When using Infor CRM SLX, the search function by itself in the upper right is known as SpeedSearch. It allows you to search all entities at once (or at least all the entities you are chosen to include in SpeedSearch), rather than searching a specific field for a specific entity. While both kinds of searches can be useful for certain purposes, SpeedSearch is very handy when you know a name, but are not sure what level of entity it is attached to. In order for SpeedSearch to work, it needs to index all that data. This is one of the big jobs that is performed by the SpeedSearch Server service. This indexing is set to run overnight, but under certain circumstances it can fail, or run long; which can leave the indexes out of date, or stop SpeedSearch from working entirely.
Assuming Infor CRM SLX SpeedSearch was set up correctly, the LAN Administrator program is the first place to check if either the SpeedSearch service won’t stay running, or if the indexes aren’t completing. (See post on SpeedSearch setup here.) To do this, go to Manage > SpeedSearch Configuration and select the Schedules tab. Here you can see if the Indexes are running like they should, and also set the schedule on which they run. I have seen issues caused by all the indexes running at once overwhelming the SpeedSearch Service, causing it to fail. I have also seen indexes taking to long to run that they did not complete before they were scheduled to kick off again, which can cause them to fail entirely. These aren’t the issues SpeedSearch can have, but the are two of the more common ones when it otherwise appears to be working.
The default schedule for almost all of the indexes is to run a full index each Saturday at 12:00 AM (which is basically Friday night at midnight), and to run incremental indexes Monday through Friday at 12:00 AM. Verify if your indexes are using that schedule or a custom one, and then look at the Last Run entry for each index. If the Last Run time matches the expected time, your indexes are running as expected. If the the Last Run time it an earlier date than the schedule would dictate, then you need to address why you indexes are not running properly.
When troubleshooting indexes that are failing to run, the first thing I generally do is pick one index that did not successfully complete the last time it was scheduled to run and test if it completes when I manually run it by itself (using the run now button). If that works, I will take a look at the time it takes the larger indexes to run a full index. If they running so long they are not completing before the system tries to run an incremental index, I need to change up the schedule so that the index has time to complete, generally by running incremental indexes less often for those entities.
If it appears the indexes have plenty of time to complete, but just don’t, it might be worth a try to stagger when the indexes run so they do not run all at once. You can try moving some of the full indexes to 12 AM Sunday instead of Saturday, and/or set some of them them to run at 3 AM or 6 AM than all at 12 AM (for either just full or both incremental and full). While there are certainly other, more complicated issues that can prevent SpeedSearch indexes from running, I have been able to solve many SpeedSearch issues by making sure the indexes are able to run and have time to complete.
