These are chat archives for nightscout/intend-to-bolus

24th
May 2016
Andrew Baugh
@baughaw
May 24 2016 00:06
@scottleibrand Thanks Scott, I missed your comments. That is exactly what I was looking for! I am fed up with the Carelink stick. :)
mm22dl
@mm22dl
May 24 2016 00:31
@jaylagorio : has it been difficult for many people to get an OpenAPS compatible pump?
Dana Lewis
@danamlewis
May 24 2016 00:32
@mm22dl yes. There's a lot more who would loop if they could find compatible pumps.
mm22dl
@mm22dl
May 24 2016 00:56
:-(
jaylagorio
@jaylagorio
May 24 2016 01:03
@mm22dl Had to step away, but yeah, I thought this would be more straightforward than it is but @danamlewis 's right.
Paul Martin
@Spazholio
May 24 2016 01:08
@danamlewis So far, doing a temp target for 2 hours during dinner, set to a target of 80 seems to be doing the trick. I ate a good amount of protein, and haven’t really gone over 123. So for a sample size of 1 meal and 1 person, this experiment has been an unqualified success. =)
Dana Lewis
@danamlewis
May 24 2016 01:09
:)
Jason Curry
@mccgm
May 24 2016 01:57
On NS, I'm noticing the time on boluses is an hour off. I've verified my pump time is correct, combed through all my settings and ini files, but I have no clue what's causing this. Any ideas?
Dana Lewis
@danamlewis
May 24 2016 02:09
What time does your Pi/rig think it is?
Paul Martin
@Spazholio
May 24 2016 02:15

Is this still the “best” way to set up the pumphistory-zoned.json file? I’ve got something fairly different in my openaps.ini:

openaps report add monitor/pumphistory-zoned.json JSON tz rezone monitor/pumphistory.json

Mine is: openaps use tz rezone --timezone "EDT" --adjust "missing" --date "timestamp dateString start_at end_at created_at" --astimezone monitor/pumphistory.json
JaysonEwer
@JaysonEwer
May 24 2016 02:18
You
JaysonEwer
@JaysonEwer
May 24 2016 02:24
Another form of y'uns is yins. https://en.m.wikipedia.org/wiki/Yinz I have family that lives near Pittsburgh,PA, which is pronounced Picsburgh by locals. 😜
Dana Lewis
@danamlewis
May 24 2016 02:24
;p
Jason Curry
@mccgm
May 24 2016 02:41
@danamlewis Pi knows what time it is, everything is right on except those dots. The time in my pumphistoryzoned is correct, but on NS is an hour plus. :confused:
Paul Martin
@Spazholio
May 24 2016 02:47
@bewest My current monitor/pumphistory.json file is generated via openaps use pump iter_pump_hours 3. Now, I know that you’re probably twitching pretty hard right now, so if you could give me the way you’d prefer to do it via oref0 I’d be happy to use that instead. =)
Dana Lewis
@danamlewis
May 24 2016 02:55
@mccgm there was some other people time zone trouble shooting a few hours ago, any of that stuff working?
ryanchen26
@ryanchen26
May 24 2016 03:27
@danamlewis No. It looks like someone sold me a defective 722 pump. The A33 error is bad news. Now I'm concerned this error is common and the 2nd-hand market is full of these duds. I'll try again at some point.
Dana Lewis
@danamlewis
May 24 2016 03:38
ok. someone ( @jaylagorio) is looking for old dud pumps to see if he can learn something from the firmware, if you end up with not being able to send it back to the seller for a refund.
Nathan Morse
@morsen
May 24 2016 03:43
btw, i keep wanting to read oref0 as oreo :)
and i was just able to take advantage of clockset.sh in oref0 git, so thank you everyone :) :)
Dana Lewis
@danamlewis
May 24 2016 03:43
@morsen have you seen there's a Seattle restaurant called orfeo? Confusing :)
Nathan Morse
@morsen
May 24 2016 03:44
haha, not yet!
and this means i just got my first suggested.json, so am very excited
ddaniels1
@ddaniels1
May 24 2016 03:45
:) :sparkles:
Dana Lewis
@danamlewis
May 24 2016 03:45
:+1:
Nathan Morse
@morsen
May 24 2016 03:45
that restaurant has a really good rating
ddaniels1
@ddaniels1
May 24 2016 03:51
@danamlewis is it temp-basal.json on pumphistory.json that drives the temp basal data displayed on NS, any idea?
Saskia
@Missdazzle
May 24 2016 04:19
Good morning America and hello world! Mrs. Robot woke up and :pig: :pig: :pig: :dancer: :dancer: :muscle: :muscle: :tada:
Mein erste sichtbare Loop-Nacht.PNG
David Cintron
@loudestnoise
May 24 2016 04:20
:+1: :+1: :+1: :+1: :+1: :+1:
Nathan Morse
@morsen
May 24 2016 04:24
first enact :) :)
after pizza, no less
Saskia
@Missdazzle
May 24 2016 04:28
:smile: :smile:
Jason Curry
@mccgm
May 24 2016 04:35
@morsen :100: :clap:
Matt Pressnall
@logichammer
May 24 2016 04:43

