

2017-03-09 13:28:38 GMT <AFaust> And how complex is it?

2017-03-09 13:29:05 GMT <AFaust> Because result sets for that will also be cached...

2017-03-09 13:29:17 GMT <alfbot> angelborroy: Sent 15 minutes ago: <yregaieg> is there an ETA for hotel/train deals for Beecon ? Should we make plans or wait for deals to be published ?

2017-03-09 13:30:35 GMT <angelborroy> yregaieg better make your plans, it’s really hard to deal with the Council and so

2017-03-09 13:31:24 GMT <AFaust> douglascrp: Only 77 points to go - you can do it!

2017-03-09 13:36:48 GMT <douglascrp> AFaust, what

2017-03-09 13:36:50 GMT <douglascrp> 77 for what

2017-03-09 13:36:59 GMT <douglascrp> I really don't follow those numbers anymore

2017-03-09 13:37:05 GMT <douglascrp> but let me check

2017-03-09 13:38:12 GMT <douglascrp> ah, to reach dave :D

2017-03-09 13:38:21 GMT <douglascrp> I don't know why I get so many points

2017-03-09 13:38:45 GMT <douglascrp> does the "like" and "mark as useful" give that much points?

2017-03-09 13:38:58 GMT <AFaust> I mostly follow to see how active other people are (e.g. if there is any sign of life from Dave who kind of has me worried...)

2017-03-09 13:38:59 GMT <douglascrp> my level of validator is high

2017-03-09 13:39:21 GMT <douglascrp> but helper is far from yours for example

2017-03-09 13:39:48 GMT <AFaust> The "mark as useful" certainly gives quite a bit of points - for both the validator and the one who wrote the comment

2017-03-09 13:40:07 GMT <douglascrp> I think "like" and "mark as useful" should give less points

2017-03-09 13:40:13 GMT <AFaust> 3 points per "helpful" that you mark

2017-03-09 13:40:24 GMT <douglascrp> at least less than what is given for who answered

2017-03-09 13:41:10 GMT <AFaust> I actually find that bit fair - I have seen some threads where people devolved into a live chat and such are really hard to follow

2017-03-09 13:41:28 GMT <AFaust> At least that way it gives incentive to write fewer answers that are qualitatively of higher value

2017-03-09 13:41:51 GMT <douglascrp> sometimes I that is not fair

2017-03-09 13:42:14 GMT <douglascrp> ah, got your point

2017-03-09 13:42:16 GMT <AFaust> I mean you do get a point for each comment

2017-03-09 13:43:21 GMT <AFaust> So you have to spam only 8 comments out to get the same points as a single helpful one (assuming only one person marked it as helpful, which gives the commenter 7 points)

2017-03-09 13:43:48 GMT <douglascrp> too much rules :D

2017-03-09 13:44:12 GMT <AFaust> They aren't really "rules" - just some weighted incentives...

2017-03-09 13:48:02 GMT <douglascrp> AFaust, and I gave some points for others

2017-03-09 13:49:04 GMT <AFaust> you are still +10 net positive with points received vs given away

2017-03-09 13:49:21 GMT <douglascrp> where do you see all those things?

2017-03-09 13:49:42 GMT <AFaust> angelborroy is quite generous - he gave away 160 more than he received

2017-03-09 13:49:49 GMT <douglascrp> I see

2017-03-09 13:50:01 GMT <douglascrp> ah, now I saw

2017-03-09 13:50:14 GMT <douglascrp> just below my profile picture

2017-03-09 13:52:57 GMT <AFaust> The one thing I find is very off is the overall level progression. I mean, how likely is it that we will ever have someone with 15 million points? How long does Alfresco intend to keep the platform alive?

2017-03-09 13:53:29 GMT <yregaieg> AFaust: Does it make sense if I have a lot of reads on my system, and very few alter to set TX caches to a low number and overall caches to high values eg default values for TX caches, 6-10x the shared caches ?

2017-03-09 13:54:51 GMT <yregaieg> AFaust: You can create a bot to weight other user interractions (mark as useful for example) and do the same for every post having at least 3 users interracted the same way ...

2017-03-09 13:55:20 GMT <yregaieg> It won't bump into a million right away but it can give quite fast progression <grin>

2017-03-09 13:56:12 GMT <AFaust> yregaieg: I obviously meant by following a "useful interaction pattern"

2017-03-09 13:56:47 GMT <AFaust> There is no real benefit to lowering the txn cache sizes

2017-03-09 13:57:42 GMT <AFaust> txn cache sizes are just to limit the memory used in the worst case, and do not affect the minimum memory used

2017-03-09 13:58:58 GMT <AFaust> It is difficult to define generic rules for cache sizing - it obviously depends on how big your structures are, how much heap you are willing to allocate (keep in mind the GC overhead) and how your use case will fill/use the caches

2017-03-09 14:00:08 GMT <AFaust> In most customer scenarios I have come across, I would only modify the size of the shared caches that relate to nodes (entity, properties, aspects) - and these are derived from the size of heap I allow the JVM to allocate

2017-03-09 14:02:37 GMT <yregaieg> I was thinking the same

2017-03-09 14:03:07 GMT <yregaieg> these are generally the ones with the highest eviction rate

2017-03-09 14:03:56 GMT <yregaieg> is there a function you can share for suggesting caches sizes depending on JVM memory ?

2017-03-09 14:06:41 GMT <yregaieg> or is it more like a feeling thing supported with experience (and by experience I mean trial and error) ?

2017-03-09 14:11:41 GMT <AFaust> There are some basic assumptions you can make and use for a rough estimate, but in real life it depends on the complexity of your data model. E.g. how many aspects will each node have on average, how many properties, what structure will the values of these properties have...

2017-03-09 14:13:21 GMT <AFaust> I found the easiest way to find out what I could use as the basis of my size calculations is just to load a few tens of thousands of nodes into caches and use a heap analysis tool to see how much memory each cache was retaining. Then divide that by the number of entries for an average to use in estimations.

2017-03-09 14:17:08 GMT <_daniele_> good afternoon to all alfresco experts and not...

2017-03-09 14:18:07 GMT <_daniele_> I have a question, just a curiosity for sdk developers: why from sdk 2.2 has been abandoned spring loaded for hot reload ?

2017-03-09 14:26:30 GMT <angelborroy> _daniele_ because of lack of support from part of Spring

2017-03-09 14:26:47 GMT <angelborroy> it looks like there was many issues unresolved in Spring Loaded

2017-03-09 14:48:02 GMT <douglascrp> _daniele_, with the new SDK 3.0 you will be able to use another library for hot reload, hotswap

2017-03-09 14:48:21 GMT <douglascrp> _daniele_, http://hotswapagent.org/

2017-03-09 14:48:22 GMT <alfbot> Title: | HotswapAgent (at hotswapagent.org)

2017-03-09 14:49:01 GMT <douglascrp> _daniele_, if you want to try it, follow this guide https://docs.google.com/document/d/1p5QR1Q8pVvMT9fh_aID5nraD5NP6ZrU1ztcgE_zHR-4/pub

2017-03-09 14:49:02 GMT <alfbot> Title: Alfresco SDK 3.0 Beta - User Guide (at docs.google.com)

2017-03-09 14:49:15 GMT <douglascrp> and you can replace beta4 with beta5 now

2017-03-09 14:51:54 GMT <angelborroy> AFaust i’ve identified the problem with AOS

2017-03-09 14:52:17 GMT <angelborroy> AFaust special characters in URLs are not identified by Office

2017-03-09 14:52:43 GMT <angelborroy> AFaust for instance: https://server/alfresco/aos/Sitio/it-business-solutions/documentLibrary/Downstream/Activities/Proceso_Fin_de_A%c3%b1o_2016.docx

2017-03-09 14:52:56 GMT <angelborroy> AFaust any hint?

2017-03-09 14:53:17 GMT <AFaust> Hmm - I know of an issue with double-encoded characters in URL that affects a customer of mine

2017-03-09 14:53:54 GMT <angelborroy> Let me dig a little more, just to know if you’ve faced this before

2017-03-09 14:53:57 GMT <angelborroy> AFaust thanks

2017-03-09 14:55:19 GMT <angelborroy> AFaust definitively it’s not Alfresco related, it’s Apache related

2017-03-09 14:55:21 GMT <AFaust> angelborroy: My issue is https://issues.alfresco.com/jira/browse/ALF-21758

2017-03-09 14:55:24 GMT <yregaieg> AFaust: If I remember right, that was related to a particular Office version /cc angelborroy

2017-03-09 15:08:02 GMT <IanW1> AFaust: angelborroy: did you try the fix in https://issues.alfresco.com/jira/browse/ACE-5259

2017-03-09 15:08:54 GMT <AFaust> IanW1: I did use it to verify it was causing the problem

2017-03-09 15:09:30 GMT <AFaust> But together with customer decided to not use it in production - after all, the encodeURI() call was added to fix another issue and the customer didn't want to run risk of other issues

2017-03-09 15:10:06 GMT <AFaust> Luckily, Chrome is not a primary browser for that customer, so most users are not affected...

2017-03-09 15:12:31 GMT <bhagyas> angelborroy: Spring loaded has since come a long way

2017-03-09 15:12:34 GMT <bhagyas> ;)

2017-03-09 15:15:24 GMT <bhagyas> Instead of fixing the tag serialisation issue, they dumped spring loaded instead

2017-03-09 15:15:33 GMT <bhagyas> but its fine, since JRebel works really well with Alfresco

2017-03-09 15:17:13 GMT <angelborroy> it looks like a JK issue

2017-03-09 15:17:19 GMT <angelborroy> as using proxypass works

2017-03-09 15:29:17 GMT <_daniele_> douglascrp... thank you very very much!

2017-03-09 15:31:18 GMT <yregaieg> bhagyas: are you using jRebel ?

2017-03-09 15:31:34 GMT <yregaieg> how would you recommend it ?

2017-03-09 15:33:08 GMT <bhagyas> yregaieg: very much

2017-03-09 15:35:14 GMT <yregaieg> well it looks too expensive for personal investment

2017-03-09 15:35:38 GMT <yregaieg> I thought there were some community edition that was announced some time back

2017-03-09 15:35:58 GMT <yregaieg> maybe I got confused with something else

2017-03-09 15:43:57 GMT <iwkse> yregaieg: there's hotswap agent, http://hotswapagent.org/

2017-03-09 15:43:58 GMT <alfbot> Title: | HotswapAgent (at hotswapagent.org)

2017-03-09 15:44:26 GMT <yregaieg> iwkse: that does not support context reloading AFAIK

2017-03-09 15:45:02 GMT <iwkse> yregaieg: it doesn't support xml beans

2017-03-09 15:45:17 GMT <AFaust> angelborroy: In my case there is no JK - always using mod_proxy

2017-03-09 15:45:29 GMT <angelborroy> I know, it works with mod_proxy

2017-03-09 15:45:43 GMT <angelborroy> Trying to find right jkoptions config

2017-03-09 15:46:46 GMT <AFaust> And for the issue my customer is having it fails with mod_proxy as well, so that issue I linked is not related to the proxy at all

2017-03-09 15:47:01 GMT <angelborroy> yep

2017-03-09 15:47:14 GMT <angelborroy> it’s fixed for 201702 (which is the version I’m using)

2017-03-09 15:48:27 GMT <douglascrp> yregaieg, bhagyas I am using jrebel too, and it works really well

2017-03-09 16:11:32 GMT <alfbot> resplin: Sent 1 day, 7 hours, and 4 minutes ago: <yregaieg> /chanserv access list #alfresco says that fcorti is finally a mod on the channel ! And the topic is already updated ! Thanks !

2017-03-09 16:13:34 GMT <yregaieg> oops

2017-03-09 16:13:55 GMT <yregaieg> I did say that to alfbot in private

2017-03-09 16:16:43 GMT <AFaust> apparently not

2017-03-09 16:16:49 GMT <AFaust> bad alfbot, bad

2017-03-09 16:34:15 GMT <angelborroy> AFaust finally it was not related to modjk

2017-03-09 16:34:40 GMT <angelborroy> AFaust I forgot to include URIEncoding="UTF-8" at AJP connector in Tomcat

2017-03-09 16:34:47 GMT <AFaust> muaha

2017-03-09 16:34:53 GMT <angelborroy> I know, I know

2017-03-09 16:35:10 GMT <angelborroy> As I’ve installed 100 times Alfresco, I’m working on-the-fly

2017-03-09 16:35:22 GMT <angelborroy> But reading a manual is always worthy

2017-03-09 16:35:32 GMT <angelborroy> my own manual, I mean

2017-03-09 16:55:38 GMT <_daniele_> angelborroy thank you

2017-03-09 19:55:01 GMT <eswbitto> Is the cloud sync feature only for enterprise edition or does CE have it as well?

2017-03-09 19:55:13 GMT <eswbitto> Trying to tune out services that I don't need.

2017-03-09 19:55:28 GMT <douglascrp> eswbitto, I am almost sure that it's enterprise only

2017-03-09 19:56:25 GMT <eswbitto> douglascrp thanks! what about the eager creation of home folders "system.usages.enabled=false" I'm not really sure that that is.

2017-03-09 19:59:43 GMT <douglascrp> eswbitto, if I am not wrong, that is used when you have the LDAP sync enabled

2017-03-09 20:00:01 GMT <douglascrp> it will create all the user home folders when users are created

2017-03-09 20:00:18 GMT <douglascrp> if you left if as false, it will create the user home folder only when the user logs in

2017-03-09 20:01:02 GMT <douglascrp> and the usages enabled will keep control over the total amount of content each user can create inside alfresco

2017-03-09 20:01:23 GMT <douglascrp> I don't know if know about it, but you can set that in the user profile

2017-03-09 20:01:27 GMT <eswbitto> douglascrp ah ok....hmm....I don't see where that would impact server performance. Either way it would still need to be created.

2017-03-09 20:01:40 GMT <douglascrp> if you set it as false, it will ignore it

2017-03-09 20:10:25 GMT <eswbitto> douglascrp do you know if disabling the activities feeds and post will mess with site invitations?

2017-03-09 20:13:18 GMT <douglascrp> no

2017-03-09 20:13:22 GMT <douglascrp> those are different things

2017-03-09 20:13:26 GMT <douglascrp> no related

2017-03-09 20:13:34 GMT <eswbitto> thanks

2017-03-09 20:14:53 GMT <eswbitto> douglascrp have you ever disabled the activities feed? If you have was there any significant changes on server performance?

2017-03-09 20:15:20 GMT <douglascrp> eswbitto, I disabled it once, but not to check the performance

2017-03-09 20:15:30 GMT <douglascrp> I did that only to eliminate the annoying emails

2017-03-09 20:15:37 GMT <eswbitto> ok

2017-03-09 20:15:45 GMT <douglascrp> well, that is what the users thought about the emails :D

2017-03-09 20:15:51 GMT <eswbitto> lol

2017-03-09 20:15:53 GMT <douglascrp> so I have no personal opinion about that

2017-03-09 20:16:00 GMT <douglascrp> they asked me to disable it, and I did so

2017-03-09 20:18:52 GMT <eswbitto> douglascrp maybe you might have gone through this. So I'm using zabbix to monitor server performance and I can graph out a time period of memory usage. Well couple of weeks back alfresco became non responsive and I looked and it was out of heap space.

2017-03-09 20:19:03 GMT <eswbitto> I restarted the service and all was well

2017-03-09 20:19:50 GMT <eswbitto> Then it happened again about a week later....I looked at graphs and from the time that I restart the server to another restart the memory gradually gets used and it never really frees up...or I don't ever see a drop in usage.

2017-03-09 20:19:54 GMT <douglascrp> haven't you noticed the reason?

2017-03-09 20:20:24 GMT <eswbitto> So I'm trying to disable non-essential services to try and determine what is keeping the memory.

2017-03-09 20:20:29 GMT <douglascrp> the ideal would be to extract a thread dump when the memory is almost completely used

2017-03-09 20:20:44 GMT <douglascrp> do you have any custom module installed?

2017-03-09 20:21:17 GMT <eswbitto> I'm going to put the javamelody on production. I was playing around with it in staging and got a working authentication going on the webpage.

2017-03-09 20:21:39 GMT <douglascrp> try the ootb support-tools instead

2017-03-09 20:21:51 GMT <douglascrp> you will have the memory usage graph available

2017-03-09 20:22:01 GMT <douglascrp> and you will be able to extract the thread dump

2017-03-09 20:22:06 GMT <eswbitto> ok

2017-03-09 20:22:14 GMT <douglascrp> do you know what I mean?

2017-03-09 20:22:20 GMT <douglascrp> do you know the tool?

2017-03-09 20:22:25 GMT <eswbitto> no, but I was going to look it up.

