API Connection has been spotty last three business days

contacts

#1

We have had what appear to be outages of some sort that last about 1 hour when trying to update our contacts. This is causing issues when sending welcome emails, resetting passwords, and other events on our site. We haven't deployed any new code during the time frame we've had these intermittent issues and they resolve on their own with no action from our team.

The response we are getting is

Net::ReadTimeout: Net::ReadTimeout
179 def update_contact(user)
180 contact = Hubspot::Contact.find_by_id(user.integration.hubspot_vid)
181 contact.update!(properties(user))
182 user.integration.hubspot_synced_at = DateTime.now
183 user.integration.save!


#2

Hi @Nick_Baker,

That appears to be logging specific to your setup; do you have the raw response from HubSpot?


#3

I'm not sure if this is related but I'm also getting issues that might be similar which have appeared over the last 5-7 days.

My issue is that all of a sudden I'm getting "Webhook failed due to connection timeout. Will retry" in workflow webhooks for files and responses that were working fine before that. I've never had this error before and nothing has changed in the response times of my server.


#4

Hi @SAlford and @Nick_Baker ,

I realize I never mentioned this before: There weren't any outages of the HubSpot API over the past week or so. You can check on the uptime of HubSpot by navigating to https://status.hubspot.com/.

@SAlford is it possible that your server sleeps during long periods of inactivity? Some hosting providers like Heroku will sleep a service during periods of inactivity, which can potentially cause timeouts. HubSpot retries the webhook a number of times though, so you shouldn't really see permanent failures unless you're seeing server downtime.


#5

@Derek_Gervais Looks like we had a rate limiting issue. Makes sense since we had issues around the same time every day, a point in time which many events occur and we fire off a bunch of requests to hubspot.