@danamlewis / @scottleibrand -- confirming that we want three separate pages for Autosens, MA, and AMA.

Realized that I combined Autosens and AMA into one page because the branch you are on for oref0 matters at the Autosens level. I can figure out how to word it, though, so you pick the right branch. I see that the documentation for Autosens and MA just got updated so I will use that. I'm actually going to work on this on the flight out on Wednesday. And should have something to merge up that night. Let me know if you need anything else from me and thanks!

Matthias
@ceben80
May 24 2016 04:44
Yeah @Missdazzle, congrats :star2:
Matt Pressnall
@logichammer
May 24 2016 04:46
@Missdazzle, that battery at 100% in the upper left? What value application setting is that to turn on? And what does it represent? Phone battery?
And congrats!
Frank Best
@efbest
May 24 2016 05:26
@Missdazzle congrats! :clap: :smile:
Boris and Kayley Raskin
@boryanvbu
May 24 2016 05:28
Congrats @Missdazzle !
Stephan
@MosiGitHub
May 24 2016 05:44
@Missdazzle very nice ... congrats :clap:
I woke up and one of my rigs (that should have been responsible during nights) stopped working tonight. Before I try a restart, I wanted to ask you wether anybody has an idea what the reason for the problem is:
openaps loop
mmtune: Traceback (most recent call last):
File "/usr/local/bin/openaps-use", line 8, in <module>
execfile(file)
File "/home/edison/src/openaps/bin/openaps-use", line 63, in <module>
app( )
File "/home/edison/src/openaps/openaps/cli/init.py", line 44, in call
self.run(self.args)
File "/home/edison/src/openaps/bin/openaps-use", line 57, in run
output = app(args, self)
File "/home/edison/src/openaps/openaps/uses/init.py", line 92, in call
return self.method.selected(args)(args, app)
File "/home/edison/src/openaps/openaps/uses/init.py", line 31, in call
return self.method(args, app)
File "/home/edison/src/openaps/openaps/uses/use.py", line 44, in call
self.before_main(args, app)
File "/home/edison/src/openaps/openaps/vendors/medtronic.py", line 61, in befo re_main
self.setup_medtronic( )
File "/home/edison/mmeowlink-source/mmeowlink/vendors/mmeowlink.py", line 84, in setup_medtronic
setup_medtronic_link(self)
File "/home/edison/mmeowlink-source/mmeowlink/vendors/mmeowlink.py", line 65, in setup_medtronic_link
link = builder.build(radio_type, port)
File "/home/edison/mmeowlink-source/mmeowlink/link_builder.py", line 16, in bu ild
return SubgRfspyLink(port)
File "/home/edison/mmeowlink-source/mmeowlink/vendors/subgrfspylink.py", lin e 56, in __init
self.open()
File "/home/edison/mmeowlink-source/mmeowlink/vendors/serial_interface.py", li ne 27, in open
self.check_setup()
File "/home/edison/mmeowlink-source/mmeowlink/vendors/subg_rfspy_link.py", lin e 72, in check_setup
self.serial_rf_spy.sync()
File "/home/edison/mmeowlink-source/mmeowlink/vendors/serial_rf_spy.py", line 108, in sync
raise CommsException("Could not get subg_rfspy state or version. Have you go t the right port/device and radio_type?")
mmeowlink.exceptions.CommsException: Could not get subg_rfspy state or version. Have you got the right port/device and radio_type?
No CGM data.
Stephan
@MosiGitHub
May 24 2016 05:54
Another thought concerning the different braches: I have one rig with AMA running but we use this rig only when I am at home and having an eye on the kid. In the other rig the "old" dev branch (MA and auto.sens) was running, but every time I go on a business trip I remove the auto-sens feature to make this rig running safer from my point of view. Why, because I got the impression that a lot of stuff could wrongly influence the auto-sens. That could be wrong, but I expect that if the family gives corrections as "BE" or the system don't knows when the kid needed some extra sugar to prevent him from getting low, the results could be wrong. That is the reason why I even thinking of removing MA from this rig and use the loop in its basic function - as a keep it simple loop. So I really like to have the possibilty to use different branches for different purposes. And I like AMA with its temp targets as well, but you have to take care that the system has all the necessary informations.
Scott Leibrand
@scottleibrand
May 24 2016 06:10
@MosiGitHub we definitely agree that there will be some people or situations for which AMA, autosens, etc. will be appropriate, and others for which it isn't. our thought has been that anyone should be able to run the latest code, and either enable or disable such features through configuration. so you could have a different openaps.ini for different situations, one with the data provided to oref0 to enable those features, and one without.
@logichammer I'm no longer sure we need three pages long-term, but we will until AMA is merged to dev (and maybe master). perhaps we should focus on the minimal set of changes required to the existing meal-assist docs to add the two new arguments, plus the other parts of your new docs that outline how to upload and view all the new AMA data, etc. if you wanted to make that an update to the current autosens and meal assist docs, with your changes and additions living in the new ama-docs branch and getting merged to dev when we merge oref0 ama to dev, that might be the most straightforward approach. if so, you're going to need to rebase your changes (figuratively if not literally) on top of the latest dev version of the autosens and meal assist page(s). I'll leave it up to you whether you want to do the split of that current doc into two as you make your edits, or leave that as something separate.
Saskia
@Missdazzle
May 24 2016 06:28
Thank you for your congratulations! Still some work to do with getting rid of the error messages... :smile:
@logichammer Yes, that´s the battery of my phone. I don´t have a Dexcom receiver but a xDrip box which sends the glucose data to my phone.
mm22dl
@mm22dl
May 24 2016 06:41
Hi people! Question of the day: how can one supervise OpenAPS while being offline? Do you only use Nightscout when online? Is there any other option? And how do you guys try and make sure you're most of the time online? Anyone built an ad-hoc network on their Pi?
Stephan
@MosiGitHub
May 24 2016 06:47
@mm22dl I have bought an Huawei 8377 for being online and if we loop without internet connection you can at least check your system via terminal (JuiceSSH)
@scottleibrand I have three different openaps.ini files on that rig :-D
but three problem is, that it is tough to keep all three up to date ...
Stephan
@MosiGitHub
May 24 2016 07:25
By the way: because of this part of the error message: "Could not get subg_rfspy state or version. Have you go t the right port/device and radio_type?")
mmeowlink.exceptions.CommsException: Could not get subg_rfspy state or version. Have you got the right port/device and radio_type? .... I decided to flash the TI stick again. Now it is communicating with the pump again...
jaylagorio
@jaylagorio
May 24 2016 10:28
@ryanchen26 I'm really sorry to hear the pump you bought went bad, but if you'd like to donate it to science (https://www.facebook.com/groups/cgminthecloud/permalink/945749455546311/) it would be a huge help. I don't know what that error is and I hope you're able to recover, but if not PM me for details.
C-Ville
@C-Ville
May 24 2016 10:46
@mm22dl I can see your problem and while I agree with @MosiGitHub that you can check the terminal output, I would prefer a graph solution. Unfortunately I failed to have NS run on my pi, which would be my first solution right now. At work I do not always have cell phone reception, and I definetly like the information of the NS website.
Nathan Morse
@morsen
May 24 2016 10:49
All, I’ve sporadically gotten that error: 'Pump' object has no attribute 'model'
Chris Oattes
@cjo20
May 24 2016 10:50
danamlewis: are the plans still to go to niche on Thursday? Or are there too many people for that?
Nathan Morse
@morsen
May 24 2016 10:50
i’ve fixed it three times now just by futzing around at the command line until eventually it starts working again. i don’t know what the cause or the fix is. does anyone else? When it was in that state, ‘pump Session’ was giving me sane values with an empty ‘model’ attire, and ‘pump model’ was giving me “”, with a sane value for ‘expires’ in pump.ini. So, no clues were had on my part. :)
Nathan Morse
@morsen
May 24 2016 10:57
Also, I’m confused about why the commands listed in the loop aliases here ( https://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-2/loop-and-retry-logic.html ) seem so different than the ones listed in the ‘full sequence’ here ( https://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-2/Using-oref0-tools.html#checking-your-reports ). May anyone please shed some light? thank you!
C-Ville
@C-Ville
May 24 2016 10:59
did anyone ever had the problem, that the mealbolus provided by the Bolus Calculator would not show up, instead the Carbs will show as Carb Correction. The AP algo seem to calculate with the IOB from the meal, but the IOB pill and the BWP pill from NS are way off. There it shows 0 IOB.
C-Ville
@C-Ville
May 24 2016 11:04
@morsen I figured hat is part of the "you have to understand what you are doing" part ;-). To be honest I assume that is due to different people contributing to the text and that it is very time consuming to always perform consistency checks in a fast growing document
Nathan Morse
@morsen
May 24 2016 11:12
well sure, but it doesn’t seem that having conflicting suggestions is helpful, and I want to understand the differences and the reasoning behind them
Dana Lewis
@danamlewis
May 24 2016 11:59
@cjo20 yes we are still going to niche :)
@cville from my understanding the bwp does not use the net iob which is why it will be didferent
Scott Leibrand
@scottleibrand
May 24 2016 12:07
I think @jasoncalabrese et. al. fixed the BWP to display OpenAPS IOB in the latest versions.
he might also be able to comment on the difference between meal bolus records and carb corrections, as he's been working on fixing some issues with duplicate carb correction entries on x23 pumps, which last I heard he thought were due to the carbs being entered and then the bolus wizard canceled
Andrew Baugh
@baughaw
May 24 2016 12:23
This message was deleted
Nathan Morse
@morsen
May 24 2016 12:43
So often getting this: AttributeError: 'Pump' object has no attribute ‘model’. I looked at the code briefly and am not sure whether it’s possibly an issue with openaps or with decocare. Who is the best person to talk to about this? I still don’t know how it randomly ‘resolves itself’. Thanks!
Dana Lewis
@danamlewis
May 24 2016 12:44
Is pump in good range when you get that error?
Paul Martin
@Spazholio
May 24 2016 12:49
I get it as well. I chalk it up to some kind of radio interference that gunks up the transmission between my rig and my pump. Periodically resetting the TI stick seems to mitigate it nicely.
Jason Calabrese
@jasoncalabrese
May 24 2016 13:02
@C-Ville @scottleibrand I think the issue I'm seeing is only from an x23, there's another similar issue where the pump history doesn't include the insulin till the bolus is complete
I have a check for that in my loop
Colin Lennon
@colinlennon
May 24 2016 13:03
Hi @C-Ville - what issues did you find when setting up Nightscout on your Pi? I might be able to help. I've got a Pi rig set up for offline use, with Nightscout running on it
