Latest Contributions by JonathanQuinn

‎03-24-2017

Take a look at this blog post, which discusses using the Windows Task Scheduler and Python installed on the machine. It's somewhat similar to running a script using the command prompt, you'd pass in the path to the python.exe and then the script as an argument.

‎03-24-2017

You shouldn't really be concerned with the resources the Web Adaptor will use. The system requirements for the Web Adaptor don't list CPU or RAM requirements because it's simply a reverse proxy running within an application pool. A setting that may be of interest to you within the application pool is the number of worker processes. However, it would be difficult to give definitive answers on performance as there are a lot of factors that go into those types of answers. For example, how many concurrent users will be accessing the services at peak time? How responsive are the services behind the web adaptors? What else is running in IIS? Your best approach is to run load tests using JMeter or another load testing application in a QA environment that mimics your production environment. The only way you'll really know if your configuration is good is if you test it.

‎03-23-2017

I don't think you'll need to do anything with the map service after updating the geometry type. The only thing you'll probably need to deal with is locks on the data, which you'll get around by stopping the services. For your other question, they don't need to be republished, but you won't be able to add a field if the option to disable schema locking isn't checked within the map service properties. Lots of negatives there, so to add a field while a service is using the data, disable schema locking.

‎03-23-2017

No, casing shouldn't matter. All that really matters is that the machine is reachable and the Data Store service is running. Can you take a backup of Data Store and then restore it manually? That's really what the DR tool is doing, it just bundles up Portal and Server.

‎03-23-2017

If you call on the variable without printing, it'll be correct as a Result object. If you print the variable you'll see the backslash is not escaped: >>> gdb = arcpy.CreateFileGDB_management("C:/Temp","blah.gdb") >>> gdb <Result 'C:/Temp\\blah.gdb'> >>> print(gdb) C:/Temp\blah.gdb >>> arcpy.Exists(gdb) True >>> arcpy.env.workspace = gdb Runtime error Traceback (most recent call last): File "<string>", line 1, in <module> File "c:\program files (x86)\arcgis\desktop10.5\arcpy\arcpy\geoprocessing\_base.py", line 541, in set_ self[env] = val File "c:\program files (x86)\arcgis\desktop10.5\arcpy\arcpy\geoprocessing\_base.py", line 601, in __setitem__ ret_ = setattr(self._gp, item, value) RuntimeError: Object: Error in accessing environment <workspace> >>> arcpy.env.workspace = str(gdb).replace("\\","/") >>> arcpy.env.workspace u'C:/Temp/blah.gdb'‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍ There seems to be different implementations of arcpy.Exists and arcpy.env.workspace, as the latter seems to look at the result as a string.

‎03-23-2017

Just to provide more insight, if you print the gdb variable, you'll get "<basedir>\<gdb_name>". The OS separator between the out file path and FGDB name isn't escaped properly, as it's set to a backslash, which is why you see that runtime error when trying to use the gdb variable as is. In one line, this worked: gdb = str(arcpy.CreateFileGDB_management(outDir,outName)).replace("\\","/") You can replace the single backslash, (escaped with a double backslash), with a forward slash.

‎03-23-2017

I would use ProcMon to determine the file activity that occurs when Server attempts to write to that location. You can filter for the specific directory path so you only return traffic to that directory. I assume you'll see some errors in the log.

‎03-23-2017

Do the logs shed any more light on the problem? This is a single machine site?

