Announcement

Collapse
No announcement yet.

Discussions related to HomeSeer Beta builds 3.0.0.369->3.0.0.423

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

    Originally posted by concordseer View Post
    Similar setup here except Raspbian Stretch Lite on Mono 5.2. Similar uptime too. Where have you pulled the memory usage from. Have you run Htop in a terminal session.

    From the about screen. I can take a look from terminal after work


    Sent from my iPhone using Tapatalk

    Comment


      Originally posted by Kerat View Post
      From the about screen. I can take a look from terminal after work


      Sent from my iPhone using Tapatalk
      I wouldn't pay too much attention to the About memory usage. Htop will give you a better idea on how memory usage is allotted. You probably are aware of that anyway. It lets you see how each plugin is performing and how quickly each one is rising.

      How many different plugins are you running. I'm running Netcam, CM15 and Concord.

      Comment


        I have seen that in the Homeseer about page the Mb of memory matches the HTop HSConsole Mb of memory.

        See attached picture. Note too that the about page does not show plugins memory in Linux.

        Day #2 here running on Linux and HS3 is doing better today at V.383 than previous versions from what I can see. This box is using an SD card for all of the OS and RW's. The RW's on an SD card are much slower and more resource intensive than an SSD and do hit overall CPU / RAM utilization which in turn will provide less resources for Homeseer 3 to run. This is a best guesstimate as I do bench marks on the iSeries Homeseer Computer with a much faster CPU and much more RAM. The RPI is a quad core CPU running in 32bit mode with 1Gb of RAM and a slow SD card for OS media.

        HTop as mentioned above also shows individual plugins memory. Virtual memory per Linux folks should not be utilized as a bench mark.
        Total memory is 560M of 1.94 Gb of RAM. This is total RAM used by the OS and anything running. This reference should only be used if you are getting close to the total amount of RAM and shouldn't really have much to do with Homeseer 3 running unless Homeseer 3 is eating up a large percentage of total RAM. If the total RAM is exceeded then you will see Swap memory being utilized. If graphing / logging the up and down is normal with Homeseer Console and application plugins. If Mb of memory is not released and keeps climbing then it is an issue.

        In this screen picture: HSConsole is using 97.4 Mb, OmniPlugin is using 81Mb, ZWave is using 150Mb and BLL lan is using 51 Mb.
        Wax and wane with an incremental slow up was HSConsole until V.383. ZWave was similiar but higher with previous versions of HS3.
        I have not had the issue of Homeseer 3 crashing on two computers to date. I am looking to test on an RPi 32 bit Wheezy, Jessie or Stretch as I have only tested on Ubuntu 16.04 64 bit on an ARM (2Gb RAM) and Intel CPU (16Gb RAM).

        [ATTACH]64376[/ATTACH]

        Note too that Homeseer Memory used is rounded off at 98 Mb. At over 100 Mb then you see the Mb numbers.

        HTop HSConsole memory used is not rounded off and is at: 97436 which really is 97.44 Mb used.

        In another thread here folks are stating that current version / beta version is no longer listed under managing plugins. Personally never paid much attention to this and noticed the issue this morning with V.383 running on Linux (so that it is a Windows and Linux Homeseer 3 issue).

        Will add another two plugins (X10 and WeatherXML) for a total of 5 and 3 HSTouch screens to this build of Homeseer lite in the next few days.

        [ATTACH]64379[/ATTACH]
        [ATTACH]64380[/ATTACH]

        Added in the CM11A X10 plugin this morning 10th of November, 2017. It is using around 40Mb. It did add to total ram which was 560Mb and now is 600Mb.
        Note too that all I am doing is visual bean counting here using HTop. No logging or graphing of utilization.
        Last edited by Pete; November 10, 2017, 10:20 AM.
        - Pete

        Auto mator
        Homeseer 3 Pro - 3.0.0.548 (Linux) - Ubuntu 18.04/W7e 64 bit Intel Haswell CPU 16Gb
        Homeseer Zee2 (Lite) - 3.0.0.548 (Linux) - Ubuntu 18.04/W7e - CherryTrail x5-Z8350 BeeLink 4Gb BT3 Pro
        HS4 Lite - Ubuntu 22.04 / Lenovo Tiny M900 / 32Gb Ram

        HS4 Pro - V4.1.18.1 - Ubuntu 22.04 / Lenova Tiny M900 / 32Gb Ram
        HSTouch on Intel tabletop tablets (Jogglers) - Asus AIO - Windows 11

        X10, UPB, Zigbee, ZWave and Wifi MQTT automation-Tasmota-Espurna. OmniPro 2, Russound zoned audio, Alexa, Cheaper RFID, W800 and Home Assistant

        Comment


          Odd memory footprint

          I have a strange memory footprint on my hs3 server. When all is running quiet since let's say 2 to 3 days, w10 performs a kind of optimisation and shrinks memory use as you can see below. My metrics are coming from a child process of tasklist. This is consistent with w10 task manager and with the brilliant app "process explorer" from microsoft sysinternals, so I am confident on the reality of phenomenum...
          I knocked at pro's doors for explanations with no success. If anyone can give a clue it would be fine cause its iching my brain. Anyway at the moment HS3 memory management is ok for me.



          red nodejs, blue HS3, green HSPI plugins, black nginx

          JMJ
          Current Date/Time: 10/11/2017 17:08:37
          HomeSeer Version: HS3 Pro Edition 3.0.0.382
          Operating System: Microsoft Windows 10 Professionnel - Work Station
          System Uptime: 6 Days 9 Hours 24 Minutes 38 Seconds
          IP Address: 192.168....
          Number of Devices: 133
          Number of Events: 49
          Available Threads: 400
          HSTouch Enabled: True
          Event Threads: 0
          Event Trigger Eval Queue: 0
          Event Trigger Priority Eval Queue: 0
          Device Exec Queue: 0
          HSTouch Event Queue: 0
          Email Send Queue: 0
          Anti Virus Installed: Windows Defender

          Enabled Plug-Ins
          30.0.0.36: RFXCOM
          3.0.1.152: Z-Wave
          Last edited by jmj09; November 10, 2017, 11:34 AM.

          Comment


            Well it been System Uptime: 1 Day 16 Hours 3 Minutes 20 Seconds with Beta v383 (Win 10) when I started my HS3.exe was around 133,096 (KB) in size and now my system is fluctuating around 178,500-180,284 (KB), my HS3 "about" states HomeSeer Memory Used: 176 Mbyte. So we're in the ball park.

            Obviously this beats my v3.68 memory leak that would clock-in at over 1,310,936 in less than a 24 hour period, so kudos to Rich for plugging the memory leak.

            @Rich, for Window users is there some tools/app that you suggest that we use to standardize how we monitor and report metrics that mean something to you whilst you debug/develop the next update? The reason I ask is that @jmj09 is using some (Windows) graphing tool(s) to show his metrics which is obviously different than what others are showing.

            @jmj09, would you be so kind and explain what your using to track your metrics and in particular why did you pick the four items that your currently displaying?

            Ok dumb question, can someone explain to this poor HS3 "peasant" what the following HS3 units means and what are the normal upper/lower bounds one should expect? Thanxs in advance Mike
            • Threads
            • Handles
            • Modules
            • Process Working Set
            Computer: CUK Intel NUC7i7BNH
            Op System: Windows10 Pro - Work Station
            HS Version: HS4 Pro Edition 4.2.19.0

            Plug-Ins: AK Weather 4.0.5.25,APCUPSD 3.3.2.3,BLBackup 2.0.63.0,BLEditor 2.0.11.0,BLGData 3.0.55.0,BLOccupied 2.0.28.0,BLShutdown 1.0.6.0,Blue-Iris 3.1.3.33206,Device History 3.2.0.2,EasyTrigger 3.0.0.76,Harmony Hub 4.0.14.0,iRobot 5.22.41.1,JowiHue 4.0.8.7,Nest 3.0.0.34,NetCam Plugin 1.0.0.5,PHLocation2 3.0.0.64,Pushover 4.0.10.0,Random 3.0.0.2,Restart 1.0.0.7,Ring 1.0.0.9,SDJ-Health 3.1.1.3,Sonos 3.1.0.59,Sonos4 4.0.1.12,UltraCID3 3.0.6681.34300,UltraMon3 3.0.6554.33094,UltraNetCam3 3.0.6413.20219,Unifi 4.0.32.0,Zigbee 4.0.11.0,Z-Wave 4.0.3.0, and Jon00 scripts.

            Comment


              HS3 is a .NET application and at times the system will do a "garbage collection" and free up resources that are not longer being used. This is normal with .NET and I suspect that is what you are seeing.

              Originally posted by jmj09 View Post
              I have a strange memory footprint on my hs3 server. When all is running quiet since let's say 2 to 3 days, w10 performs a kind of optimisation and shrinks memory use as you can see below. My metrics are coming from a child process of tasklist. This is consistent with w10 task manager and with the brilliant app "process explorer" from microsoft sysinternals, so I am confident on the reality of phenomenum...
              I knocked at pro's doors for explanations with no success. If anyone can give a clue it would be fine cause its iching my brain. Anyway at the moment HS3 memory management is ok for me.



              red nodejs, blue HS3, green HSPI plugins, black nginx

              JMJ
              Current Date/Time: 10/11/2017 17:08:37
              HomeSeer Version: HS3 Pro Edition 3.0.0.382
              Operating System: Microsoft Windows 10 Professionnel - Work Station
              System Uptime: 6 Days 9 Hours 24 Minutes 38 Seconds
              IP Address: 192.168....
              Number of Devices: 133
              Number of Events: 49
              Available Threads: 400
              HSTouch Enabled: True
              Event Threads: 0
              Event Trigger Eval Queue: 0
              Event Trigger Priority Eval Queue: 0
              Device Exec Queue: 0
              HSTouch Event Queue: 0
              Email Send Queue: 0
              Anti Virus Installed: Windows Defender

              Enabled Plug-Ins
              30.0.0.36: RFXCOM
              3.0.1.152: Z-Wave
              💁‍♂️ Support & Customer Service 🙋‍♂️ Sales Questions 🛒 Shop HomeSeer Products

              Comment


                Thats right for .net softwares, but how about nginx and nodejs in sync ?
                For nodejs I can monitor internal garbage collection "small" and "big" ones, and the shrinking process appears at a different pace.
                And at the end, HS3 is encapsulated in less than 30k and still healthy, this is fine but I fear a border effect to burst any day in my family consensus.
                A bit of paranoia dont hurt, does it ?
                JMJ

                Comment


                  Originally posted by rprade View Post
                  [ATTACH]64010[/ATTACH]
                  Randy, I see that your using Jon's Database Charting utility... setup questions; what are you using to monitoring your CPU % , I'm assuming "Jon00PerfMon" or are you using something else? Secondly what update times(s) are you using in regards to graphing and devices? I'm assuming that your running the Charting plugin continuously or only during specific test monitoring time periods?
                  Computer: CUK Intel NUC7i7BNH
                  Op System: Windows10 Pro - Work Station
                  HS Version: HS4 Pro Edition 4.2.19.0

                  Plug-Ins: AK Weather 4.0.5.25,APCUPSD 3.3.2.3,BLBackup 2.0.63.0,BLEditor 2.0.11.0,BLGData 3.0.55.0,BLOccupied 2.0.28.0,BLShutdown 1.0.6.0,Blue-Iris 3.1.3.33206,Device History 3.2.0.2,EasyTrigger 3.0.0.76,Harmony Hub 4.0.14.0,iRobot 5.22.41.1,JowiHue 4.0.8.7,Nest 3.0.0.34,NetCam Plugin 1.0.0.5,PHLocation2 3.0.0.64,Pushover 4.0.10.0,Random 3.0.0.2,Restart 1.0.0.7,Ring 1.0.0.9,SDJ-Health 3.1.1.3,Sonos 3.1.0.59,Sonos4 4.0.1.12,UltraCID3 3.0.6681.34300,UltraMon3 3.0.6554.33094,UltraNetCam3 3.0.6413.20219,Unifi 4.0.32.0,Zigbee 4.0.11.0,Z-Wave 4.0.3.0, and Jon00 scripts.

                  Comment


                    Originally posted by rprade View Post
                    I have had 4 instances of CPU utilization going through the ceiling in the last several weeks. Each time it was the Z-Wave plug-in that was consuming the extra, going from a normal 1-3% CPU to over 50%. Disabling/enabling the plug-in cures it. I have yet to figure out what triggers it in the first place. The most recent time it did it was at 1:30PM last Tuesday. At 3:30PM HS CPU spiked to 65% and that is when an event automatically disabled/enabled the Z-Wave plug-in.


                    [ATTACH]64010[/ATTACH]
                    Randy, I see that your using Jon's Database Charting utility... setup questions; what are you using to monitoring your CPU % , I'm assuming "Jon00PerfMon" or are you using something else? Secondly what update times(s) are you using in regards to graphing and devices? I'm assuming that your running the Charting plugin continuously or only during specific test monitoring time periods?
                    Computer: CUK Intel NUC7i7BNH
                    Op System: Windows10 Pro - Work Station
                    HS Version: HS4 Pro Edition 4.2.19.0

                    Plug-Ins: AK Weather 4.0.5.25,APCUPSD 3.3.2.3,BLBackup 2.0.63.0,BLEditor 2.0.11.0,BLGData 3.0.55.0,BLOccupied 2.0.28.0,BLShutdown 1.0.6.0,Blue-Iris 3.1.3.33206,Device History 3.2.0.2,EasyTrigger 3.0.0.76,Harmony Hub 4.0.14.0,iRobot 5.22.41.1,JowiHue 4.0.8.7,Nest 3.0.0.34,NetCam Plugin 1.0.0.5,PHLocation2 3.0.0.64,Pushover 4.0.10.0,Random 3.0.0.2,Restart 1.0.0.7,Ring 1.0.0.9,SDJ-Health 3.1.1.3,Sonos 3.1.0.59,Sonos4 4.0.1.12,UltraCID3 3.0.6681.34300,UltraMon3 3.0.6554.33094,UltraNetCam3 3.0.6413.20219,Unifi 4.0.32.0,Zigbee 4.0.11.0,Z-Wave 4.0.3.0, and Jon00 scripts.

                    Comment


                      Originally posted by MNB View Post
                      Randy, I see that your using Jon's Database Charting utility... setup questions; what are you using to monitoring your CPU % , I'm assuming "Jon00PerfMon" or are you using something else? Secondly what update times(s) are you using in regards to graphing and devices? I'm assuming that your running the Charting plugin continuously or only during specific test monitoring time periods?
                      I am using Jon00's Database Charting and PerfMon. The database charting is full time, one sample every 5 minutes and it just samples the PerfMon devices. The charting lets me go back in history up to 180 days and the PerfMon graphic gives me a 24 hour snapshot.

                      I also generate Pushover messages when the CPU climbs above 40%. If it remains above 60% for more than 5 minutes, HomeSeer and the computer are restarted. I've never seen a memory problem.

                      My Z-Wave plug-in has been going nuts once or twice a week for a few weeks running 3.0.1.150. While it runs about 2-5% normally, when it has a problem it can gobble 40-60% of the CPU. I just updated to 3.0.1.155 yesterday.
                      HS4 Pro, 4.2.19.0 Windows 10 pro, Supermicro LP Xeon

                      Comment


                        I am using Jon00's Database Charting and PerfMon. The database charting is full time, one sample every 5 minutes and it just samples the PerfMon devices. The charting lets me go back in history up to 180 days and the PerfMon graphic gives me a 24 hour snapshot.
                        Firstly thanxs for the comeback. I'm assuming that your PerfMon devices are also sampling every 5 minutes?