I've been meaning to add some info on that to the docs... still on my to-do list!
Jason Calabrese
@jasoncalabrese
May 24 2016 13:05
&& test $(cat monitor/status.json | json bolusing) == false)
C-Ville
@C-Ville
May 24 2016 13:10
@jasoncalabrese thanks, I will check later today I do have a 715. I will also have to check older reports if the Bolus was ever correctly shown when using the bolus calculator
@colinlennon thanks for offering your help. I tried the git repo from https://github.com/dhermanns/rpi-nightscout and it did not work for me, latest try was somewhat different, where people mention how to setup docker on jessie. But then I got stuck.
Scott Leibrand
@scottleibrand
May 24 2016 13:35
@C-Ville if you don't have that bolusing check, the data will sometimes upload while the bolus is ongoing, and if that happens, it will never upload the bolus, as it thinks it already has a record from that timestamp.
Colin Lennon
@colinlennon
May 24 2016 13:51
@C-Ville yeah, I set up docker on Jessie then used that repo. I think I had to comment out a few lines in docker-compose.yml to get it working
SSL_KEY and SSL_CERT
I commented those out and I left SSL_CA commented out (it already is)
That worked for me. Might be worth a try
Jarred Yaw
@jyaw
May 24 2016 14:24
Anybody saw the pebble core kickstarter? That seems like an excellent platform for an nightscout uploader replacement, perhaps even a pi/Edison replacement?...
C-Ville
@C-Ville
May 24 2016 15:03
@colinlennon thanks, I have a long weekend coming up starting un thursday, now I have a project :-)
David Cintron
@loudestnoise
May 24 2016 15:09
@jyaw Just checked it out. Looks pretty neato. I’d love to see what people will do with it. That would be a great uploader for Nightscout. So tiny and has 3G!
Dan Piessens
@dpiessens
May 24 2016 15:36
@jyaw Have a url? :)
Nathan Morse
@morsen
May 24 2016 15:59
@danamlewis Thanks for your reply early. Wonderful child woke up early… :)
Dana Lewis
@danamlewis
May 24 2016 16:00
No worries. I'm in Chicago today :)
Nathan Morse
@morsen
May 24 2016 16:01
@danamlewis You know what? I’m assuming that he isn’t moving in bed, but that’s not a good assumption. I’ll check again tonight. But, at the same time, “use pump Session” works (with an empty model value) when the whole thing is in this state, but is that other Session stuff merely client-side?
@Spazholio Resetting the stick with oref0-reset-usb? (btw, I’m still using carelink)
Paul Martin
@Spazholio
May 24 2016 16:23
Correct, using oref0-reset-usb.
Lynne Butts
@flutelynne_twitter
May 24 2016 17:41
I set up the RPi2 as much as I can, I believe. Have a 515, but then a G5. I am still waiting on the CareLink USB. Tried setting up for Nightscout to read, added the variables, etc., but I cannot verify until Carelink is here?
Scott Leibrand
@scottleibrand
May 24 2016 17:42
Do you have the G5 part working?
Nathan Morse
@morsen
May 24 2016 17:47
@Spazholio thx
Lynne Butts
@flutelynne_twitter
May 24 2016 17:56
@scottleibrand I wasn't sure how to set that up. Guessing it is more than just plugging in the receiver to the Pi
LadyViktoria
@LadyViktoria
May 24 2016 18:12
cool
Scott Leibrand
@scottleibrand
May 24 2016 18:18
@flutelynne_twitter you'll need to switch to a G5 branch of dexcom_reader. @bewest do you have a pip command for that?
C-Ville
@C-Ville
May 24 2016 18:23
Sounds interesting the web ex
Ben West
@bewest
May 24 2016 18:48
pip install git+git://github.com/loudnate/dexcom_reader.git'#'loudnate/g5-receiver' minus typos
Lynne Butts
@flutelynne_twitter
May 24 2016 18:53
@bewest do I write '#' or is that the receiver SN? Also, is there a ' definitely at the end? Thanks
Ben West
@bewest
May 24 2016 19:08
sudo pip install git+git://github.com/loudnate/dexcom_reader.git'#loudnate/g5-receiver' like that :-)
Lynne Butts
@flutelynne_twitter
May 24 2016 19:14

sudo pip install git+git://github.com/loudnate/dexcom_reader.git'#loudnate/g5-receiver'
Downloading/unpacking git+git://github.com/loudnate/dexcom_reader.git
Cloning git://github.com/loudnate/dexcom_reader.git to /tmp/pip-ptAb9D-build
Running setup.py (path:/tmp/pip-ptAb9D-build/setup.py) egg_info for package from git+git://github.com/loudnate/dexcom_reader.git

warning: no files found matching 'CHANGES.*'
warning: no directories found matching 'etc'

Requirement already satisfied (use --upgrade to upgrade): pyserial in /usr/lib/python2.7/dist-packages (from dexcom-reader==0.0.7)
Installing collected packages: dexcom-reader
Found existing installation: dexcom-reader 0.0.8
Uninstalling dexcom-reader:
Successfully uninstalled dexcom-reader
Running setup.py install for dexcom-reader

warning: no files found matching 'CHANGES.*'
warning: no directories found matching 'etc'

Could not find .egg-info directory in install record for dexcom-reader==0.0.7 from git+git://github.com/loudnate/dexcom_reader.git
Successfully installed dexcom-reader
Cleaning up...
This is what I got.

So is it done if it says "cleaning up"
Lynne Butts
@flutelynne_twitter
May 24 2016 19:55
When I check on the receiver status, I get EGV
"could not parse EGV record"
Andrew Baugh
@baughaw
May 24 2016 20:26

I switched over to mmeowlink from the carelink stick. MMtune runs well... etc. My loop runs fine until it tries to report suggested.json.

reporting enact/suggested.json
Traceback (most recent call last):
  File "/usr/local/bin/openaps-use", line 6, in <module>
    exec(compile(open(__file__).read(), __file__, 'exec'))
  File "/home/pi/src/openaps/bin/openaps-use", line 63, in <module>
    app( )
  File "/home/pi/src/openaps/openaps/cli/__init__.py", line 51, in __call__
    self.run(self.args)
  File "/home/pi/src/openaps/bin/openaps-use", line 57, in run
    output = app(args, self)
  File "/home/pi/src/openaps/openaps/uses/__init__.py", line 92, in __call__
    return self.method.selected(args)(args, app)
  File "/home/pi/src/openaps/openaps/uses/__init__.py", line 31, in __call__
    return self.method(args, app)
  File "/home/pi/src/openaps/openaps/uses/use.py", line 45, in __call__
    output = self.main(args, app)
  File "/home/pi/src/openaps/openaps/vendors/medtronic.py", line 317, in main
    results = self.upload_program(program)
  File "/home/pi/src/openaps/openaps/vendors/medtronic.py", line 410, in upload_program
    return self.pump.model.set_temp_basal(**program)
  File "/home/pi/src/decoding-carelink/decocare/models/__init__.py", line 205, in set_temp_basal
    result = self._set_temp_basal(**basals)
  File "/home/pi/src/decoding-carelink/decocare/models/__init__.py", line 25, in __call__
    self.response = inst.session.query(self.msg, **kwds)
  File "/home/pi/src/decoding-carelink/decocare/session.py", line 101, in query
    self.execute(command)
  File "/home/pi/mmeowlink-source/mmeowlink/handlers/stick.py", line 225, in execute
    return sender(command)
  File "/home/pi/mmeowlink-source/mmeowlink/handlers/stick.py", line 139, in __call__
    self.prelude()
  File "/home/pi/mmeowlink-source/mmeowlink/handlers/stick.py", line 110, in prelude
    resp = Packet.fromBuffer(buf)
  File "/home/pi/mmeowlink-source/mmeowlink/packets/rf.py", line 81, in fromBuffer
    rftype   = buf[0]
IndexError: bytearray index out of range

Any ideas?