2017-03-09 20:23:05 GMT <douglascrp> http://www.zylk.net/en/web-2-0/blog/-/blogs/ootb-support-tools-addon-for-alfresco-community

2017-03-09 20:23:06 GMT <alfbot> Title: - OOTB Support Tools addon for Alfresco Community - zylk (at www.zylk.net)

2017-03-09 20:23:25 GMT <douglascrp> the project`s page https://github.com/OrderOfTheBee/ootbee-support-tools

2017-03-09 20:23:26 GMT <alfbot> Title: GitHub - OrderOfTheBee/ootbee-support-tools: OOTBee Support Tools addon to extend set of administrative tools on Repository- and Share-tier (at github.com)

2017-03-09 20:23:40 GMT <eswbitto> oh cool

2017-03-09 20:24:15 GMT <douglascrp> I believe that one will help you

2017-03-09 20:24:31 GMT <eswbitto> wow....looking this over.....thanks so much!

2017-03-09 20:24:52 GMT <douglascrp> let us know what you think

2017-03-09 20:25:02 GMT <eswbitto> did you help develop it?

2017-03-09 20:25:08 GMT <douglascrp> no

2017-03-09 20:25:20 GMT <douglascrp> I do not have the required skills for that :P

2017-03-09 20:25:26 GMT <eswbitto> lol

2017-03-09 20:25:33 GMT <douglascrp> the only thing I did was to translate it into portuguese

2017-03-09 20:25:48 GMT <douglascrp> AFaust started it, and other guys helped

2017-03-09 20:25:54 GMT <douglascrp> thank the magicians

2017-03-09 20:25:58 GMT <douglascrp> not me

2017-03-09 20:26:07 GMT <eswbitto> I will do so

2017-03-09 20:26:19 GMT <douglascrp> the wizards :D

2017-03-09 20:27:01 GMT * AFaust prefers "Jedi"

2017-03-09 20:27:09 GMT <douglascrp> AFaust, uhauhauhauhauha

2017-03-09 20:27:27 GMT <eswbitto> The force is strong in this one.

2017-03-09 20:27:51 GMT <AFaust> There is a reason why I (._.) ~ ︵ ┻━┻ instead of (╯°□°)╯︵ ┻━┻ when I am displeased

2017-03-09 20:29:22 GMT <AFaust> eswbitto: Either way, feel free to try it out. We are currently working on the documentation in the wiki of the project about all the tools we already have, and as you may see have already quite a bit of a feature / improvements idea list

2017-03-09 20:30:28 GMT <eswbitto> AFaust I'll take a look. Wow...this looks fantastic.

2017-03-09 20:31:54 GMT <eswbitto> AFaust quick question....can this be restricted to the Alfresco_admin group?

2017-03-09 20:35:44 GMT <AFaust> It already is

2017-03-09 20:36:08 GMT <AFaust> All the tools are implemented as web scripts that have the <authentication>admin</authentication> tag set

2017-03-09 20:37:25 GMT <AFaust> This also applies to the web scripts that only provide data - if you find any tool that is accessible by a non-admin, please file an issue in the project.

2017-03-09 20:38:28 GMT <eswbitto> AFaust duly noted.

2017-03-09 20:42:37 GMT <eswbitto> AFaust so to create the amp files I'll need to put this in my SDK?

2017-03-09 20:44:06 GMT <AFaust> The module is available from Maven Central - you can simply add dependencies / overlays to your existing AIO project and it will fetch the pre-built release version

2017-03-09 20:44:09 GMT <AFaust> See http://search.maven.org/#search|ga|1|org.orderofthebee.support-tools

2017-03-09 20:44:10 GMT <alfbot> Title: The Central Repository Search Engine (at search.maven.org)

2017-03-09 20:44:51 GMT <eswbitto> coolbeans

2017-03-09 21:10:48 GMT <douglascrp> later guys

2017-03-09 22:02:22 GMT *** warlock is now known as Guest20019

2017-03-09 22:02:35 GMT <Guest20019> hallo alle zusammen. eine Frage spricht einer deutsch und kann mir bei Alfresco Community Edition helfen?

2017-03-09 22:04:31 GMT <Guest20019> keiner lust und zeit?

2017-03-09 22:06:20 GMT <Guest20019> bb & gn8

2017-03-09 22:48:34 GMT <resplin> AFaust: You are probably about to head to bed.

2017-03-09 22:48:37 GMT <resplin> Two follow up items.

2017-03-09 22:49:01 GMT <AFaust> Not yet - it isn't 1am yet, the earliest time I have consistently made it to bed...

2017-03-09 22:49:26 GMT <resplin> That is my preferred schedule, but I haven't been able to enjoy it since starting to work with the European team at Alfresco.

2017-03-09 22:49:46 GMT <resplin> I discussed with the Engineering Architects compatibility with 4-byte Unicode. They did not feel like it is a priority, and we haven't had any customer requests for it. So it isn't worth filing any additional issues there.

2017-03-09 22:50:23 GMT <resplin> However, we expect to have most of that code in GitHub soon, and would take pull requests fixing these sorts of problems (assuming tests etc)

2017-03-09 22:50:33 GMT <AFaust> Then I'll have to look forward to any contribution / collaboration improvements you may announce at BeeCon, so I can maybe fix this myself in a couple of hours...

2017-03-09 22:51:06 GMT <resplin> I'm hoping you will be able to ahead of BeeCon, and then we can tell attendees at BeeCon about the experience.

2017-03-09 22:51:20 GMT <AFaust> That would be even better...

2017-03-09 22:51:22 GMT <resplin> But I did want to make sure we discussed it before you bothered doing anything.

2017-03-09 22:51:25 GMT <resplin> Or anything more.

2017-03-09 22:51:37 GMT <AFaust> And not just because then you already have initial feedback to incorporate into any questions at BeeCon

2017-03-09 22:52:00 GMT <resplin> Agreed

2017-03-09 22:52:08 GMT <resplin> Second item is ALF-21854.

2017-03-09 22:52:24 GMT <resplin> We weren't clear on how to reproduce it. We have fixed those types of problems in the past, and aren't sure where exactly you are seeing them.

2017-03-09 22:52:47 GMT <resplin> So I set it to Need Info and assigned it to you, but I noticed that you assigned it back but I don't see a comment.

2017-03-09 22:52:56 GMT <AFaust> I have updated the description itself

2017-03-09 22:53:01 GMT <resplin> Ah, I see that now.

2017-03-09 22:53:11 GMT <resplin> Thank you.

2017-03-09 22:53:33 GMT <resplin> At first I thought that I had forgotten to assign it to you, so I assigned it back. Then I realized that I should talk to you about it.

2017-03-09 22:54:27 GMT <AFaust> To be honest, when engineers have reported in the past that they have "fixed" those types of problems, they probably meant they have spent "the least amount of effort to build the most minimalistic workaround or facade to cure the symptoms"

2017-03-09 22:54:30 GMT <resplin> Issue is updated, and we'll review it next week.

2017-03-09 22:55:20 GMT <resplin> Interesting. It's on the boundary between Share and Repo. I don't know which team should get it.

2017-03-09 22:55:23 GMT <resplin> I'll look into that.

2017-03-09 22:55:25 GMT <resplin> Thanks.

2017-03-09 22:55:43 GMT <AFaust> The issue is solely on the Repo-tier.

2017-03-09 22:55:57 GMT <AFaust> The Share-tier has only been able to hide the extent of the issue since forever...

2017-03-09 22:56:11 GMT <yreg> resplin, AFaust that's really cool

2017-03-09 22:56:57 GMT <yreg> (I meant alfresco finally being open to pull request)

2017-03-09 22:57:14 GMT <yreg> btw, sorry for bumping in in the middle of the discussion

2017-03-09 22:57:35 GMT <resplin> yreg: No apology needed.

2017-03-09 22:57:40 GMT <resplin> AFaust: Thanks for the clarification.

2017-03-09 22:57:59 GMT <yreg> AFaust, may I annoy you a bit more about caches <grin>

2017-03-09 22:58:23 GMT <resplin> yreg: We are getting closer to having the tooling in place to accept pull requests, but we still have a number of tooling and organizational challenges

2017-03-09 22:58:27 GMT * yreg realizes he has been a bit annoying with the amount he has been asking around related to caches

2017-03-09 22:58:45 GMT <resplin> We need to get our test framework public so people can write tests, and set up some sort of CI build on pull requests.

2017-03-09 22:58:49 GMT <AFaust> yreg: Go ahead - who would you ask otherwise for such arcane information?

2017-03-09 22:59:06 GMT <yreg> resplin, nothing comes easy, but believe me, it is more than worth it

2017-03-09 22:59:09 GMT <resplin> And we need to organizationally prioritize the effort involved with reviewing, incorporating, and supporting them.

2017-03-09 22:59:24 GMT <resplin> But we are finally making tangible progress.

2017-03-09 22:59:47 GMT <resplin> yreg: I completely agree. I just don't want people to get too excited just yet.

