10.4.1 vs 10.5

[DaveTenney]

All,

  looking for some real world input from those who have used 10.4.1 for server and have recently upgraded to 10.5

we have been trying to vet 10.4.1 in our staging environment to be used in our production environment and at this point in time we cannot vet 10.4.1. i dont want to get into details here on that, we are providing a lengthy response to this directly to ESRI.

what i would like to know is how people feel about 10.5 so far? were you experiencing issues with 10.4.1 that seem to be resolved with 10.5? i would assume we are fairly close to a 10.5.1 release (around the time of the UC). again, just looking for some real world users and your input.

thanks,

dave

[MichaelVolz]

Does your organization utilize an SDE database(s)?  If so, is it Oracle, SQL Server, or something else?

[DaveTenney]

SQL Server

[DavidColey]

Hi Dave - I have upgraded a 10.4.1 instance to 10.5 Arc Enterpise with both the ArcServer and ImageServer roles on my edn machine.  We do not have a staging environement per say, but I mimic my 3-server production environment by placing the config, system and data-store directories on a shared file server, with the web adaptor on a seperate web server.  We have not yet moved to Portal, so for the time being we have not moved to the relational and tile data stores or federated the server with portal.

So far 10.5 is functioning just as 10.4.1 does.  I am able to perform caching, serve mosaice datasets as image services, publish print and other gp services.  Manager is virtually the same, securing services functions the same. Using the self-signed cert for the internal server address works the same as does using the CA cert for the web adaptor.  So far that's about it.  No issues, but I wasn't having any at 10.4.1 either.

[MichaelVolz]

David Coley:

Do you have any Oracle SDE databases in your shop?

[DavidColey]

Hi Michael, no we're a windows server 2008 and sql 2014 shop.  No issues with registering sql sde connections for Colletor, for example

[DaveTenney]

we were noticing behavior when we would stop locator services and rebuild said locators.

      after we have rebuilt the locators and put the data in the appropriate folder locations and try to start the service back up, the service never fully comes back.

i've watched the esri logs while the process works...

   1) service stopped

   2) data copied over and locators rebuilt

   3) service started

         i can see in the esri logs that it believes the service has come back online but when i log into the admin site the configuration status is still on "stopped". so it seems to be a disconnect between what is actually happening and what the service thinks has happened.

has anyone else experienced this?

i know some have suggested the truncate and append approach. we have been working our process for almost 3 years now without an issue, now we try to upgrade to 10.4.1 and we saw this behavior. NOW, we were are also seeing this in 10.5!

[MichaelVolz]

Dave:

It sounds like you are deleting the data tied to your address locator and copying new data to the same location.  A truncate and append would also work, but geocode services need to be stopped when an address locator is rebuilt anyhow, as the geocode service puts locks on the address locator.

In what ArcGIS version did you last get your address locator/geocode service update process to work properly?

[DaveTenney]

10.2.2 is the only release that we've been able to reliably run the process (been running for over 2yrs)

   we are aware that we need to get away from 10.2.2, so decided to start testing other releases. unfortunately, we are now experiencing the above issue. 

it's just a very strange behavior and extremely difficult to troubleshoot. we had almost 25 successful runs in 10.5 and then out of nowhere, we started seeing this behavior. 

Many have suggested the truncate/append approach and we have discussed it and have decided that it would not be an option for us to go that route. So we need to be able to figure out why were getting different signals about whether or not a service has really started back up or not.

[DavidColey]

If I might suggest a small tweak to your process.  I've been doing this since file-based geolocators came out.  What I don't try to do is rebuild my locaters in place.  I perform a feature classs to feature class for our address points and streets from our SDE to a local filegdb.  I then perform the rebuilds in the local directory.  I stop locater services and overwrite the entire folder to my Servers' registered folder, then restart.  Since the locators are configured to with relative paths for the data, they just contain the new data.  This has never failed on me (yet!)