Andrew Baugh
@baughaw
May 24 2016 20:40
ok, I gave it a bit of reboot magic and it is now working. who knows...
ddaniels1
@ddaniels1
May 24 2016 23:07
Help, unplugged and replugged my Edison. Apparent Corruption, ran oref0-fix-git-corruption
error: object file .git/objects/a1/75ca48c60ce2a9a9620c43e897b53d38a39938 is empty
error: object file .git/objects/a1/75ca48c60ce2a9a9620c43e897b53d38a39938 is empty
fatal: loose object a175ca48c60ce2a9a9620c43e897b53d38a39938 (stored in .git/objects/a1/75ca48c60ce2a9a9620c43e897b53d38a39938) is corrupt
Error: Could not find last valid commit; aborting.
any suggestions?
ddaniels1
@ddaniels1
May 24 2016 23:22
this seems very bad. Now
doubled@ubilinux:~/Documents/try5$ oref0-fix-git-corruption
fatal: bad object HEAD
JaysonEwer
@JaysonEwer
May 24 2016 23:33
@ddaniels1 have you tried deleting that object?
The loose object that is.
amoskane
@amoskane
May 24 2016 23:34
O yes, the dreaded corrupted git repo scenario!
Jason Calabrese
@jasoncalabrese
May 24 2016 23:34
you could do the brute force method by going up a directory, rm -rf try5/.git and openaps init try5
since it looks like the normal tools to fix the issue aren't working for you
amoskane
@amoskane
May 24 2016 23:35
If that doesn’t work, or I get frustrated (takes 30 minutes or so), I just delete the repo on the pi and re-pull it. I’ve got that down to 1-2 minutes now. :)
You just have to restore your pump.ini file since that’s not a part of the repo (i put a gitignore rule on it to keep it private)
ddaniels1
@ddaniels1
May 24 2016 23:37
working now, with git reset
doubled@ubilinux:~/Documents/try5$ oref0-fix-git-corruption
fatal: bad object HEAD
Error: Could not find last valid commit; aborting.
doubled@ubilinux:~/Documents/try5$ oref0-reset-git
Attempting to fix git corruption.  Please wait 15s.
fatal: bad object HEAD
Error: Could not find last valid commit; aborting.

oref0-fix-git-corruption: no process found
fatal: bad object HEAD
Saving backup to: /tmp/git-1464132576
Initialized empty Git repository in /home/doubled/Documents/try5/.git/
How do I avoid this in the future? For Edison this time I didn't press the button to shut it down, just pulled plug, assume that was the issue.
amoskane
@amoskane
May 24 2016 23:39
Ya, when my pi battery dies I get this, and sometimes even if I pull the charge wire off the battery I get an unexpected reboot. Happens to me a lot.
MAkes me want to install a 'sudo shutdown' red button on the thing!
Paul Martin
@Spazholio
May 24 2016 23:41
…does such a thing exist? Because I absolutely want one. =)
C-Ville
@C-Ville
May 24 2016 23:44
There are ups systems for pi, which have a triggef to safely shutdown, when power to pi fails, but i have not tried these yet
Paul Martin
@Spazholio
May 24 2016 23:46
I wonder how they work? Is it some sort of USB thing that plugs into the power port that provides 30s of power? Just enough to power down cleanly?
amoskane
@amoskane
May 24 2016 23:49
Love this article, thanks!
ddaniels1
@ddaniels1
May 24 2016 23:57
I have an Edison rig not PI
amoskane
@amoskane
May 24 2016 23:59
You are a couple of weekends ahead of me, @ddaniels1 !