ChanServ changed the topic of #yocto to: Welcome to the Yocto Project | Learn more: https://www.yoctoproject.org | Join us or Speak at Yocto Project Summit (2021.11) Nov 30 - Dec 2, more: https://yoctoproject.org/summit | Join the community: https://www.yoctoproject.org/community | IRC logs available at https://www.yoctoproject.org/irc/ | Having difficulty on the list or with someone on the list, contact YP community mgr ndec
Ad0 has quit [Ping timeout: 250 seconds]
xmn has joined #yocto
<fray> BTW reproduced my fetch problem from this morning.. no idea why it's happening again.. but at least I can keep digging into it
<fray> well confirmed.. it is a mirrors issue.. enters try_mirrors and doesn't seem to exit it
Tokamak has joined #yocto
<fray> found it.. ENAMETOOLONG in bb.utils.py triggers an endless loop
sakoman has quit [Quit: Leaving.]
vmeson has quit [Quit: Konversation terminated!]
vmeson has joined #yocto
Net147 has quit [Quit: Quit]
Net147 has joined #yocto
Net147 has joined #yocto
Net147 has quit [Changing host]
Emantor_ has quit [Quit: ZNC - http://znc.in]
Emantor has joined #yocto
RobertBerger has joined #yocto
rber|res has quit [Ping timeout: 256 seconds]
xmn has quit [Quit: ZZZzzz…]
sakoman has joined #yocto
jmiehe1 has joined #yocto
Tokamak has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
jmiehe has quit [Ping timeout: 250 seconds]
jmiehe1 is now known as jmiehe
troth has quit [Ping timeout: 256 seconds]
sakoman has quit [Quit: Leaving.]
troth has joined #yocto
davidinux has joined #yocto
kroon has joined #yocto
<RP> fray: someone was going to send me a patch related to that but never did send the v2 :(
deurzen has joined #yocto
davidinux has quit [Ping timeout: 264 seconds]
bradfa has quit [Ping timeout: 256 seconds]
bradfa has joined #yocto
nohit has quit [Ping timeout: 264 seconds]
nohit has joined #yocto
florian_kc has joined #yocto
florian_kc has quit [Ping timeout: 264 seconds]
camus has joined #yocto
Wouter0100 has quit [Read error: Connection reset by peer]
Wouter0100 has joined #yocto
gsalazar_ has quit [Ping timeout: 240 seconds]
hogspeed[m] has joined #yocto
goliath has joined #yocto
hogspeed[m] has left #yocto [#yocto]
paulg has quit [Remote host closed the connection]
kroon has quit [Quit: Leaving]
xmn has joined #yocto
camus has quit [Quit: camus]
camus has joined #yocto
sakoman has joined #yocto
<RP> jaskij[m]: leaving separate from bitbake seems reasonable to me, I'd be happy with a standalone tool for this
<jaskij[m]> RP: Cool. I'm still developing this, so once I feel it's finished I'll write here again.
<jaskij[m]> sidenote: I tried using PyDot to get pure Python parsing (as opposed to calling libgraphviz through a wrapper). Didn't work. The parse times were too long to consider.
chrfle has quit [Ping timeout: 260 seconds]
chrfle has joined #yocto
<fray> RP if you are still here.. I found two issues.. the lock one (which I've got a patch for), and also there are two commits in the mirror code that cause the mirror to go _nuts_ in my config.. Why, I have no idea..
<fray> if I revert the two commits then suddenly it all works, but I'm sure the bug they fixed also comes back
<fray> Was going to work on the patch in a few minutes and get them sent to bitbake-devel for discussion
<RP> fray: which two patches?
wwilly has joined #yocto
<fray> sec.. it's on my work VPN side, so it'll be a minute for me to find it
<fray> (poky commits)
<fray> fb2300c144bde
<fray> 4939a419823
<fray> remove those two and it fixes the issue..
<fray> so what I was seeing was in the mirror config, it decided to use a filename of: (looking it up in the log)
<RP> fray: I thought it would be those somehow. The first one you meantion was meant to fix the second
<fray> yup
<fray> it takes a url that has already been converted from "git" syntax to mirror syntax. then mangles the crap out of it..
<RP> fray: that does look somewhat entertaining :/
<fray> What it seems to be doing is for each letter in the /git2_git.qemu.org.git.seabios.git..tar.gz it basically grabs a letter from that string and inserts it
<fray> repeat for EACH letter..
<fray> so if you take that string and do a sed s/git2_git.qemu.org.git.seabios.git..tar.gz//g on it.. you get a BLANK result
<fray> so ya.. that is what is causing the ENAMETOOLONG
<RP> fray: its obviously looping uncontrolled. I don't have ideas offhand other than diving in and trying to figure out what the code is doing so a mailing list post and heopfully a way we can replicate would be good
<RP> right, strings like that would just overflow the filenames
<fray> but that too long is a problem that also should be fixed.. I should have a patch to the list in about 10-15 minutes for that
<RP> coll
<RP> cool
<fray> ONCE I found/fixed that.. it was easy to realize it was the mirror url stuff that was the problem and then just bisect from what changed in gatesgarth (where I know this works)
<fray> the bad part of this all, it works "sometimes".. i.e. if I do a bitbake -c cleanall on this recipe, it will start working.. I have NO IDEA why
<fray> it seems to only fail on a fresh build, which likely means it may even be task order related (again, I don't see any reason for that to be true, but thats the behavior I've seen)
jmiehe has quit [Ping timeout: 268 seconds]
<fray> Do you need me to send a report on those two patches to bitbake-devel?
<fray> need/want
<RP> fray: yes please. Is there a way I could reproduce?
<fray> I'm not sure.. that is the biggest issue right now.. I'll see if we have stuff that is public that might be able to
<fray> (we just pushed a bunch of preliminary honister workt o the public, so there is finally a chance for it)
<fray> ok name too long sent.. if you get too copied I apologize in advance.. I've got a mail config issue (no working reverse DNS from my ISP, so sometimes the mail stops, sometimes it doesn't)
<RP> fray: for the long lockfile name, I think we cna just truncate it to fit FWIW
<fray> I thought about that.. but I'm inclined to just fail.. and if it becomes a real-world problem, THEN we can truncate it
<RP> fray: it is a real world issue elsewhere :/
<fray> up to this point, the code has been the way it's been since 2017 or before.. so I don't think anyone is really going to care
<RP> hence the previous discussion
<RP> I can try and sort something out
<fray> the other places (like sstate-cache) we've always said the caller needs to figure out the truncation strategy.. so thats what I did here.. left it to the caller..
<fray> might be cleaner if we can except out way back out.. but I followed ht example of the 'can't write to the disk'
<fray> anyway, that was my thought process here
<fray> lol got an alert "motion detected at your front door".. go to the camera, don't see anyone.. replay.. squirrel running with a walnut full speed past my front door..
<RP> fray: sstate is a bit different to this though and the consequences of lock overlap aren't that bad
<RP> fray: I get a lot of motion detection from raindrops rolling on camera domes :)
<fray> ok.. email sent to bitbake-devel with details on the mirror stuff.. I'll spend a bit to see if I can reproduce it, but my regular build machine is offline.. so we'll see if I can
sakoman has quit [Quit: Leaving.]
<fray> All of my other cameras I have "motion" detection off, but I do have "object" detection on. It's pretty good for filtering out squirrels, while good at telling me someone is in my driveway (either by vehicle or walking)
sakoman has joined #yocto
FredO2 has quit [Remote host closed the connection]
FredO2 has joined #yocto
jmiehe has joined #yocto
jmiehe has quit [Client Quit]
dev1990 has joined #yocto
<fray> RP got you a reproducer.. it's a bit sketchy, but should give you a chance to reproduce it..
<fray> ok steps have been sent
florian_kc has joined #yocto
florian_kc is now known as florian
camus has quit [Quit: camus]
florian has quit [Ping timeout: 264 seconds]
ant__ has joined #yocto
sakoman has quit [Quit: Leaving.]
tangofoxtrot has quit [Remote host closed the connection]
tangofoxtrot has joined #yocto
sakoman has joined #yocto
florian has joined #yocto
manuel1985 has quit [Remote host closed the connection]
manuel1985 has joined #yocto
deurzen has quit [Quit: Leaving]
aivkiv has joined #yocto
florian has quit [Ping timeout: 264 seconds]
sakoman has quit [Quit: Leaving.]