Overlapping Times in basal Tab leading to wrong iob? Dev-i

[Ede1975]

It seems tbrs have not been cancelled, Times are overlapping in basal tab Possibly aaps was out of reach of dash a couple of Times during the period from 14:53 to 16:53

[Ede1975]

Update: i tested a couple of dev-i Versions i Compiled over the last days but did not use before, ran into the same issue. Installed Master and had no issue with that Version. Then i finally rebooted the Phone (didnt before), reinstalled latest dev-i afterwards and the issue did not occur again so far

[MilosKozak]

is it possible you changed pump meanwhile?

[Ede1975]

You mean from dash to some other pump or virtual pump? No, definately not.

[MilosKozak]

do you have logs prior this one?

[MilosKozak]

note: where is this from 17:02:30.229 [RxCachedThreadScheduler-53] D/DATABASE: [PumpSyncImplementation.syncTemporaryBasalWithPumpId():274]: Updated TemporaryBasal TemporaryBasal(id=28452, version=3, dateCreated=1649602950162, isValid=true, referenceId=null, interfaceIDs_backing=InterfaceIDs(nightscoutSystemId=null, nightscoutId=6251f4b9215b0200041e4d4c, pumpType=OMNIPOD_DASH, pumpSerial=4241, temporaryId=null, pumpId=-8976659072097442112, startId=null, endId=-7142919318749436032), timestamp=1649602949159, utcOffset=7200000, type=NORMAL, isAbsolute=true, rate=1.5, duration=1800000)

[Ede1975]

do you have logs prior this one?

No, unfortunately not. I am still on the same Version, did not happen again. In case it occurs again i will add new logs.

[emmatovar27]

Hi There! I have the same issue. This happens every time the TBR is set for 30 min. The rest of TBR looks to work fine. AndroidAPS_LOG_1650416536289log.zip

Let me know if there is any additional information is needed.

I'm using the dev branch from Philoul to test Autotune.

[vanelsberg]

Could not confirm but now seeing similar, quite a lot of this overlapping TBR's at certain times. Am running DASH while building from Philoul's Autotune PR (merged with latest dev).

Will switch back building form current dev to see if it reproduces with some helpfull logging @MilosKozak Any suggestions on how to "debug" this. Assuming at least logfiles - anything else to watch for? (Btw: Will add logs as soon as I can repro this on current dev branch)

[nicoleloop]

I am experiencing the same issues with my Dash Pods, as well as terrible bluetooth connections with the last 4 pods. In the range of the lowest, 51% to the highest being 63% connection quality.

[MilosKozak]

@vanelsberg yes, try dev and provide logs

[robertrub]

I started looking and found that I have the same bug

The logs

[vanelsberg]

@MilosKozak Omipod DASH, now running latest dev builds from yesterday/today. Also seeing overlapping TBR. (Note that I am running an outdated version of NS - do not think this is relevant?)

Have 2 sets of logging with additional screenshots Also adding .zip for the complete AAPS logging for today:

1.zip 2.zip morelog.zip

---

Latest dev:

Just one example:

[vanelsberg]

Did not come to checking logs myself, maybe tomorrow... @MilosKozak If you need additional info or specific testing, no problem, let me know...

[robertrub]

It is a killer... I'm not sure going on using dev

Of course, if AAPS was NOT acting on negative IOB, I wouldn't be in a crashing hypo at the moment... One more proof that acting on on neg IOB is not good... ;)

[robertrub]

I went back to Master 3.0.0.1 and overlapping TBRs stopped.

[vanelsberg]

Screenprint showing overlap + logging (lots - had logging of events enabled - switched this off now) 20220427.zip

[robertrub]

After 12h of use, I confirm that the TBR overlapping issue does not exist in master v3.0.0.1.

[mick711]

Last dev i [Uploading AndroidAPS.log…]()

[nicoleloop]

I was woken up 3 times last night by low bg events alarms. Does anyone have any idea what's happening? Can't we just revert the last month of dev updates and start over? I am worried that this is going to cause something worse than just having alarms go off, like not being any to wake up because AAPS hadn't recognized or recorded 118 +111 minutes of basal! My low event last night was so pronounced that I walked into the wall of my own bedroom and then crumpled on the floor. My bg registered 26mg/dL when I checked it with my meter. At this point I would support reverting all of the last month's worth of commits on the Dev branch, as this is a very dangerous bug!

