Cannot connect on first launch

[tormento]

I have one minor bugs on Windows:

I use nightly builds. After every manual update, i.e. unpacking the .exe
and replacing older files, on the first launch I get the "automatic check
for updates" windows. I can accept or decline but the following effect is
that the first connection to cloud goes timeout and I get lower window red
message "Cannot connect to cloud server, working with local copy". I have
to close Subsurface and on the next use everything goes ok.

[Dirk Hohndel]

I use nightly builds. After every manual update, i.e. unpacking the .exe and replacing older files, on the first launch I get the "automatic check for updates" windows. I can accept or decline but the following effect is that the first connection to cloud goes timeout and I get lower window red message "Cannot connect to cloud server, working with local copy". I have to close Subsurface and on the next use everything goes ok.

That's strange. Can you run this with '-v -v' and share the log from an unsuccessful attempt?

Thanks

/D

[tormento]

I am currently running 4.8.0-4. The latest version I can see on GitHub is
4.8.0-2. Is it a regression from the version I am currently using?

Post by tormento

Post by tormento
I use nightly builds. After every manual update, i.e. unpacking the .exe

and replacing older files, on the first launch I get the "automatic check
for updates" windows. I can accept or decline but the following effect is
that the first connection to cloud goes timeout and I get lower window red
message "Cannot connect to cloud server, working with local copy". I have
to close Subsurface and on the next use everything goes ok.
That's strange. Can you run this with '-v -v' and share the log from an
unsuccessful attempt?
Thanks
/D

[Dirk Hohndel]

I am currently running 4.8.0-4. The latest version I can see on GitHub is 4.8.0-2. Is it a regression from the version I am currently using?

You got the continuous build from Linus' PR. So you were ahead of master. I have merged a couple of PRs in the last half hour, including Linus', so you should see a new continuous binary show up fairly soon that's once again 4.8.0.4. I'm looking through the other PRs and am trying to fix the non-detection of the OSTC2 that was reported this morning, so the version number should keep going up in the next few hours :-)

/D

[tormento]

Sent a private with log.

Tell me if you need other.

Alberto

Post by tormento

Post by tormento
I use nightly builds. After every manual update, i.e. unpacking the .exe

and replacing older files, on the first launch I get the "automatic check
for updates" windows. I can accept or decline but the following effect is
that the first connection to cloud goes timeout and I get lower window red
message "Cannot connect to cloud server, working with local copy". I have
to close Subsurface and on the next use everything goes ok.
That's strange. Can you run this with '-v -v' and share the log from an
unsuccessful attempt?
Thanks
/D

[Dirk Hohndel]

Thanks for sending the log. The error is really weird. We get the correct response back from the server; we use a personalized silly variation of the "I'm a teapot" protocol to recognize our cloud server - and we get the right response back, yet the code seems to think it's an error.
I need to stare at that code a bit longer, but this makes no sense, TBH.

Would you mind creating a GitHub issue and attaching that Log there - this way others can take a look and maybe help me see what we are doing wrong.

Thanks

/D

Post by tormento
Sent a private with log.
Tell me if you need other.
Alberto

I use nightly builds. After every manual update, i.e. unpacking the .exe and replacing older files, on the first launch I get the "automatic check for updates" windows. I can accept or decline but the following effect is that the first connection to cloud goes timeout and I get lower window red message "Cannot connect to cloud server, working with local copy". I have to close Subsurface and on the next use everything goes ok.

That's strange. Can you run this with '-v -v' and share the log from an unsuccessful attempt?
Thanks
/D
_______________________________________________
subsurface mailing list
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

[tormento]

Couldn’t be simply a timeout or handshake error caused by the update check?
It happens only on that event.

Anyway I need to study a bit GitHub as I have never contributed directly.
Could you please give me also an idea on how to anonymoize log? Personal
data there :)

Post by Dirk Hohndel
Thanks for sending the log. The error is really weird. We get the correct
response back from the server; we use a personalized silly variation of the
"I'm a teapot" protocol to recognize our cloud server - and we get the
right response back, yet the code seems to think it's an error.
I need to stare at that code a bit longer, but this makes no sense, TBH.
Would you mind creating a GitHub issue and attaching that Log there - this
way others can take a look and maybe help me see what we are doing wrong.
Thanks
/D
Sent a private with log.
Tell me if you need other.
Alberto

Post by tormento
I use nightly builds. After every manual update, i.e. unpacking the

.exe and replacing older files, on the first launch I get the "automatic
check for updates" windows. I can accept or decline but the following
effect is that the first connection to cloud goes timeout and I get lower
window red message "Cannot connect to cloud server, working with local
copy". I have to close Subsurface and on the next use everything goes ok.
That's strange. Can you run this with '-v -v' and share the log from an
unsuccessful attempt?
Thanks
/D
_______________________________________________

subsurface mailing list
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

[tormento]

Tried my first bug report to GitHub :)

Here's what you could add to the GitHub entry - if this is too much work,
I'll be happy to do that for you... it's simply a way to ensure that we
have a record of the issue and other developers have a reminder to take a
look...
Subsurface v4.8.0-80-gdc2f0f8e48e1,
built with libdivecomputer v0.7.0-devel-Subsurface-NG
(02560a7e7fe82919d584d3edbf3876f90382052c)
built with Qt Version 5.9.2, runtime from Qt Version 5.9.2
built with libgit2 0.26.0
[...]
sync with remote
git storage: Sync with cloud storage
delete proxy setting
Cloud storage: checking connection to cloud server
Checking cloud connection...
setGeometry: Unable to set geometry 120x30+885+485 on
QWidgetWindow/'QMessageBoxClassWindow'. Resulting geometry: 450x96+885+485
(frame: 8, 31, 8, 8, custom margin: 0, 0, 0, 0, minimum size: 450x96,
maximum size: 450x96).
git storage: Waiting for cloud connection (1 second(s) passed)
git storage: Waiting for cloud connection (2 second(s) passed)
git storage: Waiting for cloud connection (3 second(s) passed)
git storage: Waiting for cloud connection (4 second(s) passed)
git storage: Cloud connection failed
connection test to cloud server failed
QNetworkReply::NetworkError(ProtocolInvalidOperationError) "Error
transferring
https://cloud.subsurface-divelog.org//make-latte?number-of-shots=3 -
server replied: I'm a teapot" 418 "Linus does not like non-fat milk"
Cloud storage: unable to connect to cloud server
git storage: Can't reach cloud server, working with local data
git storage: Load dives from local cache
git load handling file 00-Subsurface
This snippet contains all the relevant data and nothing that is personal
info, I think.
/D
Couldn’t be simply a timeout or handshake error caused by the update
check? It happens only on that event.
Anyway I need to study a bit GitHub as I have never contributed directly.
Could you please give me also an idea on how to anonymoize log? Personal
data there :)

Post by Dirk Hohndel
Thanks for sending the log. The error is really weird. We get the correct
response back from the server; we use a personalized silly variation of the
"I'm a teapot" protocol to recognize our cloud server - and we get the
right response back, yet the code seems to think it's an error.
I need to stare at that code a bit longer, but this makes no sense, TBH.
Would you mind creating a GitHub issue and attaching that Log there -
this way others can take a look and maybe help me see what we are doing
wrong.
Thanks
/D
Sent a private with log.
Tell me if you need other.
Alberto

Post by tormento
I use nightly builds. After every manual update, i.e. unpacking the

.exe and replacing older files, on the first launch I get the "automatic
check for updates" windows. I can accept or decline but the following
effect is that the first connection to cloud goes timeout and I get lower
window red message "Cannot connect to cloud server, working with local
copy". I have to close Subsurface and on the next use everything goes ok.
That's strange. Can you run this with '-v -v' and share the log from an
unsuccessful attempt?
Thanks
/D
_______________________________________________

subsurface mailing list
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface
_______________________________________________

subsurface mailing list
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface