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

2017-02-22 00:09:08 GMT <alfbot> resplin: Sent 1 day, 12 hours, and 32 minutes ago: <angelborroy> : https://issues.alfresco.com/jira/browse/ALF-21833 what about Community resolution? It’s marked as “Fix Version/s: None”

2017-02-22 00:13:51 GMT <grendal_prime> ok i cannot seem to get this libre office and alfresco integration to do squat

2017-02-22 00:14:39 GMT <grendal_prime> squat...being the technical term of squat = anything close to the desired effect.

2017-02-22 00:14:48 GMT <resplin> grendal_prime: Which one are you playing with?

2017-02-22 00:14:56 GMT <grendal_prime> 50d

2017-02-22 00:15:06 GMT <resplin> Which libreoffice integration?

2017-02-22 00:15:30 GMT <grendal_prime> the one i have tried needs java app and well chrome does not supprot nappi any more

2017-02-22 00:15:43 GMT <resplin> Which one is that?

2017-02-22 00:15:51 GMT <grendal_prime> https://addons.alfresco.com/addons/libreoffice-online-edit-module

2017-02-22 00:15:52 GMT <alfbot> Title: LibreOffice Online Edit Module | Alfresco Add-ons - Alfresco Customizations (at addons.alfresco.com)

2017-02-22 00:16:46 GMT <grendal_prime> ive got management working with this finally...but the process of downloading and uploading is making them want to kill me.

2017-02-22 00:16:57 GMT <resplin> I believe that this one has had some recent attention:

2017-02-22 00:17:02 GMT <resplin> https://addons.alfresco.com/addons/edit-online-libreoffice-openoffice

2017-02-22 00:17:03 GMT <alfbot> Title: Edit online with Libreoffice / Openoffice | Alfresco Add-ons - Alfresco Customizations (at addons.alfresco.com)

2017-02-22 00:17:49 GMT <resplin> It's also worth looking at LibreOffice's ability to browse CMIS repositories. That lets you integrate with Alfresco without any add-on.

2017-02-22 00:18:48 GMT <grendal_prime> "the cmis browser blows" i think was the exact phrase that was used

2017-02-22 00:19:11 GMT <resplin> I'm glad you are already aware of it.

2017-02-22 00:19:23 GMT <grendal_prime> thing is if it used cimis to launch from the share than that would make sence.

2017-02-22 00:20:10 GMT <resplin> I'm curious to hear what you think of the Zylk module.

2017-02-22 00:25:57 GMT <grendal_prime> i need to compile it i believe.

2017-02-22 00:30:02 GMT <grendal_prime> is there a amp for it somewhere

2017-02-22 00:37:51 GMT <grendal_prime> ok ill compile it...or try to anyway

2017-02-22 00:58:05 GMT <grendal_prime> ok found the jar i can just add to tomcat. The frustrating part...would not be a big deal to just use the smb server for there editing, however it seems to be two slow.

2017-02-22 01:00:15 GMT <grendal_prime> sorry..what i mean to say is, there are a few windows users, however most of these guys are using ubuntu linux and i just have it mount up the smb shares from our filer. although i cannot get the alfresco smb server to auto mount they could at least click on a desktop launcher. but so far every time i have tried to open a document via alfrescos smb server it runs terribly slow.

2017-02-22 01:01:56 GMT <grendal_prime> the smb method of connecting to alfresco would be these guys preferred method of connecting...as they are all from a windows background and seem to understand smb shares fairly well.

2017-02-22 01:06:46 GMT <grendal_prime> ya zilk method complains about java in my browser as well

2017-02-22 01:15:49 GMT <grendal_prime> ya thats from 2003 it does not work with current version of chrom

2017-02-22 01:15:58 GMT <grendal_prime> sorry 2013

2017-02-22 01:35:28 GMT <grendal_prime> none of these work

2017-02-22 01:41:52 GMT <grendal_prime> im going to have to figure out why the smb server is acting so sketch

2017-02-22 01:50:27 GMT <grendal_prime> resplin ...this is the exact error i get from the Java zylk module (Java must be enabled in your browser to use the "Edit in LibreOffice" functionality.)

2017-02-22 07:23:21 GMT <bhagyas> WTF of the day: enterpriseRetryExceptions in RetryingTransactionHelper.java:143

2017-02-22 07:23:45 GMT <bhagyas> Would love to see an explanation why Enterprise needs a different exception than Community

2017-02-22 07:36:04 GMT <yreg> Morning guys !

