RFI / Bulk Option to Convert Leads into Targets?

Unless Lucene got installed out of the box I’m not 100% confident that it was actually there.

So far I’m super impressed by Elastic.

Quick question: How often CAN I run the indexer in the Scheduler? And does it do a full index from scratch every time or just the new or updated records?

eg, is there any performance hit to letting it run every minute or every five minutes?

Lucene is the “Advanced search” in SuiteCRM, it is installed out-of-the-box. The directory listing you posted above shows that it worked on your system for a while, then stopped working.

I am not sure that Elastic even needs those Scheduler jobs. I believe they are for Lucene. I would try disabling them, if you note that Elastic keeps working, then you’re good.

I wonder if some package accidentally got uninstalled as part of apt at some point ? I could find no evidence that Lucene had been installed or anything running – what process fingerprint do you see in a ps aux ?

I’ll toy around with the Scheduler – I do know that if i insert a new record its not searchable right away and I have not tried disabling the job and checking to see if / when it does show up

Lucene Search is a PHP package that lives inside the SuiteCRM directory, not installed by the OS.

It runs when a bean is saved (under Apache process) and when cron.php runs (from SuiteCRM schedulers, which are CLI PHP).

So basically it’s not something that you will distinguish from SuiteCRM itself.

Elastic Search is different, it’s more of a separate software that runs it’s own processes.

That’s helpful.

Now, it seems that it stopped working quite awhile ago – any thoughts on what could cause that?

The good news here is that ES was relatively easy to install and stitch in and appears to be superior to core and advanced.

The last question, for the most part as far as this thread goes, is how frequently should someone re-run the indexing process ?

-dvd

I don’t know why Lucene stopped working, that would take a debugging/troubleshooting session which is hard to conduct without the live system, looking at logs, errors, etc.

As I said, if you have elastic working, just stay with that and don’t worry about the inferior Lucene system.

After the original full indexing, indexing shouldn’t really be necessary, since all saves will automatically index the new or changed data. But there might be other things that throw the index off-sync, I’d start by setting a nightly re-indexing in the scheduler, and leave it there if there are no apparent issues. I assume that Elastic is quite optimized in terms of reindexing, normally it will do differential indexes, not full indexes.

1 Like

hi. just saw this thread. can you post the steps used/needed to run elastic? are there already steps for this?

thanks

A simple Google search would take you there faster :wink: