JRExecAction() function behavior in Netcool Impact

Just a short post today to set the record straight on the use of the JRExecAction() function in Netcool Impact policies.

The product documentation states that the JRExecAction() function returns nothing, but sets the variables ExecError and ExecOutput corresponding to stderr and stdout from the script/command that you run. However, the function itself returns the data written to stdout in addition to setting ExecOutput and/or ExecError. So if you try the following, you'll see that the same output is logged twice:


results = JRExecAction("/bin/ls", "/", false, 1000);
log(0,"Frank results = " + results);
log(0,"ExecOutput results = " + ExecOutput); 


Incidentally, the first use above is really the more common use in practice. You will find very few implementations where the ExecOutput variable is referenced.

How to reconfigure your ServiceNow MID server when your developer instance changes URLs

If your developer ServiceNow instance is hibernated, it may be assigned a new URL. This will break your MID server(s).

To fix this:

cd /servicenow/<midservername>/agent

vi config.xml

search for the old URL

update it

if the password has changed, put the new password in the file unencrypted. On startup, it will get encoded and encrypted.

./stop.sh

./start.sh

Go to your instance to check the status of your MID servers.

Customizing the ServiceNow Netcool Connector

UPDATE 5/1/2020

You can update this script through the ServiceNow GUI. The navigator item to select is "Script Files" under "MID Server". And the name of the script is NetcoolConnector. Edit in the GUI and it gets updated on all of your MID servers.

Background

The ServiceNow Netcool Connector (introduced at some point before the New York release) allows you to pull events from a Netcool ObjectServer into ServiceNow. The connector is a process that runs on a MID Server. Within the ServiceNow interface, there are only a few configuration options (userid/password, JDBC URL, how often to run, etc.). However, there are no filters to configure. That's because the connector is a straightforward Groovy script that you can edit as needed on the MID Server.

Details

The Netcool Connector script is found on the MID Server in the file .../agent/scripts/Groovy/NetcoolConnector.groovy. Some of the interesting parts of the script are the actual query that's run:

query = "select top 3000 Identifier,Node,NodeAlias,AlertKey, Manager,Agent,AlertGroup,Severity,Type,Summary,Acknowledged,LastOccurrence,StateChange,SuppressEscl from alerts.status where StateChange > " + lastTimeSignature + " and Manager not like '^.*Watch\$' order by StateChange asc";

That is exactly the query that's run, and you can edit it to include custom fields, for example. To complete the customization, you also need to update the createEvent() function to actually include those custom fields in the event that's created in ServiceNow. In there you can also do any hard-coded transforms that are required or anything else.

Short script to generate continuous random events to Netcool OMNIbus

So you need to generate some events to OMNIbus? It's really easy with the nco_postmsg shell script that's installed on your ObjectServer host:

while (true); do nco_postmsg -user root -password n3tc00l "Identifier='my_identifier_${RANDOM}'" "Node='mynode${RANDOM}'" "Severity=5" "Manager='nco_postmsg'" "Summary='An event occurred again'" "ImpactFlag=1" "AlertGroup='FRANK_${RANDOM}'"; sleep 10; echo sending another event; done

Update the user and password parameters, and you can add any fields you need. It simply generates an event every 10 seconds.

Why you should use MXOSINCIDENT instead of OSLCINCIDENT for creating and querying ICD incidents

In re-reading the documentation and looking through the system, I finally recognized why the MXOSINCIDENT is a much better choice for integration than OSLCINCIDENT. The OSLCINCIDENT OSLC Resource only includes 11 attributes (fields) for an incident. Those are:

modifiedBy
modified
classStructureId
title
internalPrioirity
created
creator
priority
status
shortId
identifier


and that's it. So when you GET an incident via the OSLCINCIDENT URL (.../maximo/oslc/oslcincident/...), those are, essentially, all of the fields that you'll see for that incident. For example, here is the data I retrieved for incident number 1003 in my system:

dcterms:creator
rdf:resource "http://was.gulfsoft.rocks:80/maximo/oslc/os/sccdoslcperson/_TUFYQURNSU4-"
oslc:shortId "1003"
dcterms:title "First incident"
dcterms:created "2019-11-27T10:43:18-05:00"
dcterms:modified "2020-02-11T11:10:06-05:00"
sccd:statusdate "2019-11-27T10:43:18-05:00"
rdf:about "http://was.gulfsoft.rocks:80/maximo/oslc/os/oslcincident/_SU5DSURFTlQvMTAwMw--"
dcterms:identifier "9"
prefixes
sccd "http://jazz.net/ns/ism/helpdesk/sccd#"
rdf "http://www.w3.org/1999/02/22-rdf-syntax-ns#"
dcterms "http://purl.org/dc/terms/"
oslc "http://open-services.net/ns/core#"
_rowstamp "17235351"
oslc:modifiedBy
rdf:resource "http://was.gulfsoft.rocks:80/maximo/oslc/os/sccdoslcperson/_TUFYQURNSU4-"
sccd:status_description "New"
sccd:status
rdf:resource "http://was.gulfsoft.rocks:80/maximo/oslc/os/sccdoslcsyndomain/_SU5DSURFTlRTVEFUVVMvTkVXL35OVUxMfi9_TlVMTH4vfk5VTEx_L05FVw--"

However, here's the data retrieved for the SAME incident using the MXOSINCIDENT URL (.../maximo/oslc/os/mxosincident/...):

spi:ticketid "1003"
spi:status "NEW"
spi:pluspporeq false
spi:virtualenv false
spi:hasactivity false
spi:relatedtoglobal false
spi:changeby "MAXADMIN"
spi:accumulatedholdtime 0
spi:selfservsolaccess false
_rowstamp "17235351"
spi:class_description "Incident"
spi:inheritstatus false
relatedrecord_collectionref "http://was.gulfsoft.rocks:80/maximo/oslc/os/mxosincident/_SU5DSURFTlQvMTAwMw--/incidentrelrec"
multiassetlocci_collectionref "http://was.gulfsoft.rocks:80/maximo/oslc/os/mxosincident/_SU5DSURFTlQvMTAwMw--/incidentmultiassetloci"
spi:isknownerror false
spi:oncallautoassign false
spi:createdby "MAXADMIN"
spi:cinum "RBA_PAYROLL"
spi:statusdate "2019-11-27T10:43:18-05:00"
spi:affectedperson "MAXADMIN"
spi:class "INCIDENT"
spi:createwomulti "MULTI"
spi:description "First incident"
ticketspec_collectionref "http://was.gulfsoft.rocks:80/maximo/oslc/os/mxosincident/_SU5DSURFTlQvMTAwMw--/ticketspecclass"
spi:creationdate "2019-11-27T10:43:18-05:00"
spi:actlabhrs 0
spi:historyflag false
spi:sitevisit false
rdf:about "http://was.gulfsoft.rocks:80/maximo/oslc/os/mxosincident/_SU5DSURFTlQvMTAwMw--"
spi:createwomulti_description "Create Multi Records"
prefixes
rdf "http://www.w3.org/1999/02/22-rdf-syntax-ns#"
spi "http://jazz.net/ns/ism/asset/smarter_physical_infrastructure#"
oslc "http://open-services.net/ns/core#"
spi:outageduration 0
spi:isglobal false
spi:multiassetlocci
0
localref "http://was.gulfsoft.rocks:80/maximo/oslc/os/mxosincident/_SU5DSURFTlQvMTAwMw--/incidentmultiassetloci/0-5"
spi:recordclass "INCIDENT"
spi:progress false
rdf:about "http://childkey#SU5DSURFTlQvTVVMVElBU1NFVExPQ0NJLzU-"
spi:langcode "EN"
spi:multiid 5
spi:isprimary true
spi:performmoveto false
prefixes
rdf "http://www.w3.org/1999/02/22-rdf-syntax-ns#"
spi "http://jazz.net/ns/ism/asset/smarter_physical_infrastructure#"
oslc "http://open-services.net/ns/core#"
_rowstamp "17235352"
spi:hasld false
spi:pmchgassesment false
spi:recordkey "1003"
spi:cinum "RBA_PAYROLL"
spi:externalsystem_description "EVENT MANAGEMENT"
spi:pmscinvalid false
spi:reportdate "2019-11-27T10:43:18-05:00"
spi:hassolution false
spi:actlabcost 0
spi:changedate "2020-02-11T11:10:06-05:00"
spi:status_description "New"
spi:externalsystem "EVENTMANAGEMENT"
spi:affecteddate "2019-11-27T10:43:18-05:00"
spi:template false
spi:ticketuid 9
spi:reportedby "MAXADMIN"

To me, MXOSINCIDENT provides much better data.

When you Reset Windows, Microsoft saves a list of all of the applications removed

The Windows 1909 (November 2019) Windows 10 update didn't like one of my laptops at all. After trying multiple times to upgrade to it, I finally gave up and decided to take the route of choosing "Reset Windows" and keeping all of my data. That worked on the first try, so I was already happy about that. Then I noticed an HTML file on my desktop named "Removed Apps". I opened it, and much to my surprise, it listed ALL of the apps that were deleted by the reset. Happily, I still had all of them in my Downloads folder, so I can easily re-install them. This was a happy surprise, since I was certain that the reinstall part (and figuring out everything that needed to be installed) would be the worst part of the process.

Automating processes in a small company using free utilities

Introduction

I want to share some of the automation that I've written for my company to help with some of the back office processing. The reason that I wrote this automation myself is that I looked at numerous offerings on the market, and none fit exactly what we needed. Once I made that realization, I came up with automation using tools we already own or that are free to use, specifically:

Google Sheets, including Google Script
Windows Batch scripting
Perl scripting
Bookmarklets in JavaScript
Quickbooks IIF files

It looks like a lot (and it may very well be a lot), but for my background, this path required no more work than learning an off-the-shelf product.

The Payroll Process

Not everything is automated. We do have automated time sheets, but there are several pieces of data that need to be manually combined and calculated. Where we started with the automation was the spreadsheet in which this data was being captured. The data was already in a straightforward format, so it just needed to be cleaned up a little. Specifically, some of the entries just needed to have some specific keywords, and we put the spreadsheet into Google Sheets. We made the decision to move from Excel so that we could always access the spreadsheet and also because I personally enjoy programming in JavaScript (Google Scripting) than in Visual Basic (MS Office Scripting).

The process starts with my CFO filling out the Google Sheet using data from the few different sources that we have. She then presses the "Create IIF" menu item (created with a Google Script implementation of the onOpen() function). That creates both an IIF file for QuickBooks (that actually needs a little more processing - more on that later) and the a file containing the JavaScript for a bookmarklet that is used later in the process. It also sends an email out to each

The IIF file has to then be downloaded into a specific directory which contains a windows .BAT file, a shell script and a Perl script. The .BAT file is simply used to call the shell script, which calls the Perl script to actually parse the file. That Perl script mainly just adds TAB characters in the correct places to match the correct IIF format. I wasn't able to make the TAB characters actually stay in the downloaded file via Google Script, so this is the workaround I came up with.

The contents of the JavaScript file are the implementation of the "SetACH" bookmarklet that has already been defined in her browser. That bookmarklet is used to fill in the appropriate fields on our bank's ACH web page. To do this, my CFO logs into our bank and navigates to the appropriate page, then clicks the SetACH bookmarklet, which fills in all of the appropriate values.

Better Than a Packaged Product?

In a word, Yes. Every COTS (commercial off-the shelf) product requires some amount of configuration, which, in my experience, would have taken more time than creating the custom solution I have. Additionally, all of the COTS products I investigated included tons of capabilities that we simply don't need or that don't specifically work for us. As you can see, we only needed a tiny bit of functionality, and I didn't want to pay for a bevy of additional features that we wouldn't use. Specifically, I used the same evaluation process that I help customers with every day. I identified the specific functions I needed to perform and reports that I needed to generate, and evaluated the choices based on those. I also evaluated the existing tools and my development expertise with each to determine that I was capable of producing a working solution. We've now had this automation in place for three years, and it has required minimal upkeep.