Sent from my T-Mobile 5G Device

-------- Original message -------- From: mick711 @.> Date: 4/27/22 11:12 AM (GMT-08:00) To: nightscout/AndroidAPS @.> Cc: nicoleloop @.>, Comment @.> Subject: Re: [nightscout/AndroidAPS] Overlapping Times in basal Tab leading to wrong iob? Dev-i (Issue #1597)

[Screenshot_20220427-200812044]https://nam12.safelinks.protection.outlook.com/?url=https%3A%2F%2Fuser-images.githubusercontent.com%2F441734%2F165591901-2e3f638e-c5e5-4dfd-b8b0-5fb0d90467c2.jpg&data=05%7C01%7C%7C5024ee6af1194ee636a308da2879867b%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637866799664227550%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=RhK6M4cbf94%2BAg6vDRRDv1u%2BLvGmvHI9r1%2FXnv4QbsI%3D&reserved=0 [Screenshot_20220427-200823761]https://nam12.safelinks.protection.outlook.com/?url=https%3A%2F%2Fuser-images.githubusercontent.com%2F441734%2F165591913-1e9528f2-e7a2-4e71-b9be-9ac884803096.jpg&data=05%7C01%7C%7C5024ee6af1194ee636a308da2879867b%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637866799664227550%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=W65pNoxF%2F1b1w2hOgZR3elvP4uXFzwyLFBgsjHkZ4vk%3D&reserved=0 Last dev i Uploading AndroidAPS.log…

— Reply to this email directly, view it on GitHubhttps://nam12.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnightscout%2FAndroidAPS%2Fissues%2F1597%23issuecomment-1111327953&data=05%7C01%7C%7C5024ee6af1194ee636a308da2879867b%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637866799664227550%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=ecNolkgsWHfjR5Scmp9Vkw8nPrqfvVnBNYwVEEocg3I%3D&reserved=0, or unsubscribehttps://nam12.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAIU7LKXG74UTSL6S2DGITJLVHF7RXANCNFSM5TA755MA&data=05%7C01%7C%7C5024ee6af1194ee636a308da2879867b%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637866799664227550%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=LQoWbDsp3YLfF854EMZZFsmSrGpWKadZD6yAr9zn%2FKc%3D&reserved=0. You are receiving this because you commented.Message ID: @.***>

[nicoleloop]

Sent from my T-Mobile 5G Device

-------- Original message -------- From: nicoleloop @.> Date: 4/27/22 3:38 PM (GMT-08:00) To: nightscout/AndroidAPS @.> Cc: nicoleloop @.>, Your activity @.> Subject: Re: [nightscout/AndroidAPS] Overlapping Times in basal Tab leading to wrong iob? Dev-i (Issue #1597)

I was woken up 3 times last night by low bg events alarms. Does anyone have any idea what's happening? Can't we just revert the last month of dev updates and start over? I am worried that this is going to cause something worse than just having alarms go off, like not being any to wake up because AAPS hadn't recognized or recorded 118 +111 minutes of basal! My low event last night was so pronounced that I walked into the wall of my own bedroom and then crumpled on the floor. My bg registered 26mg/dL when I checked it with my meter. At this point I would support reverting all of the last month's worth of commits on the Dev branch, as this is a very dangerous bug!

Sent from my T-Mobile 5G Device

-------- Original message -------- From: mick711 @.> Date: 4/27/22 11:12 AM (GMT-08:00) To: nightscout/AndroidAPS @.> Cc: nicoleloop @.>, Comment @.> Subject: Re: [nightscout/AndroidAPS] Overlapping Times in basal Tab leading to wrong iob? Dev-i (Issue #1597)

[Screenshot_20220427-200812044]https://nam12.safelinks.protection.outlook.com/?url=https%3A%2F%2Fuser-images.githubusercontent.com%2F441734%2F165591901-2e3f638e-c5e5-4dfd-b8b0-5fb0d90467c2.jpg&data=05%7C01%7C%7C5024ee6af1194ee636a308da2879867b%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637866799664227550%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=RhK6M4cbf94%2BAg6vDRRDv1u%2BLvGmvHI9r1%2FXnv4QbsI%3D&reserved=0 [Screenshot_20220427-200823761]https://nam12.safelinks.protection.outlook.com/?url=https%3A%2F%2Fuser-images.githubusercontent.com%2F441734%2F165591913-1e9528f2-e7a2-4e71-b9be-9ac884803096.jpg&data=05%7C01%7C%7C5024ee6af1194ee636a308da2879867b%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637866799664227550%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=W65pNoxF%2F1b1w2hOgZR3elvP4uXFzwyLFBgsjHkZ4vk%3D&reserved=0 Last dev i Uploading AndroidAPS.log…