2017-02-22 09:31:36 GMT <bhagyas> AFaust: Any known methods to invalidate the parent association caches?

2017-02-22 09:31:42 GMT <bhagyas> :/

2017-02-22 09:32:05 GMT <AFaust> You mean that single crappy cache that is not a real, proper cache?

2017-02-22 09:32:54 GMT <bhagyas> yup

2017-02-22 09:32:56 GMT <bhagyas> exactly that

2017-02-22 09:33:10 GMT <AFaust> You can only clear it along with all the other caches in the NodeDAO

2017-02-22 09:33:12 GMT <bhagyas> I've already used all known expletives, so can't insult it anymore

2017-02-22 09:33:20 GMT <bhagyas> xD

2017-02-22 09:33:52 GMT <AFaust> The NodeBulkLoader interface it implements has a "clear()" operation "FOR TESTING ONLY"

2017-02-22 09:34:15 GMT <bhagyas> urgh!

2017-02-22 09:34:16 GMT <bhagyas> :/

2017-02-22 09:34:37 GMT <bhagyas> It's not even thread safe

2017-02-22 09:34:55 GMT <AFaust> Thread safe is one thing - transactionally safe is another

2017-02-22 09:35:07 GMT <AFaust> Or "not being" any of these

2017-02-22 09:35:10 GMT <bhagyas> so good luck if you're iterating the parentAssocs while requesting parent Assocs

2017-02-22 09:35:14 GMT <bhagyas> yeah :|

2017-02-22 09:37:16 GMT <bhagyas> Any way to get around this crap?

2017-02-22 09:40:00 GMT <bhagyas> AFaust: What happens if you wrap it inside a retryingTranasactionHelper?

2017-02-22 09:40:51 GMT <AFaust> It does not matter - the implementation is completely unaware of transactions - it is a purely in-memory data structure based on regular HashMap instances

2017-02-22 09:41:06 GMT <AFaust> But the cache internally uses a Java ReadWriteLock

2017-02-22 09:41:32 GMT <AFaust> So at least the clear() operation is atomic for all other code that uses the cache

2017-02-22 09:41:41 GMT <bhagyas> fcorti: Do you have any good idea of this?

2017-02-22 09:42:42 GMT <bhagyas> AFaust: Looks like the NodeBulkLoader clears out everything

2017-02-22 09:42:50 GMT <AFaust> I said it would

2017-02-22 09:43:08 GMT <bhagyas> yeah, thought there would be an option to clear 'just' the parent cache

2017-02-22 09:43:14 GMT <bhagyas> what about using FileFolderService?

2017-02-22 09:43:19 GMT <bhagyas> to fetch the parents

2017-02-22 09:43:21 GMT <bhagyas> :|

2017-02-22 09:43:21 GMT <AFaust> Nope - not unless you use reflection to retrieve the private field

2017-02-22 09:44:15 GMT <bhagyas> FFS- ffs xD

2017-02-22 09:44:16 GMT <AFaust> I believe the FileFolderService offers no operations regarding parents...

2017-02-22 09:45:04 GMT <bhagyas> hmpf.. yeah

2017-02-22 09:45:13 GMT <bhagyas> I remembered a listFolders method

2017-02-22 09:45:19 GMT <bhagyas> but that goes downwards

2017-02-22 09:45:57 GMT <AFaust> Also, I ignore it completely 99% of the time. It is one of the services I actually despise because it seriously and unecessarily restricts some use cases if used, and I wish Alfresco would use it less for some of their core functionality.

2017-02-22 09:46:39 GMT <fcorti> bhagyas not yet... if you open a discussion in the community portal I can try to involve engineers from the platform for a discussion

2017-02-22 09:46:39 GMT <bhagyas> I kinda like that FFS takes care of permissions and some lookups

2017-02-22 09:46:57 GMT <bhagyas> fcorti: Is this not a community portal?

2017-02-22 09:46:58 GMT <bhagyas> ;)

2017-02-22 09:47:05 GMT <bhagyas> hehe

2017-02-22 09:47:12 GMT <fcorti> I feel like it is not a reply like "do this" but instead of a discussion

2017-02-22 09:47:37 GMT <bhagyas> ah

2017-02-22 09:47:39 GMT <bhagyas> :)

2017-02-22 09:47:46 GMT <fcorti> nope, this is not a community portal... and you know it ;-)

2017-02-22 09:47:49 GMT <bhagyas> AFaust: What about the getPath method?

2017-02-22 09:48:07 GMT <bhagyas> I have a feeling that it might skip over the parentCaches..but again you might only end up with just the primary Parent

2017-02-22 09:48:29 GMT <AFaust> bhagyas: What permissions does it take care of? And with regards to lookups, sure it may help with trivial path lookups but IMHO you should rarely depend on path-based resolution/lookups

2017-02-22 09:48:59 GMT <bhagyas> AFaust: Agreed, but Alfresco is far from a perfect implementation of a good node service

2017-02-22 09:49:18 GMT <bhagyas> that can give proper path based resolution

2017-02-22 09:49:30 GMT <AFaust> getPath definitely uses the cache

2017-02-22 09:49:35 GMT <AFaust> That is one of the primary use cases

2017-02-22 09:49:50 GMT <bhagyas> getPath is overridden by a number of classes

2017-02-22 09:50:08 GMT <bhagyas> and difficult to know which one is actually applied at a given time

2017-02-22 09:50:11 GMT <AFaust> And again: I usually don't care about paths so rarely have a need to resolve them...

2017-02-22 09:50:32 GMT <bhagyas> AFaust: If you have to get ParentAssocs in a thread safe way, what would you do?

2017-02-22 09:50:32 GMT <AFaust> I meant NodeService.getPath()

2017-02-22 09:50:43 GMT <bhagyas> (and transactionally safe)

2017-02-22 09:51:01 GMT <bhagyas> (apart from re-writing Alfresco)

2017-02-22 09:51:41 GMT <bhagyas> Right now, I've hit a point where the whole solution depends on this freaking associationCache

2017-02-22 09:51:58 GMT <bhagyas> *parentAssociationCache

2017-02-22 09:52:33 GMT <bhagyas> and guess what, its all in a readonly transaction as well

2017-02-22 09:52:35 GMT <bhagyas> :/

2017-02-22 09:52:37 GMT <AFaust> Yeah - I have not realized before that you can poison your parallel operations with it...

2017-02-22 09:53:08 GMT <AFaust> The only way to ensure you have transactionally safe parent associations is by ensuring each node has more than 2000 parents

2017-02-22 09:54:11 GMT <AFaust> But the parentAssociationCache should use the node version (DB column, not actual version) to avoid inconsistencies / stale data

2017-02-22 09:55:12 GMT <AFaust> And the cache key even has the changeTxnId of the current transaction as a differentiator

2017-02-22 09:55:29 GMT <AFaust> So I don't see exactly where your problem would be...

2017-02-22 09:56:57 GMT <bhagyas> yeah

2017-02-22 09:57:19 GMT <bhagyas> for some reason its returning in random

2017-02-22 10:05:04 GMT <AFaust> random order? or random data?

2017-02-22 10:07:38 GMT <bhagyas> random order

2017-02-22 10:07:42 GMT <bhagyas> and data as well

2017-02-22 10:34:13 GMT <DarkStar1> Morning everyone

2017-02-22 10:43:54 GMT <DarkStar1> Is the site name or relative url available as an object to doclib action config?

2017-02-22 10:47:40 GMT <DarkStar1> nevermind. I found a way around it :)

2017-02-22 10:47:42 GMT <Loftux> DarkStar1: Yes (but you probably need more info, you can get that if you say from where you need it;)

2017-02-22 10:48:03 GMT <Loftux> So I can keep my knowledge !

2017-02-22 10:53:22 GMT <DarkStar1> Loftux: I need to navigate back to a previous page and was going to build the url, but I remembered that the document.referrer has what I need :)

2017-02-22 10:56:31 GMT <Loftux> DarkStar1: If you need it on client side js, also check out Alfresco.constants, has lots of useful stuff.

2017-02-22 10:57:11 GMT <DarkStar1> Loftux: Yes only on the client side. I keep forgetting about that cache of gems :)

2017-02-22 10:57:21 GMT <DarkStar1> property gems*

2017-02-22 10:57:28 GMT <DarkStar1> properties*

2017-02-22 11:13:50 GMT <Hiten> Hello, I am facing an issue regarding the startup of alfresco 5.1 community instance

2017-02-22 11:14:03 GMT <Hiten> Here is the catalina.out logs http://pastebin.com/pBJ7h4Rg

2017-02-22 11:14:04 GMT <alfbot> Title: catalina.out - Pastebin.com (at pastebin.com)

2017-02-22 11:15:03 GMT <Hiten> I cannot see any exception other than the tracking failed error of solr

2017-02-22 11:16:03 GMT <Hiten> can anyone please let me know why I might me facing this issue

2017-02-22 11:18:27 GMT <Hiten> Here are the solr logs

2017-02-22 11:18:28 GMT <Hiten> http://pastebin.com/CdqC7vrs

2017-02-22 11:18:29 GMT <alfbot> Title: solr.log - Pastebin.com (at pastebin.com)

2017-02-22 13:09:06 GMT <yreg> ~since

2017-02-22 13:09:06 GMT <alfbot> yreg: Either yreg didn't leave, or no messages were sent while yreg was gone.

2017-02-22 13:33:42 GMT <DarkStar1> Aarg!! I have the folowing defined in an ftl href="${url.context}/res/dk/magenta/..... and this resolves to domain-name/share/res/share/dk/magenta/...

2017-02-22 13:34:15 GMT <DarkStar1> does res not resolve to the app context anymore?

2017-02-22 13:42:26 GMT <AFaust> DarkStar1: ?? It never has, AFAIK

2017-02-22 13:45:29 GMT <DarkStar1> AFaust: ${url.context}/res/components/dashlets/docsummary.js would suggest otherwise I don't know why it's adding the extra "share" in my webscript

2017-02-22 13:47:35 GMT <AFaust> because you use ${url.context}

2017-02-22 13:47:50 GMT <AFaust> ah - now I see it - you mean the /share after /res

2017-02-22 13:48:13 GMT <DarkStar1> yes

2017-02-22 13:48:16 GMT <AFaust> What FTL directive do you use?

2017-02-22 13:48:35 GMT <AFaust> Had something similar for a short time recently

2017-02-22 13:48:56 GMT <DarkStar1> <@markup id="js">

2017-02-22 13:48:56 GMT <DarkStar1> <#-- JavaScript Dependencies -->

2017-02-22 13:48:56 GMT <DarkStar1> <@script type="text/javascript" src="${url.context}/res/dk/magenta/lool/libreOffice-online-edit.js" group="lool"/>

2017-02-22 13:48:58 GMT <DarkStar1> </@>

2017-02-22 13:49:25 GMT <AFaust> ok - did you know that the entire prefix up to an including /res is actually not necessary and sort-of deprecated?

2017-02-22 13:49:35 GMT <DarkStar1> Funny thing is I havenNope

2017-02-22 13:49:38 GMT <DarkStar1> Nope

2017-02-22 13:50:04 GMT <DarkStar1> This I didn't know

2017-02-22 13:50:35 GMT <AFaust> Source comment: "Ideally this directive should be used without the application context, servlet mapping, controller mapping or URL rewrite mapped path"

2017-02-22 13:51:01 GMT <AFaust> So just start from dk/

2017-02-22 13:51:19 GMT <DarkStar1> When was this deprecated

2017-02-22 13:51:25 GMT <DarkStar1> Pre 5.0.d?

2017-02-22 13:51:48 GMT <DarkStar1> because then I have really not been paying attention :/

2017-02-22 13:54:47 GMT <AFaust> I said "sort-of deprecated" - there is no specific mention of this in any documentation that I know of

2017-02-22 13:55:17 GMT <AFaust> It's just that I actually make use of the openess of the source and read it sometimes

2017-02-22 13:55:34 GMT <AFaust> And then find comments like these...

2017-02-22 13:56:54 GMT <DarkStar1> Ok. I need to look up this same class for the 5.0.d source

2017-02-22 13:56:56 GMT <AFaust> I mean, even Alfresco engineers don't seem to be aware of this because most Share (and RM) code still include the url.context + res

2017-02-22 13:57:08 GMT <DarkStar1> Thanks though

2017-02-22 13:57:32 GMT <AFaust> It should work with these as well, so it likely is not related at all to the cause of your issue

2017-02-22 14:05:12 GMT <yreg> fcorti, http://stackoverflow.com/a/22980127/2307988

2017-02-22 14:05:13 GMT <alfbot> Title: sql - what is the maximum length of varchar(n) in postgresql 9.2 and which is best to use varchar(n) or text? - Stack Overflow (at stackoverflow.com)

2017-02-22 14:07:12 GMT <fcorti> yreg, ok, which is the message behind this? :-)

2017-02-22 14:12:15 GMT <bmejias> hi fellows.... I need some ideas to debug

2017-02-22 14:12:58 GMT <bmejias> there is some code deployed by consultants that would send an email to a user that gets assigned a document.

