Below are common issues customers run across with haystack connection. We recommend you take a look at the following issues and see if any of them resolve your issue. If you happen to see a different error/issue, go to our Haystack Debug doc and see what error you get when debugging it.

Issue 1: Get 500: Internal Server Error (N3.8)

...

If you are connecting to your station via browser and get the above error. You need to do the following to resolve this error. (Usually on a JACE)

• Go to your station and navigate to Platform > Platform Administration > Set Module Filter to UI + Runtime

• Station will have to be restarted

• Then try to connect to it again via browser and see if it works.

Issue 2: Get 410: Gone (N4)

...

If you get the above issue when connecting to your station/haystack/about, then you need to do the following to resolve it.

• Go to your NHaystackService and right click on it to execute 3 actions in the same order below.

  • Initialize Haystack

  • Remove Broken Refs

  • Rebuild Cache

• Then try to connect to it again via browser and see if it works.

Issue 3: Get "No suitable auth scheme for: 302 null" error when attempting to connect to haystack (FIN v3.0.9 and N4)

If you get the above issue when trying to connect to N4 on the newer FIN Stack v3.0.9, then you need to configure the settings below to get this to work.

• Open N4 workbench and create new user and set their authentication as "Basic Authentication"

• Disable Https port in the Niagara Web Service because it'll force all traffic to https which throws the error

• Try to connect now and it should work

Issue 4: Get "haystack::AuthErr: No suitable auth algorithm for: 302" error when attempting to connect haystack (FIN v2.1.15 and AX)

If you get the above issue, then you'll need to configure the below settings to make this work.

• HTTP - for just http, make sure "Https Only" (if https is enabled at the same time) is disabled and Authentication Scheme is set to Basic. Then try disable and re-abling the connector.

• HTTPS - for https, make sure to set Authentication Scheme to Basic. Then make sure to Https certificate by using Crypto page.

Issue 5: Can't connect to N4.3.58 using FIN v2.1.15

If you are trying to connect to a N4.3 station via FIN Stack. Go here and look at the Custom Haystack Pod part: Haystack Compatibility#Custom-Haystack-Pod---N4.3+-with-FIN-v2.1.15 

Issue 6: Get "sys::ParseErr: Unexpected token: eof [line 1]"

If connecting to a haystack connector and get the above error, its likely that there may be some unescaped spaces in the point paths. Its recommended to execute the Initialize, Remove Broken Refs, and Rebuild Cache actions on the haystack service.

Issue 7: Get "sys::IOErr: javax.net.ssl.SSLHandshakeException: sun.security.validator.ValidatorException: PKIX path validation failed: java.security.cert.CertPathValidatorException: signature check failed"

If you get the above error, it means you have HTTPS enabled on your station. Follow the doc below when connecting with HTTPS.

• Make sure your uri is configured as https not http

• Follow this doc to add the Https certificate needed to connect to these: Crypto

Issue 8: Get "sys::IOErr: HTTP error code: 403" with N4.3 and patch 4.3.58.22.7

If you get the above error when attempting to connect to N4.3 with the updated patch, then you need to do the below.

• Change the Authentication Scheme on the user from Basic to Digest

• Tridium has since changed how the authentication mechanism works. Here is the forum regarding this: https://project-haystack.org/forum/topic/537

Issue 9: Get "sys::ParseErr: Invalid unit name 'in/wc' ["in/wc"] [line 41]"

If you get this error, it means you have points with unit 'in/wc' and in the beginning, this wasn't supported by nhaystack. It as since been implemented around mid 2016. To resolve this, follow the steps below. Got this from this project-haystack forum: https://www.project-haystack.org/forum/topic/433

• Download a text editor such as Notepad++ or something similar if you don't currently have one because notepad won't work. You'll need to save a file as UTF-8.

• Then go to the FIN Stack directory here: C:\Program Files (x86)\FIN Stack\FIN Stack <your version>\etc\sys

• You want to open the "units.txt" file there and add the unit on line 328 if you haven't modified this file before. Else it should be for "inches_of_water"

• The line looks like this: inches_of_water, inH₂O; kg1*m-1*sec-2; 248.84

• The line looks like this after applying the unit: inches_of_water, in/wc, inH₂O; kg1*m-1*sec-2; 248.84

• Then save the file and restart the FIN Stack service.

Issue 10: Get "sys::Err: Authentication failed: 302" with AX or N4 with FIN v2.1.15

...

This error means that the user likely doesn't have the JACE commissioned to use runtime+UI (not runtimeOnly). This will display a 500 Internal Service error when hitting ip/haystack/about.

Issue 20: Get "haystack::AuthErr: 404 Not Found"

If all looks good and no error returned when debug this, then make sure to check that "/haystack/" is appended to the host in the URI scheme on the connector. The URI scheme should look something like this: http://10.10.10.100/haystack/ not http://10.10.10.100/

Issue 21: Get "haystack::AuthErr: 403 Authentication failed." with N4 and FIN v2.1.15

This error means that the user didn't enter the correct credentials and/or that specific user doesn't exist in Niagara anymore. Enter correct credentials or create new user if doesn't exist.

Issue 22: Get "auth::AuthErr: 403 Authentication failed." sporadically with sporadically with N4 and FIN

This error means that the authentication to Niagara failed. Tridium has implemented a new auto-logoff property on the users in N4. What this would do is auto-logoff the user after a certain time has elapsed. The connectors in FIN would fault briefly for a few seconds and then be back to normal when it reconnects to the station. To resolve this, the user has to disable the auto-logoff property on the user and it would then be fine since it won't have to try and reconnect to the station.

Issue 23: CPU usage is high, which sometimes can cause points to go into fault for couple seconds

This can be few different things. It could mean you have a lot of points and a lot of them being in a watch either by being trended or used in a program. Or another thing is the polling is to frequent for the JACE. To help tune this, the user can add a number tag called "haystackPollFreq" set to maybe 10-15 seconds (default is 1sec, which can be to frequent if JACE can't handle it). Make sure to include unit "s" for seconds when setting value. Then disable and re-enable the connector.