                        I also generate Pushover messages when the CPU climbs above 40%. If it remains above 60% for more than 5 minutes, HomeSeer and the computer are restarted. I've never seen a memory problem.
                        I like your CPU threshold idea(s) and may also adopt something similar myself.

                        What other database(s) have you setup and what are tracking that's useful?

                        I was able to setup a "Manual" event that allows me to start the Database Charting plugin but have yet to figure how to turn it off other than going to the tray icon and exit.
                        Computer: CUK Intel NUC7i7BNH
                        Op System: Windows10 Pro - Work Station
                        HS Version: HS4 Pro Edition 4.2.19.0

                        Plug-Ins: AK Weather 4.0.5.25,APCUPSD 3.3.2.3,BLBackup 2.0.63.0,BLEditor 2.0.11.0,BLGData 3.0.55.0,BLOccupied 2.0.28.0,BLShutdown 1.0.6.0,Blue-Iris 3.1.3.33206,Device History 3.2.0.2,EasyTrigger 3.0.0.76,Harmony Hub 4.0.14.0,iRobot 5.22.41.1,JowiHue 4.0.8.7,Nest 3.0.0.34,NetCam Plugin 1.0.0.5,PHLocation2 3.0.0.64,Pushover 4.0.10.0,Random 3.0.0.2,Restart 1.0.0.7,Ring 1.0.0.9,SDJ-Health 3.1.1.3,Sonos 3.1.0.59,Sonos4 4.0.1.12,UltraCID3 3.0.6681.34300,UltraMon3 3.0.6554.33094,UltraNetCam3 3.0.6413.20219,Unifi 4.0.32.0,Zigbee 4.0.11.0,Z-Wave 4.0.3.0, and Jon00 scripts.

                        Comment


                          Originally posted by MNB View Post
                          Firstly thanxs for the comeback. I'm assuming that your PerfMon devices are also sampling every 5 minutes?



                          I like your CPU threshold idea(s) and may also adopt something similar myself.

                          What other database(s) have you setup and what are tracking that's useful?

                          I was able to setup a "Manual" event that allows me to start the Database Charting plugin but have yet to figure how to turn it off other than going to the tray icon and exit.
                          Yes, every 5 minutes.

                          I initially was restarting the Z-Wave plug-in, but I changed it to a total system restart in case the excessive CPU usage was something else. My system restart routine reboots all three Z-Nets, shuts down HomeSeer and reboots the HomeSeer server. Any time I restart the system I use the same routine.

                          Since the DB Charting database allows for up to ten sources per database I picked 10 from PerfMon to store in the charting program. I also use the DB charting to chart various energy and temperature metrics to track heating/hot water and air conditioning temperatures and household energy consumption.
                          HS4 Pro, 4.2.19.0 Windows 10 pro, Supermicro LP Xeon

                          Comment


                            Originally posted by rprade View Post
                            Yes, every 5 minutes.

                            Since the DB Charting database allows for up to ten sources per database I picked 10 from PerfMon to store in the charting program. I also use the DB charting to chart various energy and temperature metrics to track heating/hot water and air conditioning temperatures and household energy consumption.
                            Thanxs Randy. It took a little bit to get my head around how to make the DB Charting database plugin to work.. I think setting up a temperature database and a energy consumption database should give me a leg up, as always your a great knowledge resource! Mike
                            Computer: CUK Intel NUC7i7BNH
                            Op System: Windows10 Pro - Work Station
                            HS Version: HS4 Pro Edition 4.2.19.0