2017-02-22 14:13:40 GMT <bmejias> I see the debug message on the logs: DEBUG [action.executer.MailActionExecuter] [defaultAsyncAction3] {template=workspace://SpacesStore/96ee627e-b76f-49d1-8c31-5f110e835424, subject=, cc_many=[], to_many=[xxx], comment=}

2017-02-22 14:13:45 GMT <bmejias> but the email never arrives.

2017-02-22 14:14:14 GMT <bmejias> the smtp is correctly configured as we can send emails with rules on the folder, and the activities notifications arrive every day.

2017-02-22 14:14:25 GMT <bmejias> but that particular emails does not arrive.

2017-02-22 14:14:55 GMT <bmejias> I also changed the template to a plain "Test" html file, to prevent any error on the "ftl"

2017-02-22 14:14:58 GMT <bmejias> but nothing.

2017-02-22 14:15:24 GMT <bmejias> any idea about how can I track the problem?

2017-02-22 14:16:39 GMT <yreg> Column sized in AAAR datamart, most of them should not have size limits

2017-02-22 14:17:42 GMT <yreg> I just hit an extraction problem due to the Values column in stg_audits being limited to 5000 char, where the values in my audit entry make 5254 characters !

2017-02-22 14:18:18 GMT <yreg> just one extra consideration for alflytics :)

2017-02-22 15:04:16 GMT <yreg> fcorti QQ: what is the Workflow dashboard supposed to show

2017-02-22 15:04:18 GMT <yreg> ?

2017-02-22 15:04:50 GMT <yreg> I do not have any error in the logs, but that dashboard does not seem to be working

2017-02-22 15:05:18 GMT <yreg> PS: the active workflows report shows dozens of active workflows

2017-02-22 15:28:32 GMT <fcorti> yreg => I ususally iimit the varchar to 4096 but I'll increase it

2017-02-22 15:33:50 GMT <yreg> fcorti, I am suggesting to ommit max length

2017-02-22 15:34:21 GMT <yreg> especially when you can not predict the size of the entery

2017-02-22 15:34:26 GMT <yreg> *entry

2017-02-22 15:34:44 GMT <yreg> ie auditing values

2017-02-22 15:34:57 GMT <fcorti> yreg, ok => another note taken for Alytics ;-)

2017-02-22 15:35:51 GMT <yreg> thanks

2017-02-22 15:36:09 GMT <fcorti> you are welcome... I always look for good test cases

2017-02-22 15:42:12 GMT <AFaust> fcorti: varchar - 4096 characters or bytes?

2017-02-22 15:42:59 GMT <fcorti> characters

2017-02-22 15:43:15 GMT <fcorti> I like the 2 ^ X

2017-02-22 15:43:34 GMT <AFaust> Are you sure? Even Alfresco schema itself does not properly handle character lengths for column widths

2017-02-22 15:43:55 GMT <AFaust> And "characters in what character set"?

2017-02-22 15:43:59 GMT <fcorti> I leave the management of this to PostGresql and MySql

2017-02-22 15:44:07 GMT <fcorti> In the next version also to HSql

2017-02-22 15:44:14 GMT <AFaust> They will handle that fine - when you configure it correctly

2017-02-22 15:44:31 GMT <AFaust> That is what I meant with "even Alfresco schema itself does not properly handle character lengths"...

2017-02-22 15:44:44 GMT <AFaust> when => if

2017-02-22 15:44:59 GMT <grendal_prime_> alfrescos smb server does not lock the document?

2017-02-22 15:45:00 GMT <fcorti> Using Pentago (but it's the same with other tools) the batches are on top of the DB

2017-02-22 15:45:17 GMT <AFaust> grendal_prime: Not if the client does not submit a SMB lock command

2017-02-22 15:45:55 GMT <grendal_prime_> libre office?

2017-02-22 15:46:56 GMT <AFaust> I have not coded in libre office, so don't know how it is supposed to behave

2017-02-22 15:47:27 GMT <AFaust> That is the problem with SMB and WebDAV mounting of Alfresco - you are always dependent on client applications behaving "correctly" or "as you would expect"

2017-02-22 15:47:53 GMT <grendal_prime_> Also one of my test machines is not quite 2 ghz what services will not be available with a slower processor (or is it just a performance hit.)

2017-02-22 15:50:11 GMT <grendal_prime_> AFaust, i agree however..it seems that if alfresco is going to use a smb server of its own (and seeing as how it sort of has to because of the filesystem being what it is) it would be a relatively simple enhancement to simple lock the file when it is opened.

2017-02-22 15:52:13 GMT <AFaust> grendal_prime: Alfresco cannot know for what reason a file is being opened / accessed.

2017-02-22 15:52:36 GMT <AFaust> If the file would be locked, you will quickly end up with almost all files being locked. And when do you unlock the file?

2017-02-22 15:52:36 GMT <grendal_prime_> that is not intended to be a jab, more like a feature request. And one that comes from somebody who has had to deal with allot of trouble because of lost data related to file locking or lack there of

2017-02-22 15:53:11 GMT <AFaust> I can understand the request / requirement, but from a technical point of view it simply can never work.

2017-02-22 15:53:29 GMT <AFaust> That is the reason a lot of us will almost always recommend NOT to use file-based protocols for integration.

2017-02-22 15:54:47 GMT <grendal_prime_> ya i see your point. hmm the more i think about this the more it looks like cmis needs to be a protocol in the nautilus file browser...and windows file browser as well.

2017-02-22 15:54:49 GMT <AFaust> That is also the reason why Alfresco is moving into a desktop synch approach

2017-02-22 15:55:43 GMT <grendal_prime_> temporary local folder? and an app that sends it up?

2017-02-22 15:56:30 GMT <AFaust> https://www.youtube.com/watch?v=-obswDPcE7M

2017-02-22 15:56:38 GMT <grendal_prime_> so...scenario would be

2017-02-22 15:56:46 GMT <grendal_prime_> oh cool hold on

2017-02-22 16:02:30 GMT <Frederick> Guys I'm installing alfresco in centos machine, the machine holds both alfresco (tomcat) and the database, but for some reason I seem unable to perform the database part of the setup it fails reporting: /opt/alfresco-community/postgresql/bin/createuser -p 5432 -U postgres --no-superuser --no-createdb --no-createrole alfresco

2017-02-22 16:02:30 GMT <Frederick> createuser.bin: could not connect to database postgres: FATAL: Ident authentication failed for user "postgres"

2017-02-22 16:04:11 GMT <grendal_prime_> Frederick, ident auth is the problem, you need to set up the postgresql server to accept connections from other machines

2017-02-22 16:04:46 GMT <grendal_prime_> AFaust, will there be a linux client for this desktop sync?

2017-02-22 16:05:33 GMT <grendal_prime_> Frederick, i have a link for the procedure you need to perform to get that to work.

2017-02-22 16:06:12 GMT <AFaust> Just like it is "America first" it will be "Windows first". I believe a MacOS variant is in the works and I haven't anything concerning a Linux one...

2017-02-22 16:06:35 GMT <AFaust> "haven't [heard] anything"

2017-02-22 16:06:38 GMT <grendal_prime_> bummer

2017-02-22 16:06:43 GMT <grendal_prime_> Frederick, https://www.liquidweb.com/kb/change-postgresql-authentication-method-from-ident-to-md5/

2017-02-22 16:06:44 GMT <alfbot> Title: Change PostgreSQL Authentication Method from Ident to MD5 – Liquid Web Knowledge Base (at www.liquidweb.com)

2017-02-22 16:08:28 GMT <grendal_prime_> AFaust, so, i have several mixed houses. im gonna have to look more into clients that issue locks it looks like. I think libre 5.1 has this built in and i thought that 4 had it but i dont see anything on the sniff.

2017-02-22 16:11:10 GMT <grendal_prime_> I build custom solutions that use openvpn as the core connection device for road warriors. I have in the past just used a standard samba server and limit access to it via linux user management. However there is no document indexing with that arangement and more and more people are frustrated with not having a google style search facility for there own documents.

2017-02-22 16:14:30 GMT <grendal_prime_> i have a lot of linux desktop users...developers, graphix art designers, yada yada yada. They are born into mac world alot of times and fallback to using linux opensource because of the cost and similar file sturcture , tools and programs "feel right" for them.

2017-02-22 16:16:18 GMT <grendal_prime_> I like the idea of a desktop app service...makes it possible to do additional things...colaboration options become more rich...but um no nix client? thats a little disenchanting.

2017-02-22 16:19:28 GMT <grendal_prime_> Frederick, you get that hammered out?

2017-02-22 16:19:30 GMT <AFaust> grendal_prime_: I assume you are not an Enterprise customer? That would at least give you the option of giving feedback to Alfresco that they might actually care to hear / consider...

2017-02-22 16:19:30 GMT <Frederick> grendal_prime_: ive already followed int, trying to instal all again

2017-02-22 16:22:05 GMT <grendal_prime_> AFaust, ya looked into that. i was told they minimal support they offer is for 200 clients. And i was not able to find any pricing information.

2017-02-22 16:23:12 GMT <Frederick> Cannot find Alfresco Repository on this server. (Does this application have access to alfresco-global.properties? Does this application have cross-context permissions?)

2017-02-22 16:24:22 GMT <grendal_prime_> Im thinking you may need to adjust that .properties file to point to the database server.

2017-02-22 16:25:52 GMT <grendal_prime_> i have not built any solutions that use a database outside the alfresco server itself (i use postgres allot though for web apps) and the biggest problem with postgres is the authentication process. That error there looks more like alfresco is still pointed at the local server and if postgres is not running on there that sounds like a logical error.

2017-02-22 16:27:05 GMT <Frederick> db.url=jdbc:postgresql://localhost:5432/${db.name}

2017-02-22 16:27:18 GMT <grendal_prime_> ya

2017-02-22 16:27:32 GMT <grendal_prime_> localhost...you probably need to replace with the ip address of your db server.

2017-02-22 16:27:35 GMT <Frederick> but it is not working

2017-02-22 16:27:43 GMT <Frederick> grendal_prime_: im running on localhost

2017-02-22 16:28:03 GMT <grendal_prime_> oh i though you were pointing it to a seperate db server.

2017-02-22 16:28:35 GMT <grendal_prime_> whats the operating system you are using

2017-02-22 16:32:38 GMT <grendal_prime_> AFaust, ya libre office does issue a locking command, it does not show in alfresco share as being locked though

2017-02-22 16:33:20 GMT <grendal_prime_> so the share app does not recognize the smb lock it looks like.

2017-02-22 16:33:46 GMT <AFaust> Ok - there are lock types that will not show in Share. E.g. the MS Office apps also trigger WebDAV locks that have no proper UI representation in Share

2017-02-22 16:34:10 GMT <AFaust> Also, some locks are "ephemeral", meaning they are not persisted in the database as properties of the node

2017-02-22 16:35:20 GMT <grendal_prime_> hmm well, this will be sufficent for most of the issues we would run into, The share interface is mostly going to be used for offsite access and manageing permissions to "sites"

2017-02-22 16:36:34 GMT <grendal_prime_> also searching for content. I do have concerns about quantity of sessions and performance though. Do you guys have an metrics on performance vs number of smb connections?

2017-02-22 16:37:07 GMT <AFaust> Not really - most SMB uses are very light fortunately

2017-02-22 16:37:37 GMT <AFaust> One thing to keep in mind is that each active SMB session will be associated with a specific thread on the Repository-tier.

2017-02-22 16:38:22 GMT <AFaust> So if you have a high Xss setting (which I have seen around in the community) you can was quite a lot of memory

2017-02-22 16:39:20 GMT <AFaust> Also, it depends greatly on what is done via these SMB sessions. I have once seen a single SMB session bringing down a server, because some bloke was stupid enough to run a "find" / "Windows Search" on the mounted drive...

2017-02-22 16:40:41 GMT <grendal_prime_> Idealy i would like to get the users to use the share interface for locating files and launching the editing process.. its just the fact that they have to make the additional step of locating it locally and uploading it that the users just cant seem to wrap their heads around.

2017-02-22 16:41:46 GMT <grendal_prime_> I get it...the developer types get it...be all those types are familar with versioning software, im getting pushback from admin types and secritarys.

2017-02-22 16:42:47 GMT <grendal_prime_> it seems like this was pretty well addressed with a couple of apps that no longer work because of chrome's dropping of nappi plugins..

2017-02-22 16:45:11 GMT <grendal_prime_> looks as though there are people working on cmis protocol in gvfs

2017-02-22 16:46:41 GMT <IanW1> There is a CMIS client for LibreOffice if that helps - the remote files option

2017-02-22 16:47:53 GMT <grendal_prime_> ya, we have not had much luck with it

2017-02-22 16:48:11 GMT <grendal_prime_> i got to roll, thanks guys for the help

2017-02-22 16:56:46 GMT <grendal_prime_> AFaust, i had another question about nfs file shares. In the past i have basically utilized my existing filer for storage by first nfs mounting a nfs share on the filer i then set the data path on my alfresco server to the local dir (that is mounted to the nfs share) There are several reasons for this most have to do with the backup system running on the fler.

2017-02-22 16:57:52 GMT <grendal_prime_> I know that this is not ideal as the index's like to be on a local drive for speed. but...it does work well enough in most cases.

