Connectwise Automate Feedback & Questions.

I just installed this plugin today and cannot get it to stay connected. I go into the settings for the plugin and put in my API key and secret and it actually does say "Connected" and in using Fiddler, it does get a successful response from the Sophos API and gives me a valid token with expire time. 

The problem seems to be related to the DB based on the response I got from the POST to the Automate endpoint when it is trying to INSERT into "plugin_sophos_central_pld_auth". The POST responds with a -9999 code. 

This is further shown when I click out of the settings window and go back into it, the key and secret have disappeared. So it looks like it is just not saving them.

I opened a ticket with ConnectWise and they said I would have to get with you guys. Any help would be appreciated. Thanks.

I didnt think I did, but just took a look and theres one customer we no longer have licenses with that somehow has a trial.

I have 403 errors as well.  From what I've seen it's for accounts that we don't have partner access to.  In a future release I'd like to see it smart enough to recognize this and not try to sync these accounts.  I'd settle for a manual process of de-selecting accounts we don't want to attempt to sync.

We've made some good progress at least!  403 errors are also typical when you have trial accounts as it attempts to pull the endpoints for those tenants, but then doesn't have the permissions as we have barred trial accounts from API access.  Would you be able to confirm if this is the case?  If so, I have an enhancement on the backlog to better handle this situation.

Quick followup, after running the SQL query, I can now see computers/locations again. Still getting a lot of 403 errors though

Thanks for the response. .132 also threw 403 errors, however it appeared to be otherwise functional, computers/locations appeared and could run commands against them. I will give that table drop a try and see what happens.

Hello Matt, First, let me say Happy New Year!  Thank your for reaching out and I'm sorry you have been experiencing some issues with the latest version.  Could you just confirm for me that .132 was indeed working and now .134 is throwing the 403 errors?

If this assumption is the case, we have seen this behavior on occasion with a subset of upgrades.  The fix so far, has been to apply the below SQL statement to drop the Sophos tables from the Connectwise database, from within your Connector admin console, and then reboot the machine the Connecttwise instance runs from.  Upon reboot the tables will be re-created and should populate with data once confirmation of the API key is present.

SQL Command: “drop tables plugin_sophos_central_cls_basic, plugin_sophos_central_cs_basic, plugin_sophos_central_ls_basic;”

Please do let me know if this corrects your issue or if we'll need to schedule a remote session to further troubleshoot.

Running the latest 1.0.0.134.

The log still shows near constant 403 errors with no reasons. 

The computer pane is currently empty; if I try to force a sync it fails with "Object reference not set to an instance of an object"

Overall seems like a step backwards in usability from .132

Hi Candace, sorry to hear you are experiencing this issue.  It sounds as though the global auto deploy settings are set from the Clients window.  Could you please private message me your email address, I'd like to set up a troubleshooting call with you, myself, and our SE tech lead for this integration.

Also, could you please confirm what version of the plugin you have installed and if it's not the latest .133 version, could you download that version from the wiki page and get it deployed prior to our meeting as we've made some changes to the deployment workflow and I'd like to have the latest version running to troubleshoot from.

New issue as of today. We had a client's server getting disconnected from the SQL backend - application support believed it to be Sophos. I took a look at the machine in Automate, and as of 12/1, machines with Sophos on them are having a script run multiple times per day. "Sophos Central Install and Register".  This has come out of the blue - I've not done anything with the plugin at all in the past two weeks since I originally installed and set it up.

The only thing I can see that calls this script is the "Sophos - Install Agent' Alert template which is being called by the 'Sophos Auto-Register' Monitor set which is picking up pretty much every single machine we have Sophos installed on. 

Any idea what's going on? Anyone else seeing this? It looks like the monitor set is looking for a service called 'Sophos MCS Agent', but even though it appears in the monitor set as being an issue and not having the service, it absolutely does have it and it shows in the view table correctly.