— Reply to this email directly, view it on GitHubhttps://nam12.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnightscout%2FAndroidAPS%2Fissues%2F1597%23issuecomment-1111327953&data=05%7C01%7C%7C5024ee6af1194ee636a308da2879867b%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637866799664227550%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=ecNolkgsWHfjR5Scmp9Vkw8nPrqfvVnBNYwVEEocg3I%3D&reserved=0, or unsubscribehttps://nam12.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAIU7LKXG74UTSL6S2DGITJLVHF7RXANCNFSM5TA755MA&data=05%7C01%7C%7C5024ee6af1194ee636a308da2879867b%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637866799664227550%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=LQoWbDsp3YLfF854EMZZFsmSrGpWKadZD6yAr9zn%2FKc%3D&reserved=0. You are receiving this because you commented.Message ID: @.***>

— Reply to this email directly, view it on GitHubhttps://nam12.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnightscout%2FAndroidAPS%2Fissues%2F1597%23issuecomment-1111547628&data=05%7C01%7C%7C9fe4b259f4424a7538e008da289eb091%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637866959274183606%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=JPvGszauIywrZgaOyAEV%2BxTECuaWsG%2BxqK157GVkPAw%3D&reserved=0, or unsubscribehttps://nam12.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAIU7LKRNIDZW77LU42SNT53VHG6XLANCNFSM5TA755MA&data=05%7C01%7C%7C9fe4b259f4424a7538e008da289eb091%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637866959274183606%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=icnr0rLLkBoYd3W03bMRGfsYad7l%2FwAS%2FcnGJzRXYsQ%3D&reserved=0. You are receiving this because you are subscribed to this thread.Message ID: @.***>

[jgslade]

I got an overlapping basal today. Looking into what was going on at the time, it looks like a tbr was set, an smb was delivered, an automation was triggered setting a temp target, then another tbr was set.

AndroidAPS._2022-04-2700-00-09.4.zip

[nicoleloop]

I have yet to look in the log files to verify this info, but right off the top of my head- temp basals are affected in this way, but not SMBs.

Sent from my T-Mobile 5G Device

-------- Original message -------- From: jgslade @.> Date: 4/27/22 9:33 PM (GMT-08:00) To: nightscout/AndroidAPS @.> Cc: nicoleloop @.>, Comment @.> Subject: Re: [nightscout/AndroidAPS] Overlapping Times in basal Tab leading to wrong iob? Dev-i (Issue #1597)

I got an overlapping basal today. Looking into what was going on at the time, it looks like a tbr was set, an smb was delivered, an automation was triggered setting a temp target, then another tbr was set.

[Screenshot_20220427-215332_AndroidAPS]https://nam12.safelinks.protection.outlook.com/?url=https%3A%2F%2Fuser-images.githubusercontent.com%2F29897815%2F165676730-be897c6f-7096-4623-b494-c727b1ff8a93.jpg&data=05%7C01%7C%7Ccb0a6113f4fd4702c49a08da28d03de8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637867172105993931%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=QS4fY%2BOufpBCTIIuba1Y9cKJ21Cmbm2Ya7sTwYUNEmQ%3D&reserved=0

AndroidAPS.2022-04-27_00-00-09.4.ziphttps://nam12.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnightscout%2FAndroidAPS%2Ffiles%2F8579512%2FAndroidAPS._2022-04-27_00-00-09_.4.zip&data=05%7C01%7C%7Ccb0a6113f4fd4702c49a08da28d03de8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637867172105993931%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=Sme4FGHOhOWUJmoUZFOLA9u%2FOl%2B41S963UlEcuG1PnA%3D&reserved=0

— Reply to this email directly, view it on GitHubhttps://nam12.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnightscout%2FAndroidAPS%2Fissues%2F1597%23issuecomment-1111735717&data=05%7C01%7C%7Ccb0a6113f4fd4702c49a08da28d03de8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637867172105993931%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=OMkeb9jegJYCJk2RoMFqNiXOGrRDckWt8mtC3wMIhKs%3D&reserved=0, or unsubscribehttps://nam12.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAIU7LKQK7PVB6AMMCWHIJP3VHIIJPANCNFSM5TA755MA&data=05%7C01%7C%7Ccb0a6113f4fd4702c49a08da28d03de8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637867172105993931%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=h03VKWXU5nhxAolnSBRZv7GEb3TbGLhZKPMnxLkD%2Bt0%3D&reserved=0. You are receiving this because you commented.Message ID: @.***>

[tswire]

Has anyone confirmed if this truly results in incorrect IOB? More importantly, if that IOB is used in the predictions and AAPS dosing decisions? I think it's very obvious that something is "wrong" based on the treatments tab for TB. However, I think we need to know immediately and urgently if this could be dangerous, as seen by a few examples of people going low.