‎03-23-2017

You should be using the arcpy.env.scratchFolder to construct the path to output data. I'm also using the os.path.join function to construct the path to output data. Finally, you don't need to create a feature layer again to return the data, simply return the data the updateFCPath variable is pointing to. import os filePath = arcpy.env.scratchFolder data = datetime.datetime.now() timestamp = str(data).replace("-","").replace(":","").replace(".","").replace(" ","_") events_l = arcpy.MakeFeatureLayer_management("event_lyr", "events_layer") print "Inizio segmentazione dinamica eventi" updateFCPath = os.path.join(filePath,"{0}.shp".format(timestamp)) updateFC = arcpy.CopyFeatures_management(events_l,updateFCPath) arcpy.SelectLayerByAttribute_management("event_lyr", "NEW_SELECTION", "LOC LIKE 'LO%'") arcpy.AddMessage("num_sel: " + str(arcpy.GetCount_management("event_lyr"))) print("num_sel: " + str(arcpy.GetCount_management("event_lyr"))) layer = arcpy.MakeFeatureLayer_management(Lo_UTM, "loc_lyr") sourceFC = layer sourceField = ['OR_ID','SHAPE@'] valueDict = {r[0]:(r[1:]) for r in arcpy.da.SearchCursor(sourceFC, sourceField)} updateFieldsList = ["loc", "Shape"] with arcpy.da.UpdateCursor(updateFC, updateFieldsList) as updateRows: for updateRow in updateRows: keyValue = updateRow[0] print(keyValue) if keyValue in valueDict: updateRow[1] = valueDict[keyValue][0] print(valueDict[keyValue][0]) updateRows.updateRow(updateRow) arcpy.SelectLayerByAttribute_management('event_lyr', "CLEAR_SELECTION") arcpy.SetParameter(1,updateFCPath)‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍ What is "event_lyr", used initially on line 7? If that's already a feature layer, there's no need to run the Make Feature Layer tool again, and you don't really need to store it on disk. I've also added the arcpy.AddMessage line so you see the print statement on line 4. data = datetime.datetime.now() timestamp = str(data).replace("-","").replace(":","").replace(".","").replace(" ","_") print "Inizio segmentazione dinamica eventi" arcpy.AddMessage("Inizio segmentazione dinamica eventi") arcpy.SelectLayerByAttribute_management("event_lyr", "NEW_SELECTION", "LOC LIKE 'LO%'") arcpy.AddMessage("num_sel: " + str(arcpy.GetCount_management("event_lyr"))) print("num_sel: " + str(arcpy.GetCount_management("event_lyr"))) layer = arcpy.MakeFeatureLayer_management(Lo_UTM, "loc_lyr") sourceFC = layer sourceField = ['OR_ID','SHAPE@'] valueDict = {r[0]:(r[1:]) for r in arcpy.da.SearchCursor(sourceFC, sourceField)} updateFieldsList = ["loc", "Shape"] with arcpy.da.UpdateCursor(updateFC, updateFieldsList) as updateRows: for updateRow in updateRows: keyValue = updateRow[0] print(keyValue) if keyValue in valueDict: updateRow[1] = valueDict[keyValue][0] print(valueDict[keyValue][0]) updateRows.updateRow(updateRow) arcpy.SelectLayerByAttribute_management('event_lyr', "CLEAR_SELECTION") arcpy.SetParameter(1,"event_lyr")‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍

‎03-23-2017

I would take a look at the http requests when logging in. Spin up Fiddler or use the Dev Tools to see which requests take the longest, then you can work from there to determine why they take so long.

‎03-23-2017

Was the ArcGIS Data Store validating prior to running the restore? The error is pretty specific in stating that no connection could be made to the Data Store. Do you have both the relational and tile cache data store configured?

‎03-22-2017

Are you seeing that error when going through the web adaptor? Can you try to reach the portal through 7443?

‎03-21-2017

My suggestion is again, to disable public access to your internal network until you have an understanding of web servers. More helpfully, though, don't use the reverse proxy rules within IIS. Install the Web Adaptor on the machine that's publicly accessible to the internet and register it with ArcGIS Server. You'll still need to disable Windows Authentication on the web server.

‎03-20-2017

What do you mean by "corrupt"? If you were to navigate to https://<machine>:7443/arcgis/portaladmin can you reach the machine? Do you see a javaw.exe process running when Portal is running?

‎03-20-2017

Was it a hosted service? If they were using ArcGIS Data Store, then they were also using Portal, in which case the two types of services you can publish are "hosted" and "non-hosted" services. Non-hosted services are the traditional map services that you'd publish directly to ArcGIS Server. The data can be referenced in the original location it's stored in or copied to the Server when publishing. These spin up an ArcSOC.exe process, which consumes memory and CPU on the machine. Hosted services can be published directly to the Portal and the data is copied into the ArcGIS Data Store. They don't spin up an ArcSOC.exe process, thus conserving resources on the machine, and you can publish significantly more of these types of services than traditional services. I don't work too much with GeoEvent, but a lot of clients spin up a hosted service as output, (analysis tools, Insights, etc), so perhaps GeoEvent was creating one of these types.

Online Status	Offline
Date Last Visited	‎06-08-2026 07:43 AM

My Ideas

Latest Contributions by JonathanQuinn

Re: Run Python script on a specified time schedule

Re: How much RAM does the Web Adaptor require?

Re: Do Map Services need to be republished after Migrate Storage SDEBinary to SQL Geometry?

Re: Webgisdr restore Fail

Re: Creating a geodatabase and setting it to env.workspace

Re: Creating a geodatabase and setting it to env.workspace

Re: ArcGIS Server 10.5 Cannot add machine to site due to connection failure to config-store and directories

Re: Arcgis Server Setup Wizard failed to synchronize

Re: Pythonerror ERROR 000622

Re: why is argis for server manager login slow

Re: Webgisdr restore Fail

Re: SSL certificate corruption after patching

Re: Can not access my ArcGIS server externally

Re: SSL certificate corruption after patching

Re: Enhanced map service based on ArcGIS Data Store

Re: Portal for ArcGIS upgrade from 11.5 to 12.1 no...

Re: ArcGIS Portal public maps asking for login cre...

Re: Failed to start the portal service

Re: Restore to standby environment per component (...

Re: Can't backup ArcGIS Enterprise 11.4 webgisdr s...