I did cite 2 errors in the text you quoted, emonhub is reported as not running in emoncms despite “it tells me emonhub has been restarted, more than once” and at the end of the script it tells me the “service runner process cannot be found, and yet it has been started” and yet in emoncms it shows as running. Isuspect the latter is just a hangover from the recent changes to the service-runner perhaps. It is running so it is NOT the problem the log reports. But emonhub has failed to be restarted for some reason (or crashed out very soon after the update).
Then there is the 2 minor errors about rpi-rw at the start and rpi-ro towards the end, as you say nothing to worry about and easily fixed.
I am also not concerned about the fw not updating, I have no rfm2pi or emonpi board on the bare RPi at this point. That was expected, as was the lcd warning.
But if you look closer there are these errors
Your branch is up-to-date with 'origin/master'.
nothing to commit, working tree clean
fatal: unable to access 'https://github.com/openenergymonitor/emonpi.git/': gnutls_handshake() failed: Error in the pull function.
and
The following additional packages will be installed:
python-colorzero
Suggested packages:
python-gpiozero-docs
The following NEW packages will be installed:
python-colorzero
The following packages will be upgraded:
python-gpiozero python-rpi.gpio
2 upgraded, 1 newly installed, 0 to remove and 103 not upgraded.
Need to get 133 kB of archives.
After this operation, 261 kB of additional disk space will be used.
Do you want to continue? [Y/n] Abort.
In both cases there was obviously a working network connection, both seem odd, but the point here is that this is automated, I didn’t do anything to interfere with these operations and the network was up at this time. I have just checked and the apt-get installs (https://github.com/openenergymonitor/emonpi/blob/master/emonpiupdate#L18-L19) were not done, but succeeded first time when I just manually tried it.
And the emonpi repo wasn’t up to date either.
pi@emonpi:~/emonpi $ git status
On branch master
Your branch is up-to-date with 'origin/master'.
nothing to commit, working tree clean
pi@emonpi:~/emonpi $ git describe
emonSD-22Dec2015-709-g4c1d84f
pi@emonpi:~/emonpi $ git pull
remote: Enumerating objects: 139, done.
remote: Counting objects: 100% (139/139), done.
remote: Compressing objects: 100% (23/23), done.
remote: Total 210 (delta 124), reused 121 (delta 116), pack-reused 71
Receiving objects: 100% (210/210), 58.16 KiB | 0 bytes/s, done.
Resolving deltas: 100% (135/135), completed with 24 local objects.
From https://github.com/openenergymonitor/emonpi
4c1d84f..56ce5b2 master -> origin/master
18b8859..003ad11 cydynni -> origin/cydynni
* [new branch] feature/update-tmpfiles -> origin/feature/update-tmpfiles
* [new branch] remove-reset-hack -> origin/remove-reset-hack
* [new branch] service_correction -> origin/service_correction
* [new tag] 2.9.1 -> 2.9.1
Updating 4c1d84f..56ce5b2
Fast-forward
emoncmsupdate | 565 +++++++++++++++++++++++++++----------------
emonhub-sudoers | 2 +-
emonhubupdate | 53 +++-
emonpiupdate | 7 +-
firmware/platformio.ini | 1 +
firmware/src/src.ino | 2 +
firmware/src/temperature.ino | 77 ++++--
lcd/emonPiLCD.cfg | 36 +++
lcd/emonPiLCD.py | 518 +++++++++++++++++++++++----------------
rc.local_jessieminimal | 2 +
service-runner-update.sh | 12 +-
stretch/motd | 2 +-
stretch/rc.local | 3 +-
13 files changed, 832 insertions(+), 448 deletions(-)
create mode 100644 lcd/emonPiLCD.cfg
pi@emonpi:~/emonpi $
If the emonpi repo is not successfully updated, there really in’t much point in continuing the update process as you end up in between versions or crossed version might be a better term. A warning and abort would be better at that point if not successful.
The fact it was updating from an old version of the emonpi repo is why the update is significantly different to Trystans example.
What ever the reasons behind the fail, maybe it was network, maybe this will be better managed in systemd, either way, I am just reporting what happened when I simply booted to a new image, I had not interfered with anything other than pre-adding the wifi details as explained above.