[vanelsberg]

I was woken up 3 times last night by low bg events alarms.

Same here a number of times. As far as I could tell the much-argued negative IOB problem: too much insulin causing high negative insulin buildup that AAPS will try to fill in as soon as BGs are rising - followed by another low as logic dictates. This could be a side-effect? (Suggestion: to prevent -IOB consider lowering basal rate (and maybe ISF) to prevent 0-TBR occurring for long durations?)

Not sure, but I do think we need confirmation that the TBR problems we see are not affecting IOB calculations. I could switch back to release version, but then I would not be able to help document/analyze/solve this problem. So will stay on dev as long as I think is safe enough.

[Ede1975]

Has anyone confirmed if this truly results in incorrect IOB? More importantly, if that IOB is used in the predictions and AAPS dosing decisions? I think it's very obvious that something is "wrong" based on the treatments tab for TB. However, I think we need to know immediately and urgently if this could be dangerous, as seen by a few examples of people going low.

From my point of view its obvious that it affects iob calculations. It counts (negative) insulin twice when Times overlap. If you remove an overlapping treatment (which is Not 0 yet) , the calculated iob changes instantly. Same for positive overlaps of course.

[vanelsberg]

This issue is labeled "DASH". Is it only for Omnipod DASH pump the overlapping TBR is occurring?

[MilosKozak]

i improved logging https://github.com/nightscout/AndroidAPS/commit/2d653c67ef9c32201fe0e28897af4904f9c423cb i'd need someone to reset AAPS DB (with zero IOB COB to be safe) and try latest dev Provide logs if you see this again

[vanelsberg]

i improved logging 2d653c6 i'd need someone to reset AAPS DB (with zero IOB COB to be safe) and try latest dev Provide logs if you see this again

@MilosKozak About to build, reset database and then watch for repro. Expect logging as soon as I have some.

[vanelsberg]

....Provide logs if you see this again

Any special log settings to enable?

[MilosKozak]

PUMP*

[mick711]

I try to find différence Master and Dev

[vanelsberg]

PUMP*

@MilosKozak Yesterday evening install latest dev, did database reset: until now (24hrs) no overlap. (detail: pump connection is 100%, so no fail/retries on commands - will test later if any relation with failures?)

(Expect logging as soon as I see overlapping occur)

[emmatovar27]

Same Here! :)

TBR is not overlapping anymore!

Thanks @MilosKozak

[nicoleloop]

Wonderful! Thanks Milos! 😁

Sent from my T-Mobile 5G Device

-------- Original message -------- From: emmatovar27 @.> Date: 4/30/22 1:56 PM (GMT-08:00) To: nightscout/AndroidAPS @.> Cc: nicoleloop @.>, Comment @.> Subject: Re: [nightscout/AndroidAPS] Overlapping Times in basal Tab leading to wrong iob? Dev-i (Issue #1597)

Same Here! :)

TBR is not overlapping anymore!

Thanks @MilosKozakhttps://nam12.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FMilosKozak&data=05%7C01%7C%7Cc0ccc8eeb36340b31c7108da2aebdbf3%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637869489736324622%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=LlkDL4ewME1wLOqNA2eXlT1x49GyJxP91xu%2Fmjvq%2B8o%3D&reserved=0

[TempBasalRates]https://nam12.safelinks.protection.outlook.com/?url=https%3A%2F%2Fuser-images.githubusercontent.com%2F39507448%2F166122364-6af6dc19-b192-437b-a854-db1e265a085e.png&data=05%7C01%7C%7Cc0ccc8eeb36340b31c7108da2aebdbf3%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637869489736324622%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=CQ%2F0wvcI6JSLnEp2WWSN17rr6Wk2YthdmmIs40a3YVI%3D&reserved=0

— Reply to this email directly, view it on GitHubhttps://nam12.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnightscout%2FAndroidAPS%2Fissues%2F1597%23issuecomment-1114052771&data=05%7C01%7C%7Cc0ccc8eeb36340b31c7108da2aebdbf3%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637869489736324622%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=bUWPFJPEYMVjKaWsjG247MhfXLbraeRyZR45JAdD%2FIs%3D&reserved=0, or unsubscribehttps://nam12.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAIU7LKVSDD6C3FNVKN46IMTVHWM6ZANCNFSM5TA755MA&data=05%7C01%7C%7Cc0ccc8eeb36340b31c7108da2aebdbf3%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637869489736324622%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=mcquxkPMdwrNr764%2BKkJACg6dBsHuUrscGkizLnAWXM%3D&reserved=0. You are receiving this because you commented.Message ID: @.***>

[vanelsberg]

After 2 days, with new build and database reset, still not seeing anything on TBR overlapping. Wondering if database could be a factor in causing this?

