Announcement

Collapse
No announcement yet.

Unscheduled watering following HS3 restart

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

    Unscheduled watering following HS3 restart

    Michael,

    I mentioned in another thread that the plugin starts an unscheduled watering cycle immediately after a HS3 restart--and possibly after a PI disable/enable, though I've not tested it. As far as I can tell it happens every time. My HS3 server is restarted relatively frequently for a variety of reasons, and the excess watering is problematic. It's particularly frustrating because manually stopping the cycle after a restart--if I happen to remember or notice--doesn't solve the problem; if the Area is set for round-robin cycling, it repeatedly restarts watering even after a manual stop. My start-times are all hard time-of-day, not calculated, so it seems this issue should be preventable.

    Can you please implement a fix for this? Thank you.
    -Wade

    #2
    A cycle is based upon the state as viewed at startup. State is save every minute. Can you post the debug logs for for to Hs restart and after Hs restart.

    Comment


      #3
      I was incorrect that this happens every time. I've just tried to replicate the issue and everything went back to normal after a restart. I'll try to document what is going on around the time it happens next time I observe the issue. Only thing I recall being different last time is there were rain inhibits.
      -Wade

      Comment


        #4
        Michael,

        After an orderly HS3 server reboot this morning following a windows update, an area ran unscheduled. I am emailing the log files from prior to the reboot and the new file created after. Area 3 (Garden Drip) is scheduled to run every other morning at 5:00am in 4 round robin cycles, and it ran this morning as scheduled. After the reboot at ~10:54am, it started running again.

        As background and to explain what may be in the older log file... Area 3 was running daily until I changed it yesterday to begin running every other day beginning on April 23 (tomorrow) due to it being over-watered. I expected this would cause it to not water today. Instead, it began watering this morning as per the previous schedule. (This is likely by design, I just didn't realize it would work that way). So this morning before end of the scheduled run, I stopped Area 3 and changed the every other day schedule to begin April 24 (Friday). I checked the run list and it showed correctly that next run was scheduled for Friday morning 5:00am. A few hours later the windows update/reboot problem happened.

        I've emailed the two log files to you. You will see that I've manually stopped Area 3 twice in the past few minutes. It will restart 2 more times (3rd and 4th round robin cycles) despite me manually stopping it as round robin scheduler and manual stop feature don't appear to be aware of one another.

        Thank you for looking into it.


        edit: After stopping the 4th round robin cycle, Area 3 Garden now correctly shows in the Run List for Friday morning at 5:00am.
        -Wade

        Comment


          #5
          Another thing I realized about my configuration: I had maximum days without watering set to 1 day for Area 3. Not sure how the 1 day is calculated (not more than 24 hours between waterings, or not more than 1 full day between watering days--so every other day?). In any case I don't think this is the cause of the issue but wanted to call attention to it in case it's related. I set it to 1 day about a week ago but have observed the issue with reboot for quite some time. I've now changed it to 2 days.
          -Wade

          Comment


            #6
            I understand the issue, but solution will take some consideration. In your case it occurs when there are more than one round robin area scheduled for the day and a restart occurs after the irrigation had previously happened during the day.

            Round Robin has caused problems after problems because the original design did not have provisions such such a mode. It really complicates things when going outside the basic scheduling structure. I'm not indicating that round robin is not a good thing, but that it just does not fit with the original design and with so many users I do not want to start over with a design that has the basic schedule interval including a duration that is split among multiple cycles.

            I will also look into the stop operation you described above. A user should not need to stop each RR cycle.

            A day is as viewed on the calendar and not by a number of hours.

            Comment


              #7
              Ok thank you.
              -Wade

              Comment


                #8
                I made the updates to handle the restarts with round robin. 2.20.0.11. Available at http://mcsSprinklers.com/mcsSprinkle...0_0_11_HS3.zip or http://mcsSprinklers.com/setup.zip. The logic updated in mcsSprinklers_2009.dll. Version number updated in HSPI_MCSPRINKLERS.exe

                I was confused for awhile with your setup of R15 only running in the first of two program times. I also tried to replicate the Stop problem you reported and had no luck. I was trying it on the version after the restart was fixed. I tried the Zone Status page, Stop All. I also tried the HS Area Stop button. Both resulted in the desired stop. I also looked at you debug and never saw any commands from HS (which show up in debug log starting with "Setio". If you do continue to have a problem with stopping an Area that has round robin then provide the debug and the steps you used.

                Comment


                  #9
                  Originally posted by Michael McSharry View Post
                  I made the updates to handle the restarts with round robin. 2.20.0.11. Available at http://mcsSprinklers.com/mcsSprinkle...0_0_11_HS3.zip or http://mcsSprinklers.com/setup.zip. The logic updated in mcsSprinklers_2009.dll. Version number updated in HSPI_MCSPRINKLERS.exe
                  Thanks. I'll give it a go.