2017-02-22 16:59:01 GMT <AFaust> fcorti: Regarding the Alfresco One EA - a customer of mine is unable to find a proper ZIP download that contains the SOLR 4 WAR. I'll check too but maybe you already have a link handy

2017-02-22 17:01:17 GMT <grendal_prime_> my question regarding all of this is the replication tool, im assuming i can replicate the entire storage to another alfresco server that could run as a backup?

2017-02-22 17:10:28 GMT <AFaust> Sorry grenald_prime_ - due to the customer issue with the EA I was a bit distracted just now

2017-02-22 17:11:35 GMT <AFaust> It would be save to replicate everything in the storage that is contentstore content. It is definitely not save to replicate index data WHILE the Alfresco SOLR server is running

2017-02-22 17:11:51 GMT <AFaust> only if the SOLR instance is shut down would that be save as well

2017-02-22 17:11:57 GMT <AFaust> :%s/save/safe/g

2017-02-22 17:12:10 GMT <grendal_prime_> its cool i know how that goes, i just read up on the docs. mostly looks like a sync wich works for what i want to do.

2017-02-22 17:12:36 GMT <grendal_prime_> so basically if i use it to backup the data i would have to run the indexing on the second server?

2017-02-22 17:13:25 GMT <grendal_prime_> the index would not be backed up just the data. The big question i have is ...can i use it to migrate to a newer server?

2017-02-22 17:14:07 GMT <grendal_prime_> if i replicate the data to a second server that is already indexing with solr recognize the new data being added and index it?

2017-02-22 17:14:21 GMT <grendal_prime_> the target server that is.

2017-02-22 17:19:56 GMT <AFaust> If you migrate the data while SOLR is already indexing there is no guarantee that your files will be properly full-text indexed. There is no functionality for SOLR to check if the document is available at all - it is completely abstracted from SOLR anyway (i.e. SOLR might run on a different host to begin with)

2017-02-22 17:20:15 GMT <AFaust> Migration must be complete before you start the Repository / SOLR applications

2017-02-22 17:21:47 GMT <grendal_prime_> ok so basically.... Server A runs a replication job -> Server b who is already running. I have to stop solr delete index and force re- index.

2017-02-22 17:22:02 GMT <grendal_prime_> on Server B that is

2017-02-22 17:22:44 GMT <AFaust> Just to be sure: You are not talking about the Alfresco Replication feature, right?

2017-02-22 17:23:23 GMT <grendal_prime_> yep ...I have to submit a migration plan so if the software itself manages this it looks better to them.

2017-02-22 17:23:29 GMT <AFaust> Because for that you'd need both servers to be online. But for a NFS / storage replication the server B shouldbedown

2017-02-22 17:24:15 GMT <grendal_prime_> ya understand you dont want to servers working with the same data and keystore at the same time.

2017-02-22 17:24:46 GMT <AFaust> It is not a problem of working with the data at the same time, but moving of the data while some processes may already access data that has not been replicated yet

2017-02-22 17:25:50 GMT <grendal_prime_> oh, wait...so if i have two alfresco servers both pointing to the same datastore and keystore...up and functioning....thats ok?

2017-02-22 17:30:47 GMT <grendal_prime_> ill do some more reading.. documentation is pretty good. I got to get to the property the sun is shining and i have not gotten to run my saw mill in over a week.

2017-02-22 17:31:28 GMT <AFaust> grendal_prime_ It is ok as long as they use the same database and are in an active cluster to coordinate cache updates

2017-02-22 17:32:02 GMT <grendal_prime_> hmm i need to look more into that that....after i get my smb server issue worked out.

2017-02-22 17:35:46 GMT <grendal_prime_> in the mean time

2017-02-22 17:35:49 GMT <grendal_prime_> http://resume.booksnmore.com/videos/VID_20170130_134551484.mp4

2017-02-22 17:36:16 GMT <grendal_prime_> logs got delivered yesterday and i need to build a barn.

2017-02-22 17:36:21 GMT <grendal_prime_> hehehe

2017-02-22 17:37:37 GMT <grendal_prime_> peace out...my friend (who...has no known face or even shadow as of yet)

2017-02-22 22:26:02 GMT <digcat> ~later tell fcorti, back to a conversation regarding a roadmap, this seems to be a good way to do it, https://github.com/cockroachdb/cockroach/issues/13517

2017-02-22 22:26:02 GMT <alfbot> digcat: The operation succeeded.

End of Daily Log

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