<sa7mfo>
Hello there, I want to create a directory structure to be included into a (non-root) partition specified in a wic file. Some directories should not be owned by root. What is the best approach? I thinking about creating a new image and use the rawcopy plugin in the wic file, or is there any better options? Thanks in advance
<sa7mfo>
It was not as easy as I hoped for. Any suggestions?
jmd has quit [Remote host closed the connection]
alessioigor has joined #yocto
<landgraf>
sa7mfo: that should be easy. just put files under such directory and add partition to wic with --source rootfs --rootfs-dir=${IMAGE_ROOTFS}/yourdir (do not forget to exlude them from the / partition)
<sa7mfo>
landgraf, Oh, I will try that, thank you!
<RP>
abelloni: hopefully that is the mingw issues sorted!
altru has quit [Quit: Client closed]
jmd has joined #yocto
Guest97 has joined #yocto
Saur_Home34 has quit [Quit: Client closed]
Saur_Home34 has joined #yocto
<Guest97>
I am trying to use PREMIRROR to mirror few repos. I am able to mirror repositories to an private git mirror using premirror. However when GENERATE_MIRROR_TARBALL is enabled it is created using the premirror name . When a native recipe try to fetch, it skips the pre-mirror (because the repo is downloaded previously) and then fails (it fails during a
<Guest97>
check where looking for the original url name in the tarball) and my download directory is having the mirror tarball with the private repo name in it . Any clue how to work around this ?
alessioigor has quit [Quit: Client closed]
<RP>
Guest97: it sounds a bit like a bug :/
sng has quit [Quit: No Ping reply in 180 seconds.]
sakoman has quit [Remote host closed the connection]
sakoman has joined #yocto
enok has quit [Ping timeout: 240 seconds]
alessioigor has quit [Quit: Client closed]
mbulut has quit [Ping timeout: 252 seconds]
<kilobyte_ch>
Is there a good guide on how to use git repositories with git submodules in SRC_URI? Always having weird issues with that.
<kilobyte_ch>
Weird issues with things like this: 'do_fetch: Fetcher failure: Unable to find revision a495f76317cf4c5c68e671ef3b9762624b0e3982 in branch even from upstream'
<kilobyte_ch>
When cloning the same repo in CLI and doing a git submodule update --init everything works fine. But bitbake can't deal with it. I'm using gitsm://.
applepi has joined #yocto
mbulut has joined #yocto
florian_kc has joined #yocto
alessioigor has joined #yocto
florian__ has quit [Ping timeout: 260 seconds]
goliath has joined #yocto
<kilobyte_ch>
hmm, is there any reason why a do_configure:prepend() and do_patch:prepend() behaves differently in a bitbake? With do_configure it works, but with do_patch it throws errors like that cd ${WORKDIR}/git is invalid syntax.
<JaMa>
do_patch is python task, do_configure is shell
<dl9pf>
Can you specify the minimum python version required for hashserv?
mbulut has quit [Ping timeout: 260 seconds]
<dl9pf>
i had luck running it on a ubuntu 24.04 (Python 3.12 iirc) but it failed w/ the python thats part of the buildtools in scarthgap
<RP>
dl9pf: are you asking about the client side or the server side and is this with websockets or not?
<dl9pf>
no websockets, server side
<dl9pf>
i have can run scarthgap as client if i start the server on Ubuntu 24.04 , same setup but older Server os, bang. branch scarthgap everywhere
<dl9pf>
we tried different base os in FROM w/ the same hashserv being started inside
vthor has joined #yocto
vthor has quit [Changing host]
vthor has joined #yocto
<RP>
dl9pf: what is puzzling me is that the log you share above is from the client?
<dl9pf>
yes, thats from the client. difference was the server container os
<dl9pf>
it Puzzles me as well that the client barfs , but thats my datapoint atm.
<dl9pf>
is the websocket setup documented ?
<RP>
dl9pf: are there errors in the server logging?
<dl9pf>
we saw none ...
<RP>
dl9pf: I have no idea about the docs :/
<dl9pf>
well, then i only know non-ws obbiously
<RP>
dl9pf: bitbake-hashserv will use websockets if you specify a ws:// address most likely, as per --help
<JaMa>
I'm using 24.04 for both server and client but only over unix://
applepi has quit [Quit: Client closed]
<RP>
I know we have many different os using websockets as clients. 2204 is the only one I know of issues with client side and I tried to make that clear about the error scenarios
<JPEW>
dl9pf: So, no logs from the server? Does it even run?
enok has joined #yocto
<dl9pf>
i see the startup messages
<dl9pf>
but no message when the error occurrs. prserv runs meanwhile in the same container fine
<JPEW>
How are you starting the server?
<JPEW>
(command line)
<dl9pf>
wild guess: python Version .... whats the minimum for asyncrpc to be Happy?
<JPEW>
We don't have a specific check for that; it's supposed to be the same as the minimum version of bitbake
<embetrix>
I have in some of my application the usage of dlopen and I would like to use it without referencing the version which may change over time.
<embetrix>
Is there anyway to create Non-versioned shared object symlinks for all shared objects ?
<embetrix>
on the otherside classical linux distros have that already
<kilobyte_ch>
How do I know which tasks (do_patch, do_configure) are shell and which are python? And can I run a command like 'git submodule update --init --recursive' in a python task?
<embetrix>
better to use SRC_URI gitscm:// for submodules
embetrix has quit [Quit: Client closed]
alessioigor has quit [Quit: Client closed]
<Saur>
embetrix: The versioned DSOs in /lib are typically not loaded using dlopen(), but rather by the linker when starting the application. dlopen() is more typically used to load unversioned DSOs that are used as plugin. However, they are typically not distributed in /lib, but rather in an application specific directory (which may be a subdirectory of /lib). The unversioned DSOs in /lib are normally only used when building and thus they are pa
<JPEW>
dl9pf: Ya, I tried it out. Its making the TCP connection, but something is terminating it; proxy? or the server itself doesn't like something? If the later, I would expect to see logs in the server, but maybe it needs better logging
<dl9pf>
how can we get the Output we need... strace ? add extra logging ?
<JPEW>
The latter probably
<JPEW>
You have _no_ logs from the server?
<dl9pf>
no, we tailf the logfail via docker logs -f , nothing
<dl9pf>
will dig some more.
<dl9pf>
or find a reproducer
Piraty has quit [Quit: -]
Piraty has joined #yocto
<JPEW>
dl9pf: Ya, it is bizzare. There were some Python version dependencies, but they were all older stuff like 3.5, 3.6 ish. There really hasn't been much that should keep it from working on 3.8 & later
<dl9pf>
ok, good to know . its a headscratcher...
simonew has quit [Remote host closed the connection]
Vonter has quit [Ping timeout: 268 seconds]
goliath has quit [Quit: SIGSEGV]
goliath has joined #yocto
gabriele_00 has joined #yocto
gabriele_00 has quit [Quit: gabriele_00]
mvlad has quit [Remote host closed the connection]
Kubu_work has quit [Quit: Leaving.]
jmd has quit [Remote host closed the connection]
_whitelogger_ has quit [Ping timeout: 260 seconds]
_whitelogger has joined #yocto
florian_kc is now known as florian
xmn has quit [Ping timeout: 272 seconds]
Ad0 has quit [Ping timeout: 260 seconds]
enok has joined #yocto
enok has quit [Ping timeout: 240 seconds]
florian has quit [Ping timeout: 252 seconds]
RyanEatmon has joined #yocto
<RyanEatmon>
Is the scarthgap hash equiv code currently compatible with the latest master hash equiv server?
<RP>
RyanEatmon: yes
<RyanEatmon>
Cool.
<RP>
RyanEatmon: we've a performance fix we're going to backport to the scarthgap client code but it is all compatible right now and going forward
<RyanEatmon>
I'm testing the latest master server and client to see if it fixes out performance issues.