Anyone still use this plugin?

[Blade]

I am wondering if anyone still uses this plugin

[Guest]

Just wondering. is this plugin useful only to ''patch'' a memory leak?

[Blade]

It used to be run in HS3 to lower the memory the HS3 process used to take - things have likely come a long way since then and I am wondering if anyone is running it in HS4 and if it makes a difference

[Tillsy]

It used to be run in HS3 to lower the memory the HS3 process used to take - things have likely come a long way since then and I am wondering if anyone is running it in HS4 and if it makes a difference

Yes and no - a bunch of us (at least via Facebook groups, talking with developers, etc) have had an ongoing network resource problem. eg if you add some camera feeds, or plugs that make network connections, then anywhere from 12 hours to several days later the -Wave plugin becomes non-responsive as does the HS web server (it will request authentication, but it won't deliver any content) requiring a forced restart.

I've just installed your plugin now and will let you know if it helps 👍

[langenet]

I set it up on my HS3 system a few years ago. My memory sits wonderfully low. I have an event run each night @1:00am. Perhaps I don't need it, but ...

[Blade]

Just wondering if it needs to be converted to HS4 or not

[TallTM]

I use it. I'm not sure if it's effective or not, but my HS3 system is very reliable. I will transition to HS4 in the next few months.
As the saying goes "You don't know what you've got 'til it's gone."

[jon00]

Just wondering if it needs to be converted to HS4 or not

I'm not sure it is worth your time with this one. Rich stated sometime ago the HS3 functionality would not be removed from HS4. There would be an outcry if they did!

[Tillsy]

Yes and no - a bunch of us (at least via Facebook groups, talking with developers, etc) have had an ongoing network resource problem. eg if you add some camera feeds, or plugs that make network connections, then anywhere from 12 hours to several days later the -Wave plugin becomes non-responsive as does the HS web server (it will request authentication, but it won't deliver any content) requiring a forced restart.

I've just installed your plugin now and will let you know if it helps 👍

Blade so I can say this plugin made a 90% difference to my system. HS4 for me starts at around 129MB of memory usage and slowly grows. I find once it hits to 200MB mark there is a mere hour before the system goes down - highest I've spotted while it is still running is around 216MB, but I've never been able to check the system after that because next thing I know the HS4 web service is now non-responsive. Generally within 12-24 hours, friggen pain in the ***. Turning on things that use network traffic seem to be the worse - for example adding camera feeds and having something fetch those camera feeds repeatedly will kill HS4 within hours.

BLGarbage testing for the past week has been a game changer. For around 2.5 days it would start at around 129MB and grow to maybe 169MB before BLGarbage cleaned it back down to 129MB again. This was hourly so I reduced it to 30 minutes to ensure it stayed nice and clean.

Sadly after 2.5 days the memory usage would start growing and BLGarbage didn't seem able to control that - 169MB, 174MB, 182MB, 190MB, 200MB, 216MB, POOF HS4 gone.

So what I now did was have BLGarbage doing the clean every 30 minutes AND I have set HS4 to reboot every night at 4AM. Seems to be a good combination - albiet a sad necessity, under HS3 I could be running for several months until I decided to do Rasbian or HS updates.

Sadly two agos it still happened though - by around 6pm two days ago I saw memory usage was at 210MB and I though ****, I think she's gonna blow. Sure enough next thing I walk down our passage and nothing happened, my wife opens the pantry and nothing happened, yep HS4 has gone bye bye. Had to power cycle.

But yesterday was perfectly fine - I got home quite late in the evening and checked her and she was smack on 129MB.

So two outages in 7 days instead of 7 - one of them avoidable if I had the daily reboot setup at that point.

Still not 100% reliable but a tremendous improvement in compensating for whatever is going wrong in HS4

[Bill Brower]

BLGarbage testing for the past week has been a game changer. For around 2.5 days it would start at around 129MB and grow to maybe 169MB before BLGarbage cleaned it back down to 129MB again. This was hourly so I reduced it to 30 minutes to ensure it stayed nice and clean.

Sadly after 2.5 days the memory usage would start growing and BLGarbage didn't seem able to control that - 169MB, 174MB, 182MB, 190MB, 200MB, 216MB, POOF HS4 gone.

So what I now did was have BLGarbage doing the clean every 30 minutes AND I have set HS4 to reboot every night at 4AM. Seems to be a good combination - albiet a sad necessity, under HS3 I could be running for several months until I decided to do Rasbian or HS updates.

Sadly two agos it still happened though - by around 6pm two days ago I saw memory usage was at 210MB and I though ****, I think she's gonna blow. Sure enough next thing I walk down our passage and nothing happened, my wife opens the pantry and nothing happened, yep HS4 has gone bye bye. Had to power cycle.

But yesterday was perfectly fine - I got home quite late in the evening and checked her and she was smack on 129MB.

So two outages in 7 days instead of 7 - one of them avoidable if I had the daily reboot setup at that point.

Still not 100% reliable but a tremendous improvement in compensating for whatever is going wrong in HS4

When I had issues like that, I would just do an automated nightly HSx restart. Made for a very reliable and responsive system with only two minutes of downtime in the middle of the night.