                  Originally posted by Michael McSharry View Post
                  I was confused for awhile with your setup of R15 only running in the first of two program times. I also tried to replicate the Stop problem you reported and had no luck. I was trying it on the version after the restart was fixed. I tried the Zone Status page, Stop All. I also tried the HS Area Stop button. Both resulted in the desired stop. I also looked at you debug and never saw any commands from HS (which show up in debug log starting with "Setio". If you do continue to have a problem with stopping an Area that has round robin then provide the debug and the steps you used.
                  The only way I've ever stopped a watering cycle is by clicking "Stop Area X" on the Touch Area interface. I'll test it further. Odd that you didn't see evidence in the log. I stopped it twice before downloading the log. I know it stopped and restarted because I have a trouble-shooting event that emails me with every area start and area stop--triggered by change in any of the HS Area devices.
                  -Wade

                  Comment


                    #10
                    Michael, after testing and writing up the below I realized I probably don't have the correct beta. I downloaded from the link above (http://mcsSprinklers.com/mcsSprinkle...0_0_11_HS3.zip) and it is showing 2.20.0.10 rather than .11. That probably explains the result. Please verify the files at the posted link. I don't *think* I could have mixed it up on my end, but I've certainly made worse mistakes.

                    I've not bothered emailing the logs until I straighten out the version.

                    -----------------------------------------------------------------------------------------------------

                    Michael, here's my process and the results. Times approximate, to aid review of the logs if needed.

                    1:50p
                    • Downloaded log.(~1:50pm)
                    • Stopped the plugin.
                    • Copied the new mcsSprinklers_2009.dll and HSPI_MCSPRINKLERS.exe into my HS3 root directory.
                    1:52
                    • Restarted the plugin.
                    1:53
                    • Area 4 started an unscheduled cycle.
                    1:56
                    • Opened Touch Area interface and clicked Stop Area 4. (~1:56pm)
                    • Area 4 stopped and did not restart ~23.5 minutes later (next round robin cycle) as was happening previously.
                    3:08
                    • Downloaded log.

                    The version number is updated to 2.20.0.10 on the General Status page but still shows 2.20.0.5 in the HS3 plugin manager.

                    Thanks.
                    -Wade

                    Comment


                      #11
                      HSPI_XXX.exe contains the version number visible from HS. Unless you restart HS it will not look for a new file version. Status of the mcsSprinklers_2009.dll version number. shows up in \html\mcsSprinklers\data\startuptrace.txt. The startuptrace.txt file is the only one that makes a difference. I confirmed it it 2.20.0.11. The mcsSprinklers Other page shows what version is available on mcsSprinklers.com. It pulls the date at midnight or after use of refresh button.

                      Assuming you have 2.20.0.11 can you provide the debug log?

                      Comment


                        #12
                        Originally posted by Michael McSharry View Post
                        HSPI_XXX.exe contains the version number visible from HS. Unless you restart HS it will not look for a new file version. Status of the mcsSprinklers_2009.dll version number. shows up in \html\mcsSprinklers\data\startuptrace.txt. The startuptrace.txt file is the only one that makes a difference. I confirmed it it 2.20.0.11. The mcsSprinklers Other page shows what version is available on mcsSprinklers.com. It pulls the date at midnight or after use of refresh button.

                        Assuming you have 2.20.0.11 can you provide the debug log?
                        Startuptrace.txt does in fact show 2.20.20.11. I'm sending the logs now.

                        Code:
                        True 4/23/2020 1:52:57 PM | mcsSprinklers | Version 2.20.0.11 |    V2.20.0.11/2.20.0.11 (Professional)  Copyright 2004-2020 mcsSolutions Registered with Homeseer
                        -Wade

                        Comment


                          #13
                          I think you will find that if you restart again that you will not see the issue. The problem existed with saving the state prior to shutdown and since the older version did this the newer version did what it was told thinking that irrigation had not already occurred for the Area.

                          Comment


                            #14
                            Originally posted by Michael McSharry View Post
                            I think you will find that if you restart again that you will not see the issue. The problem existed with saving the state prior to shutdown and since the older version did this the newer version did what it was told thinking that irrigation had not already occurred for the Area.
                            Michael, I updated the plugin from .11 to .13 via the updater this morning and on re-enabling the PI an area scheduled to run later today immediately started. That area is not configured for round robin cycling.

                            This information is of limited usefulness at this point as I had logging disabled. I've re-enabled logging and follow up if/when I see the behavior again.
                            -Wade

                            Comment


                              #15
                              Originally posted by Michael McSharry View Post
                              I think you will find that if you restart again that you will not see the issue. The problem existed with saving the state prior to shutdown and since the older version did this the newer version did what it was told thinking that irrigation had not already occurred for the Area.
                              After a HS3 restart this morning an unscheduled area cycle started. I've emailed the log.
                              -Wade

                              Comment

                              Working...
                              X