OSCAR rebuilding and oximeter problem
Re: OSCAR rebuilding and oximeter problem
I zipped my OSCAR_Data folder, but can't attach it - too large at 54MB. What can I leave out to make a smaller file?
If you meant a directory listing, I dumped it into a text file and attached it. When I ran Data | Rebuild, it went through several phases of what I'm used to seeing (but it flickers past too fast to track, plus I wasn't paying attention), then displayed the close/debug popup.
119GB free space on C:
If you meant a directory listing, I dumped it into a text file and attached it. When I ran Data | Rebuild, it went through several phases of what I'm used to seeing (but it flickers past too fast to track, plus I wasn't paying attention), then displayed the close/debug popup.
119GB free space on C:
_________________
Mask: ResMed AirFit™ F30 Full Face CPAP Mask with Headgear |
Additional Comments: Alg:AutoSet,Pressure:7.0-11.0,Ramp:Auto/4.0,EPR:1/FullTime,Climate:Auto |
- BlueDragon
- Posts: 546
- Joined: Sun May 27, 2018 10:20 am
- Location: Sonoran Desert
OSCAR rebuilding and oximeter problem
That is what I wanted to see--a directory listing. Make a backup copy of the entire OSCAR_Data folder in case the repair attempts fail.Dave Wise wrote: ↑Sat Jun 08, 2019 5:33 pmI zipped my OSCAR_Data folder, but can't attach it - too large at 54MB. What can I leave out to make a smaller file?
If you meant a directory listing, I dumped it into a text file and attached it. Wise_Oscar_Dir.zip
When I ran Data | Rebuild, it went through several phases of what I'm used to seeing (but it flickers past too fast to track, plus I wasn't paying attention), then displayed the close/debug popup.
119GB free space on C:
It is apparent that you willl need to rebuild your profile but, fortunately, the backup data appears still to be there. Unfortunately, I have no experience with how to handle a profile with oximetry data.
Here is what you can do for the CPAP data: Create a new profile. Then use the SD Card Import function and select a folder as your input instead of an actual SD card. Select this folder: C:\Users\David_Wise\Documents\OSCAR_Data\Profiles\David Wise\ResMed_23183180396\Backup. That should import all your data from the old profile into the new. If that fails, then we have to rethink this.
I hope one of the other SH/OSCAR experts can speak to the issue of recovering your oximetry data to the profile, or merging the new profile with the old. I suspect this is possible but have never done this myself.
_________________
Machine: AirCurve™ 10 VAuto BiLevel Machine with HumidAir™ Heated Humidifier |
Mask: Brevida™ Nasal Pillow CPAP Mask with Headgear |
Additional Comments: Dx Mar 2018 (AHI=24, RDI=54; AHI=73 supine). Started APAP June 2018, VAuto Aug 2020. |
See OSCAR for the latest release.
OSCAR Team
ResMed AirCurve 10 VAuto, F&P Brevida.
FlashAir SD and FlashPap for data transfer.
OSCAR Team
ResMed AirCurve 10 VAuto, F&P Brevida.
FlashAir SD and FlashPap for data transfer.
OSCAR rebuilding and oximeter problem
Thank you, BD, that got me back my CPAP data. I will next try to import oximeter data from .spo2 files.
_________________
Mask: ResMed AirFit™ F30 Full Face CPAP Mask with Headgear |
Additional Comments: Alg:AutoSet,Pressure:7.0-11.0,Ramp:Auto/4.0,EPR:1/FullTime,Climate:Auto |
OSCAR rebuilding and oximeter problem
First few .sp02 file imports went okay. I wish there was a way to import a group of files, once the timezone glitch is fixed.
_________________
Mask: ResMed AirFit™ F30 Full Face CPAP Mask with Headgear |
Additional Comments: Alg:AutoSet,Pressure:7.0-11.0,Ramp:Auto/4.0,EPR:1/FullTime,Climate:Auto |
Re: OSCAR rebuilding and oximeter problem
Crashed it again. I imported a CPAP session, then a bunch of .sp02 files, then commanded Data |Rebuild .
Unhandled exception at 0x00007FF8E70EAA2C (ntdll.dll) in OSCAR.exe: 0xC0000374: A heap has been corrupted (parameters: 0x00007FF8E7143180).
I hit Go in the debugger and against expectation, OSCAR resumed execution, with errors. This allowed me to capture the debug pane, which I've zipped and attached.
OS is Windows 10/64 Enterprise, running on a Thinkpad T430 with 12GB ram.
Unhandled exception at 0x00007FF8E70EAA2C (ntdll.dll) in OSCAR.exe: 0xC0000374: A heap has been corrupted (parameters: 0x00007FF8E7143180).
I hit Go in the debugger and against expectation, OSCAR resumed execution, with errors. This allowed me to capture the debug pane, which I've zipped and attached.
OS is Windows 10/64 Enterprise, running on a Thinkpad T430 with 12GB ram.
_________________
Mask: ResMed AirFit™ F30 Full Face CPAP Mask with Headgear |
Additional Comments: Alg:AutoSet,Pressure:7.0-11.0,Ramp:Auto/4.0,EPR:1/FullTime,Climate:Auto |
- Attachments
-
- Wise_Oscar_Crash_2_Debug_Pane.zip
- (4.58 KiB) Downloaded 8 times
- BlueDragon
- Posts: 546
- Joined: Sun May 27, 2018 10:20 am
- Location: Sonoran Desert
Re: OSCAR rebuilding and oximeter problem
I passed this by Phil, who has been working on oximeter import issues. He says that looking at the debug output, it looks like the problem is with the purge, which had a delete error for some reason, and then I'm guessing OSCAR got pretty confused about his ResMed sessionss - as if some of the data was left behind. I don't think it's related to the oximeter imports - but I could be wrong.Dave Wise wrote: ↑Tue Jun 18, 2019 7:33 amCrashed it again. I imported a CPAP session, then a bunch of .sp02 files, then commanded Data |Rebuild .
Unhandled exception at 0x00007FF8E70EAA2C (ntdll.dll) in OSCAR.exe: 0xC0000374: A heap has been corrupted (parameters: 0x00007FF8E7143180).
I hit Go in the debugger and against expectation, OSCAR resumed execution, with errors. This allowed me to capture the debug pane, which I've zipped and attached.
OS is Windows 10/64 Enterprise, running on a Thinkpad T430 with 12GB ram.
I (BD) will test data rebuilding to see if I can shake a problem loose. I may ask you to provide a copy of your database if you are willing.
_________________
Machine: AirCurve™ 10 VAuto BiLevel Machine with HumidAir™ Heated Humidifier |
Mask: Brevida™ Nasal Pillow CPAP Mask with Headgear |
Additional Comments: Dx Mar 2018 (AHI=24, RDI=54; AHI=73 supine). Started APAP June 2018, VAuto Aug 2020. |
See OSCAR for the latest release.
OSCAR Team
ResMed AirCurve 10 VAuto, F&P Brevida.
FlashAir SD and FlashPap for data transfer.
OSCAR Team
ResMed AirCurve 10 VAuto, F&P Brevida.
FlashAir SD and FlashPap for data transfer.
Re: OSCAR rebuilding and oximeter problem
I'm perfectly willing, BD, just let me know how to send it to you. It's 50+MB after compression.
_________________
Mask: ResMed AirFit™ F30 Full Face CPAP Mask with Headgear |
Additional Comments: Alg:AutoSet,Pressure:7.0-11.0,Ramp:Auto/4.0,EPR:1/FullTime,Climate:Auto |
- BlueDragon
- Posts: 546
- Joined: Sun May 27, 2018 10:20 am
- Location: Sonoran Desert
Re: OSCAR rebuilding and oximeter problem
I have sent you a PM about where to upload your files.
_________________
Machine: AirCurve™ 10 VAuto BiLevel Machine with HumidAir™ Heated Humidifier |
Mask: Brevida™ Nasal Pillow CPAP Mask with Headgear |
Additional Comments: Dx Mar 2018 (AHI=24, RDI=54; AHI=73 supine). Started APAP June 2018, VAuto Aug 2020. |
See OSCAR for the latest release.
OSCAR Team
ResMed AirCurve 10 VAuto, F&P Brevida.
FlashAir SD and FlashPap for data transfer.
OSCAR Team
ResMed AirCurve 10 VAuto, F&P Brevida.
FlashAir SD and FlashPap for data transfer.
- BlueDragon
- Posts: 546
- Joined: Sun May 27, 2018 10:20 am
- Location: Sonoran Desert
Re: OSCAR rebuilding and oximeter problem
I've looked at the files you sent me, and your OSCAR_Data file is apparently from after the rebuild failure, and doesn't give me enough clues about why the rebuild failure occurred. The STR.EDF file is truncated after the fourth day, so settings data from all days after that is missing and the Daily window says that the settings are "guessed."
Could you make a zip of your SD card and upload that? I'll send you a PM with the file request address.
Thanks.
Could you make a zip of your SD card and upload that? I'll send you a PM with the file request address.
Thanks.
_________________
Machine: AirCurve™ 10 VAuto BiLevel Machine with HumidAir™ Heated Humidifier |
Mask: Brevida™ Nasal Pillow CPAP Mask with Headgear |
Additional Comments: Dx Mar 2018 (AHI=24, RDI=54; AHI=73 supine). Started APAP June 2018, VAuto Aug 2020. |
See OSCAR for the latest release.
OSCAR Team
ResMed AirCurve 10 VAuto, F&P Brevida.
FlashAir SD and FlashPap for data transfer.
OSCAR Team
ResMed AirCurve 10 VAuto, F&P Brevida.
FlashAir SD and FlashPap for data transfer.
Re: OSCAR rebuilding and oximeter problem
Uploaded. Good luck!
_________________
Mask: ResMed AirFit™ F30 Full Face CPAP Mask with Headgear |
Additional Comments: Alg:AutoSet,Pressure:7.0-11.0,Ramp:Auto/4.0,EPR:1/FullTime,Climate:Auto |
- BlueDragon
- Posts: 546
- Joined: Sun May 27, 2018 10:20 am
- Location: Sonoran Desert
Re: OSCAR rebuilding and oximeter problem
Can you walk me through the process by which you import oximetry data? How do you avoid importing the same day twice (and does it matter)? How many days do you import before you rebuild your CPAP data, how do you decide how many to import, and why do you rebuild the CPAP data? Do you adjust the oximeter date or time? (I don't use an oximeter myself, so maybe I would know the answers already if I did.)
And, I don't remember whether your migrated SH data before starting OSCAR or whether you started fresh with OSCAR. If you migrated data, I may need your SH folder as well (upload to the same place).
In my testing so far, nothing has gone wrong and all the data seems ok, data rebuilds are working, etc. So I want to make sure I am going through the same steps you went through.
Thanks!
And, I don't remember whether your migrated SH data before starting OSCAR or whether you started fresh with OSCAR. If you migrated data, I may need your SH folder as well (upload to the same place).
In my testing so far, nothing has gone wrong and all the data seems ok, data rebuilds are working, etc. So I want to make sure I am going through the same steps you went through.
Thanks!
_________________
Machine: AirCurve™ 10 VAuto BiLevel Machine with HumidAir™ Heated Humidifier |
Mask: Brevida™ Nasal Pillow CPAP Mask with Headgear |
Additional Comments: Dx Mar 2018 (AHI=24, RDI=54; AHI=73 supine). Started APAP June 2018, VAuto Aug 2020. |
See OSCAR for the latest release.
OSCAR Team
ResMed AirCurve 10 VAuto, F&P Brevida.
FlashAir SD and FlashPap for data transfer.
OSCAR Team
ResMed AirCurve 10 VAuto, F&P Brevida.
FlashAir SD and FlashPap for data transfer.
Re: OSCAR rebuilding and oximeter problem
It's 100% manual, including avoiding duplicate imports.
I rebuilt because (a) one time I didn't see the imported oximeter data in Daily View until I rebuilt, and (b) I had the impression I was supposed to. Not the first time I've been wrong!
I adjust the date and time to match CPAP. Due to the GMT problem, I do it by hand. I tick the CPAP box, note the time, then tick the Oximeter box and manually enter the corrected local time.
I was using SH before switching to OSCAR, but I can't remember whether I imported the SH data or just started fresh with my CPAP SD card.
I don't have a set number of days to import Oxi data; I don't run the oximeter every day, it's just a plaything. Last time, I imported one old day and it looked okay, did another and it looked okay, then imported three or four days and rebuilt, and that was when it crashed. Note the crash was a heap corruption, which means a rogue write to memory; that is chaotically sensitive to memory size, OS patch level and run history, phase of the moon, and everything else you can think of... Rather than beat up your head trying to reproduce this one, maybe it would be better to just wait until I crash again, then see if there's a family resemblance. I will leave the debug pane enabled, and try to capture it when O falls over again.
I can't remember, do you build the Windows version of OSCAR with Visual Studio? Maybe you could send me a symbols file. Is there a "verbose malloc debug" build option?
I rebuilt because (a) one time I didn't see the imported oximeter data in Daily View until I rebuilt, and (b) I had the impression I was supposed to. Not the first time I've been wrong!
I adjust the date and time to match CPAP. Due to the GMT problem, I do it by hand. I tick the CPAP box, note the time, then tick the Oximeter box and manually enter the corrected local time.
I was using SH before switching to OSCAR, but I can't remember whether I imported the SH data or just started fresh with my CPAP SD card.
I don't have a set number of days to import Oxi data; I don't run the oximeter every day, it's just a plaything. Last time, I imported one old day and it looked okay, did another and it looked okay, then imported three or four days and rebuilt, and that was when it crashed. Note the crash was a heap corruption, which means a rogue write to memory; that is chaotically sensitive to memory size, OS patch level and run history, phase of the moon, and everything else you can think of... Rather than beat up your head trying to reproduce this one, maybe it would be better to just wait until I crash again, then see if there's a family resemblance. I will leave the debug pane enabled, and try to capture it when O falls over again.
I can't remember, do you build the Windows version of OSCAR with Visual Studio? Maybe you could send me a symbols file. Is there a "verbose malloc debug" build option?
_________________
Mask: ResMed AirFit™ F30 Full Face CPAP Mask with Headgear |
Additional Comments: Alg:AutoSet,Pressure:7.0-11.0,Ramp:Auto/4.0,EPR:1/FullTime,Climate:Auto |