Announcement

Collapse
No announcement yet.

408 Request Timeout Errors

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    408 Request Timeout Errors

    Originally posted by spud View Post

    In version 3.0.0.2 I have set the default poll interval to 300secs, I will contact Rainforest to see what they recommend.
    I will move my discussion to the new dedicated forum, Thanks!

    I am curious about Rainforest's response to the timeouts - it must be something on their server-side. Even with polling at 300s, I am still getting a lot of the 408 errors randomly. Just now looking through my log - I had (1) 408 error about 1am, then none until 6:40am, then (21) between 6:40am and 7:10am.

    #2
    I'm getting the same errors. I was at 60 seconds, I just updated the plugin to .2 and changed the value to 300, I will see if this helps. Also I notice that I didn't have any data in the homeseer energy db for this device (rainforest root) for the past 2 days
    Attached Files

    Comment


      #3
      With the legacy Rainforest Eagle I noticed with Spud's Plugin and the Python script that I wrote it seems to have a memory leak when you use the Cloud API. It is fine in the beginning and then slows down and you receive these timeout's. Even the local WebUI stops working.

      It looks like the cloud just proxies the API's and each call hits the eagle. After time this leak happens, eventually it stops responding all together. There are lots of reports of people using many different 3rd party apps and seeing it. The local non-REST API worked fine but I can't use it anymore due to the way my home is configured.

      I gave up using the Legacy Rainforest Eagle, bought a Eagle-200 and it has been rock solid. They changed the way the cloud API works, the Rainforest cloud always gets your data and then it sends it to other clouds. The API using the data in the cloud so it does not re-query the Eagle. The Eagle seems to hum away without issues.

      Comment


        #4
        Can you give some more context, why the local non-REST API (I am assuming you are were using their local socket connection?) doesn't work with the way your home is configured?

        Comment


          #5
          Originally posted by kevini View Post
          With the legacy Rainforest Eagle I noticed with Spud's Plugin and the Python script that I wrote it seems to have a memory leak when you use the Cloud API. It is fine in the beginning and then slows down and you receive these timeout's. Even the local WebUI stops working.
          This exactly describes the behavior that I am seeing. Funny that I didn't notice it before but maybe I never really looked.

          I gave up using the Legacy Rainforest Eagle, bought a Eagle-200 and it has been rock solid. They changed the way the cloud API works, the Rainforest cloud always gets your data and then it sends it to other clouds. The API using the data in the cloud so it does not re-query the Eagle. The Eagle seems to hum away without issues.
          I would be willing to give the Eagle-200 a try, if it resolves this issue.

          DSteiNeuro

          HS3Pro

          MSI Cubi Intel(R) Core(TM) i5-5200U CPU @ 2.20GHz, 2201 Mhz, 2 Core(s), 4 Logical Processor(s) 16GB DDRl RAM

          Enabled Plug-Ins
          BLRussound, BLSpeech, HSTouch Server, JowiHue, MyQ, Nest, Rain8, Squeezebox, Ultra1Wire3, UltraGCIR3, Vista Alarm, X10,Z-Wave

          Comment


            #6
            I've been running the plugin against my -200 since last week. Still rock solid no issues.

            My own script worked for a month too with no issues.

            The only stable ways with the legacy eagle is the local interface or the push. You can only push to one cloud at a time though, unlike the -200 which pushes the the rainforest cloud and then forks from there.

            Comment


              #7
              I hope that Spud is able to utilize the legacy Eagle local interface - to remove the dependence on the Rainforest Cloud. I’m polling at 300s now, and it has reduced the timeouts, but polling at 1 min would be preferred.

              Comment


                #8
                I fully agree and hope Spud can get either push or the local interface working... The issue that I have ran into seems to be common and because of the cloud. Personally I don't care if I have to use the push interface, as long the data is accurate and reliable. My personal feeling is that having a cloud option is nice however what happens when the internet goes down? Does the eagle-200 store the history and then fill in the missed the polls. I could be wrong but I don't think it does. Ultimately some error recovery or history ingestion would be ideal... I, and many others stayed away from the -200 because they pushed the cloud as the only method of connection I believe. There may be other local connections available now for the 200 but its been a long time since I've looked into it.

                Comment


                  #9
                  Well I got curious, and took a look at their documentation... It looks like they removed their local connection documentation for the legacy eagle:
                  https://rainforestautomation.com/support/developer/

                  I'll see if I downloaded a copy of it somewhere. They do have documentation though for the legacy eagle and using the "push method"

                  Interestingly enough it looks like the eagle-200 supports a local HTTP connection method and associated documentation. This has me interested enough to consider an upgrade to the 200 in the future. It would be awesome to have the 200 using the cloud api but having a secondary connection method in case the internet or cloud goes down, so you don't miss polling. In addition to this if possible getting the stored history in case of longer outages would be the holy grail!

                  Comment


                    #10
                    I have asked Rainforest Automation about these timeout issues, as well as local API, but I did not get any response yet.

                    I have just released the beta version 3.0.0.3 that fix the problem of the Eagle not appearing in the HSTouch energy screen.
                    To make this work you will need to delete your current RainforestEagle devices, and then restart the plugin so that it recreates it.

                    Click image for larger version

Name:	image1.PNG
Views:	440
Size:	73.0 KB
ID:	1251236

                    Comment


                      #11
                      Thanks Spud. I downloaded the 3.0.0.3 update, deleted the old devices and restarted the plugin. The new devices were created in HS3, and the Eagle shows up in HSTouch Energy.

                      As skarragallagher mentioned, early word was that the local interface was eliminated from the -200 Eagle device. I hadn't even considered upgrading from the legacy device based on that. If Rainforest has changed that, it might be worth a look. I'm hopeful that Rainforest will get back with you though, and have a solution to the timeouts.

                      Comment


                        #12
                        I am also noticing 408 timeouts on a Eagle Legacy-100

                        The plugin itself does not give any indication that there is a problem. It would be nice if we could test for this in an event and create some kind of alert that the data has not been received for XXX seconds.

                        Looking at the Plug-In management page everything looks good, no indication of any sort of failure. The PI management page should at least turn the icon amber if the PI is not receiving the data.

                        Aside from these minor things I think the PI looks pretty good.



                        Comment


                          #13
                          FYI the RainforestEagle plugin is now in the store and updater

                          I'm still waiting a response from Rainforest about timeout issues and local API.

                          Comment


                            #14
                            Originally posted by kevini View Post
                            I've been running the plugin against my -200 since last week. Still rock solid no issues.

                            My own script worked for a month too with no issues.

                            The only stable ways with the legacy eagle is the local interface or the push. You can only push to one cloud at a time though, unlike the -200 which pushes the the rainforest cloud and then forks from there.
                            Is your -200 still doing well? No timeout issues? I changed my -100 back down to a poll time of 180 sec before I went to bed, and by morning I was getting tons of timeouts again. I looked at my Eagle device and the Cloud LED was blinking. I didn't think to check and see if local access was still working. I had to power-cycle the Eagle to get things working again. I set the poll back to 300 sec, and still get a dozen or so timeouts daily.

                            Comment


                              #15
                              Yes, it has been solid, has not missed a beat and never had a timeout. I have the poll at 30 seconds too.

                              I saw the same issue on my Legacy Eagle with the blinking cloud LED.

                              Comment

                              Working...
                              X