[ArthurusDent]

@vanelsberg If this is the same bug I saw, it might only happen when disabling the pump für 2 h. Didn't notice it with shorter periods but I might just not have noticed. Could you maybe test 2 h, too?: Disabling the pump for 2 h, waiting for the algorithm to run once and then re-enabling the pump again?

[MilosKozak]

i believe it was corrupted data

[vanelsberg]

.. i believe it was corrupted data

@MilosKozak Was thinking the same. Had been testing autotune on separate branch. Now on dev latest dev branch with databases reset, but still: Overlap again past night (see 20220503, 02:59h, 04:59h,) Especially the 04:59h event was troublesome is it caused a lot of incorrect negative IOB buildup :-|

Coincided with two G6 abnormalities where it started alerting low (while relatively on target) needing carbs (partly unregistered). Not sure if this could be related. See screen prints G6 and FSL I have additionally active for testing/comparence

All logging/info: 20220503.zip

Screenshots;

[MilosKozak]

if I see correctly the fixing commit is not included in autotune branch https://github.com/nightscout/AndroidAPS/commit/1684b88270b34d0205f3e10ad3109e2319802532

[tswire]

if I see correctly the fixing commit is not included in autotune branch 1684b88

But he said this overrlapping was from updated Dev (as of yesterday) and after resetting DB? So that commit was 20 days ago and not sure how that fits in to this?

[jgslade]

I got some overlapping tbrs last night. Using the issue with improved logging and reset the db after updating.

The 10:34-11:04 and 10:59-11:09 one is interesting because it's the same tbr value.

AndroidAPS._2022-05-0221-47-20.1.zip

[vanelsberg]

if I see correctly the fixing commit is not included in autotune branch

(@MilosKozak FYI: Not using autotune branch but latest "dev" #7d93a2ceef)

More overlap? Going to reset database again to see if it stops,... May have to change back to release 3.0.0.1 because this getting problematic :-(

More logging & screen prints See: 07:23h 08:18h 09:13h 10:43h 15:14h 16:28h (??) .... 20220503-2.zip

About

[kristinsmotz]

yesterday found the first overlapping TBR's I have noticed. We're on Dash, and testing Dynamic ISF on the second latest version (Dynamic ISF beta 1.6.1). What was interesting to me was that it occurred in the same timeframe where I encountered the first re-occurance of my other issue (https://github.com/nightscout/AndroidAPS/issues/1667 where bolus doses (and SMB's) occasionally don't show up in the NSClient phone, only in AAPS phone and on our nightscout site), since I cleared out memory and cache related to NSClient on my phone a few days ago. Could these issues be related in any way? Since clearing database (or in my case clearing phone memory) appears to help for a period of time, could it have anything to database structure/save data file size?

[avereha]

Can you please to disable Nightscout sync and check if the issue is still there?

[vanelsberg]

Can you please to disable Nightscout sync and check if the issue is still there?

Would setting NSClient to "Paused" (checkbox) do?

[avereha]

yes

[vanelsberg]

(Quick first results. Will run this for some time trying to "leave it alone" and post new logs here once overlapping happens)

Did update to new build for #1691 around 21:45h (?). Now seeing this. Not sure: why are these most recent TBRs suddenly with only a starting time?

See "22:11-22:41 1.75U/h 30 mins": It was canceled (also on DASH history) but somehow this was not registered to AAPS?

Log files + screenshot 20220504.zip

[jgslade]

I turned off nightscout syncing this morning, checked this afternoon and I've still got overlapping basals. AndroidAPS._2022-05-0400-00-00.2.zip AndroidAPS.log

[vanelsberg]

@avereha Lots of TBR overlap (see 00:00 ... 03:00h)? With #1691, NSClient paused.

Logfiles & screenprint 20220505-1.zip

[MilosKozak]

looks like AAPS is updating old records. i added more logging to dev https://github.com/nightscout/AndroidAPS/commit/d131ceb4bd52e85d92650f6dff26692f73d822de please again update and reset db + pause NSClient