2017-03-09 22:59:49 GMT <yreg> thanks Axel

2017-03-09 23:00:01 GMT <yreg> and resplin

2017-03-09 23:00:11 GMT <resplin> I can't help with questions about caches.

2017-03-09 23:00:30 GMT <resplin> I had an engineer explain to me today that i don't actually know what I'm talking about in that area.

2017-03-09 23:01:16 GMT <yreg> -= THIS MESSAGE NOT LOGGED =-

2017-03-09 23:01:20 GMT <AFaust> resplin: I just hope that news about improvements with regards to contribution process won't be offset by completely bonkers announcements regarding Share / UI product strategy between now up to and including BeeCon.

2017-03-09 23:01:41 GMT * AFaust is kind of dreading that aspect of BeeCon...

2017-03-09 23:01:44 GMT <yreg> I meant about the contribution process

2017-03-09 23:01:50 GMT <resplin> Thanks yreg

2017-03-09 23:01:57 GMT <resplin> AFaust is likely to be disappointed.

2017-03-09 23:02:12 GMT <resplin> But I don't have as much input in that area.

2017-03-09 23:02:32 GMT <yreg> hehe

2017-03-09 23:03:16 GMT <AFaust> resplin: ahh - do you want to give me an aneurysm? Just because I am dreading something does not mean you should immediately confirm/spoil that....

2017-03-09 23:03:58 GMT <yreg> AFaust, back to the caches, I was wondering if you had any idea about how those local manual caches work, and how to ties them to the cache the belong to

2017-03-09 23:03:59 GMT <AFaust> ...or even just in the slightest hint at something I might not like...

2017-03-09 23:04:23 GMT <resplin> I hear that Germans have great healthcare.

2017-03-09 23:05:36 GMT <yreg> resplin, if they do, I bet it is nothing compared to Finland, Norway, Sweden and Denmark

2017-03-09 23:05:37 GMT <AFaust> Compared to the mess that US has everything a 2nd world country has should be considered great healthcare

2017-03-09 23:05:56 GMT <yreg> good point there

2017-03-09 23:06:01 GMT <resplin> And you are worried about a little aneurysm?

2017-03-09 23:06:29 GMT <resplin> The US philosophy is to die young before you need any pension money!

2017-03-09 23:06:34 GMT <AFaust> You know, with an aneurysm it's like with buying property: location, location, location.

2017-03-09 23:06:42 GMT <resplin> lol

2017-03-09 23:06:45 GMT <AFaust> Can be significant complications if in the wrong place

2017-03-09 23:06:48 GMT <resplin> I had never thought about it that way, and it is so true.

2017-03-09 23:07:37 GMT <resplin> Though my attempt at humor was based on the assumption that it is always complicated.

2017-03-09 23:07:43 GMT <AFaust> Ok - yreg, might have to catch me up again. "Those local manual caches" - what exactly do you mean by that?

2017-03-09 23:08:03 GMT * resplin goes back to his email

2017-03-09 23:08:06 GMT <AFaust> We did talk about the double-caching of the node cache yesterday (or the day before)

2017-03-09 23:08:35 GMT <AFaust> and the issue with loading too much data in your CSV job

2017-03-09 23:08:40 GMT <yreg> Earlier today

2017-03-09 23:09:19 GMT <yreg> ah I forgot, we are @ 00:09 already

2017-03-09 23:10:00 GMT <yreg> ok, DefaultSimpleCache has a member caches of that type

2017-03-09 23:10:21 GMT <AFaust> Ah - you were referring to the name of the Java class member "localCache"

2017-03-09 23:10:58 GMT <yreg> com.google.common.cache.LocalCache$LocalManualCache

2017-03-09 23:11:02 GMT <yreg> yes indeed

2017-03-09 23:11:26 GMT <yreg> at the end of the tree I find multiple segments

2017-03-09 23:11:42 GMT <yreg> with no indication what so ever about type of content

2017-03-09 23:11:42 GMT <AFaust> Ok - DefaultSimpleCache is basically just the Alfresco API class for a cache. Anything we show in OOTBee Support Tools - Caches view is a DefaultSimpleCache (actually, an instance of the SimpleCache API interface)

2017-03-09 23:12:24 GMT <AFaust> Since Alfresco 5.0 they are using a Google cache implementation internally - before that it was a Google ConcurrentLRUMap...