                            Plug-Ins: AK Weather 4.0.5.25,APCUPSD 3.3.2.3,BLBackup 2.0.63.0,BLEditor 2.0.11.0,BLGData 3.0.55.0,BLOccupied 2.0.28.0,BLShutdown 1.0.6.0,Blue-Iris 3.1.3.33206,Device History 3.2.0.2,EasyTrigger 3.0.0.76,Harmony Hub 4.0.14.0,iRobot 5.22.41.1,JowiHue 4.0.8.7,Nest 3.0.0.34,NetCam Plugin 1.0.0.5,PHLocation2 3.0.0.64,Pushover 4.0.10.0,Random 3.0.0.2,Restart 1.0.0.7,Ring 1.0.0.9,SDJ-Health 3.1.1.3,Sonos 3.1.0.59,Sonos4 4.0.1.12,UltraCID3 3.0.6681.34300,UltraMon3 3.0.6554.33094,UltraNetCam3 3.0.6413.20219,Unifi 4.0.32.0,Zigbee 4.0.11.0,Z-Wave 4.0.3.0, and Jon00 scripts.

                            Comment


                              After about 3 days running .383 I had one processor lock up again.



                              Current Date/Time: 11/11/2017 5:14:34 PM
                              HomeSeer Version: HS3 Pro Edition 3.0.0.383
                              Linux version: Linux HS3 4.4.0-98-generic #121-Ubuntu SMP Tue Oct 10 14:24:03 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux System Uptime: 0 Days 0 Hours 42 Minutes 59 Seconds
                              Number of Devices: 343
                              Number of Events: 90
                              Available Threads: 800
                              HSTouch Enabled: True
                              Event Threads: 0
                              Event Trigger Eval Queue: 0
                              Event Trigger Priority Eval Queue: 0
                              Device Exec Queue: 0
                              HSTouch Event Queue: 0
                              Email Send Queue: 0
                              Anti Virus Installed:

                              Enabled Plug-Ins
                              2.0.35.0: BLLAN
                              3.0.0.56: DSC Security
                              3.0.1.9: Kodi
                              1.2.0.0: Monoprice Amp
                              3.0.6510.21140: Ultra1Wire3
                              3.0.0.76: weatherXML
                              3.0.0.36: X10
                              0.0.0.6: YAMAHA-RECEIVER
                              3.0.1.154: Z-Wave
                              Attached Files
                              https://forums.homeseer.com/forum/de...plifier-plugin

                              Comment


                                I get cpu spikes a couple of times a day. Some of them are half iowait, most are pure cpu. So far I really haven't been able to tie it to anything. It happened about an hour ago, but I didn't notice it until it was over. The spikes typically last about an hour and the system and HS3 recovers OK. During the spike things are sluggish (slow response to motion sensors for example).

                                But so far .383 continues to be stable with very minimal memory increase with the set of plugins I have enabled.

                                If I try to add any of the following plugins to the current setup, I start getting callback queue errors: HSBuddy, EasyTrigger, RFXCOM. However they do not seem to have any significant or lasting impact on memory utilization. I have not tried disabling everything else and only enabling one of these at a time.

                                Current Date/Time: 11/12/2017 11:52:05 AM
                                HomeSeer Version: HS3 Pro Edition 3.0.0.383
                                Linux version: Linux rpi5855 4.9.35-v7+ #1014 SMP Fri Jun 30 14:47:43 BST 2017 armv7l GNU/Linux System Uptime: 2 Days 13 Hours 58 Minutes 12 Seconds
                                IP Address:
                                Number of Devices: 342
                                Number of Events: 184
                                Available Threads: 399
                                HSTouch Enabled: False
                                Event Threads: 0
                                Event Trigger Eval Queue: 0
                                Event Trigger Priority Eval Queue: 0
                                Device Exec Queue: 0
                                HSTouch Event Queue: 0
                                Email Send Queue: 0
                                Anti Virus Installed:

                                Enabled Plug-Ins
                                2.0.71.0: BLSpeech
                                1.2.0.0: Device History
                                3.0.1.90: PHLocation
                                3.0.6413.20219: UltraNetCam3

                                Remote plugins
                                BLLAN
                                BLRF
                                JowiHue
                                Z-Wave

                                Comment

                                Working...
                                X