2017-03-09 23:12:58 GMT <AFaust> It is best to only look at the DefaultSimpleCache instance and its "cacheName" member

2017-03-09 23:13:06 GMT <yreg> I did what you suggested earlier (disable batch loading) and the caches at ibatis level shrinked to half, still occupy 1.3 Gb hoxever

2017-03-09 23:13:07 GMT <AFaust> This should give you the name of the cache

2017-03-09 23:13:40 GMT <yreg> Can I do that from a dump ?

2017-03-09 23:13:54 GMT <AFaust> Yes

2017-03-09 23:14:59 GMT <AFaust> Regarding the ibatis cache, that was expected - all the pre-load selects are gone and no longer cached, but you likely still have DB FTS selects and anything related to ACL checking that may be cached in the outer transaction.

2017-03-09 23:16:09 GMT <AFaust> You should also be able to identify the specific SQL queries that are used as keys in the ibatis caches from the dump, so you can learn what sort of selects that are...

2017-03-09 23:16:10 GMT <yreg> found it

2017-03-09 23:16:32 GMT <yreg> That one with 1.8 Gb was for node properties cache

2017-03-09 23:16:50 GMT <AFaust> That is always the biggest

2017-03-09 23:17:28 GMT <yreg> DB FTS, If I specify in my search param that I want to use solr, Would it still go for DB FTS ?

2017-03-09 23:17:52 GMT <AFaust> If you explicitly set queryConsistency to EVENTUAL, it will always go for SOLR

2017-03-09 23:18:13 GMT <AFaust> But if it is the default and the query is DB FTS-executable, then it will go to DB

2017-03-09 23:18:23 GMT <yreg> nice, that should give me extra room

2017-03-09 23:19:28 GMT <yreg> I thought once I preceeds a column/key with @ it always goes to solr

2017-03-09 23:19:33 GMT <yreg> doesn't it ?

2017-03-09 23:20:51 GMT <AFaust> @ has nothing to do with forcing SOLR

2017-03-09 23:21:17 GMT <AFaust> @ is also completely unnecessary in general - it is a leftover from the low-level Lucene

2017-03-09 23:23:15 GMT <AFaust> The important factors are: property type / index configuration and presence of exact match / term expansion operator

2017-03-09 23:24:07 GMT <yreg> it's just listing all items of one root type

2017-03-09 23:24:08 GMT <AFaust> e.g. for most d:text properties unless you prefix the column/key with = it cannot go to DB

2017-03-09 23:24:24 GMT <AFaust> So a simple TYPE/ASPECT query then?

2017-03-09 23:24:35 GMT <yreg> yes

2017-03-09 23:24:40 GMT <AFaust> Definitely DB FTS compatible

2017-03-09 23:24:58 GMT <AFaust> and there is no way to force SOLR with rewriting the query in this case

2017-03-09 23:25:35 GMT <AFaust> unless you include another field, e.g. like a dummy cm:name:*

2017-03-09 23:26:14 GMT <AFaust> or just PROPERTIES:"{http://www.alfresco.org/model/content/1.0}name"

2017-03-09 23:26:51 GMT <AFaust> ^^ simply selects all items that have a name set - which is basically everything, so it does not restrict your base query if you combine it with AND, but it forces SOLR

2017-03-09 23:27:31 GMT <yreg> Thanks, will definitely give it a run and see how it goes with my caches

2017-03-09 23:27:56 GMT <yreg> I assume you would be interested in such feedback for your talk around caches ;()

2017-03-09 23:28:13 GMT <yreg> $;()

2017-03-09 23:28:24 GMT <yreg> *;-)

2017-03-09 23:29:00 GMT <yreg> I am so tired, it took me 3 times to get the thing right... need to go get some rest

2017-03-09 23:29:22 GMT <AFaust> anything is appreciated - even if only to serve as an anecdote

2017-03-09 23:29:23 GMT <yreg> Thanks again Axel, you've been as instructive as always

2017-03-09 23:30:21 GMT <yreg> one last thing before I go to bed

2017-03-09 23:30:28 GMT <yreg> since I still remember

2017-03-09 23:30:55 GMT <yreg> do you have any particular recommendation/tip for tuning MS SQL for Alfresco

2017-03-09 23:31:36 GMT <yreg> ("DO NOT USE IT !" a part)

2017-03-09 23:32:03 GMT <yreg> will check the logs tomorrow

2017-03-09 23:32:07 GMT <yreg> have a good night

End of Daily Log

The other logs are at http://esplins.org/hash_alfresco