=== genii-core is now known as genii-meeting
=== genii-meeting is now known as genii
=== genii is now known as genii-core
=== not_phunyguy is now known as phunyguy
[11:31] <webchat12> I suspect openssl3 is causing issues to SSL connections over VPN
[11:32] <lotuspsychje> webchat12: we told you yesterday, you should clean install 22.04 daily iso, not upgrade in this devel stage yet
[11:35] <lotuspsychje> you could at least test a daily clean, and compare with your current issue webchat12 see if you can reproduce
[11:43] <webchat12> yes, when I have a sec
[16:06] <wherestheh> Hi, I tried to use global protect with a fresh install of 22.04 and same error with SSL handshake
[16:06] <wherestheh> I suspect the error in that case is in libqt5webkit5 package
[16:08] <wherestheh> not sure why I also have problems with rdp when connecting with forticlient
[16:09] <lotuspsychje> wherestheh: you can use ubuntu-bug packagename please to file your bug?
[16:13] <wherestheh> https://bugs.launchpad.net/ubuntu/+bug/1960268
[16:13] <ubottu> Launchpad bug 1960268 in Ubuntu "SSL handshake failed - VPN SSL broken in 22.04" [Undecided, New]
[16:14] <wherestheh> what's strange is that I have problems with xfreerdp too ...
[16:15] <lotuspsychje> wherestheh: your package is wrong on your bug ID, you will have to change it to a package affected first
[16:15] <lotuspsychje> then apport-collect 1960268 after to drag in your system info & logs into your bug
[16:15] <wherestheh> I have no idea where the problem lies, is it openssl3 ? qt5webkit ?
[16:16] <lotuspsychje> not sure myself sorry, dont use vpn's myself
[17:12] <lotuspsychje> ok great wherestheh now please apport-collect 1960268 fromp a terminal if you like
[17:14] <rbasak> wherestheh: the significant change is that in 22.04 we've moved to OpenSSL 3.0. I think our packaging might also come with some new requirements, such as a minimum TLS protocol version.
[17:14] <rbasak> It might be the case that the other end doesn't support a new-enough version of TLS. That would be where I'd look first - because if that's the case, it might not be considered to be a bug in Ubuntu at all.
[17:15] <rbasak> It might also be possible to configure your system locally to support older TLS versions (thus making it work but reducing security).
[17:15] <rbasak> I'm not familiar with the details though, sorry.
[17:15] <rbasak> Hopefully that helps with your research though.
[17:16] <rbasak> Tweaking /etc/ssl/openssl.cnf might be a way to do that.
[17:16] <wherestheh> so I should make openssl3 work like openssl2, kind of ?
[17:17] <rbasak> Googling about for how to tweak that file might help.
[17:19] <lotuspsychje> wherestheh: apport collect shows you're on 21.10 now?
[17:19] <wherestheh> yesterday I found a piece to add to the cnf file and I it was written to download   http://curl.haxx.se/ca/cacert.pem but it didn't help
[17:20] <wherestheh> mmm. you're right
=== genii-core is now known as genii
[18:10] <ASC__> Hi! Is "echo 'password' | openssl passwd -6 -salt xxx -stdin" now valid way to compare/check password against hash in /etc/shadow?
[18:12] <ASC__> For 22.04.
[18:34] <ASC__> What a hash type is '$y$'?
[18:35] <ASC__> (first field of ecnrypted pass in /etc/shadow - it's hash type and it's not $6$ as in earlier versions)
[18:45] <ASC__> It's 'yescrypt'.
[22:30] <KeeperoftheKeys> Hey I just upgraded my 21.10 install to jammy, I use btrfs on luks volumes and my boot went to shit due to libgcc_s.so.1 missing which prevented cryptsetup from decrypting the luks volumes
[22:33] <KeeperoftheKeys> After booting from a jammy live medium and chrooting into my real disks I modified  /usr/share/initramfs-tools/hook-functions to also add libgcc_s.so.1 when libcryptsetup.so. is a dependency found by ldd
[22:34] <KeeperoftheKeys> it seems that cryptsetup though failing to decrypt because it has to have libgcc for libpthread it is not actually linked against libpthread (at least according to ldd)
[22:34] <KeeperoftheKeys> thus libgcc does not get added
[22:35] <KeeperoftheKeys> After this I still got dropped to an initramfs rescue shell because I guess I'm still missing something there but was able to manually mount and boot the system
[22:36] <KeeperoftheKeys> This would seem to be a bug to me...
[22:36] <sarnold> is that "still missing something" the busybox thing? https://lists.ubuntu.com/archives/ubuntu-devel/2022-February/041824.html
[22:38] <KeeperoftheKeys> That was not the error I saw, but then again there wasn't a lot in errors on the screen.
[22:38] <KeeperoftheKeys> Now that I am able to boot I don't mind rebooting to see if I can get more error info
[22:38] <KeeperoftheKeys> if you tell me where to look
[22:39] <KeeperoftheKeys> (I won't be able to chat while I'm there since this is my daily personal (non-work) driver)
[22:40] <sarnold> heh, sorry, I just saw the email the other day and wondered if you hit the same thing, I was hoping to save you a bit of time if it was ;)
[22:40] <KeeperoftheKeys> would errors from the initramfs stage appear in dmesg?
[22:40] <sarnold> probably the journal, I'm not sure about dmesg
[22:44] <KeeperoftheKeys> the actual error shown on screen had to do with amdgpu which is weird since as far as I can tell my graphics are working fine
[22:44] <KeeperoftheKeys> (amd r7 4700u)
[22:45] <KeeperoftheKeys> and there is a stacktrace of a crash related to amdgpu before the good boot
[22:45] <sarnold> hmm, was it something about firmware files being unavailable? I think I see folks mention that from time time..
[22:45] <KeeperoftheKeys> [    2.415967] UBSAN: invalid-load in /build/linux-aa0B4d/linux-5.15.0/drivers/gpu/drm/amd/amdgpu/../display/amdgpu_dm/amdgpu_dm.c:6044:84
[22:47] <sarnold> oh wow, is compiling the kernel or modules with UBSAN something we do only in -devel? or is that a more general change?
[22:47] <KeeperoftheKeys> load of value 52 is not a valid value for type '_Bool
[22:48] <KeeperoftheKeys> I would not know I'm afraid :)
[22:49] <KeeperoftheKeys> https://www.mail-archive.com/ubuntu-bugs@lists.ubuntu.com/msg5990074.html
[22:50] <KeeperoftheKeys> looks like UBSAN is a kernel >=5.15 thing
[23:30] <nonix4> How should I link a git branch to launchpad bug report these days? Haven't added a changelog entry there yet though. https://git.launchpad.net/~iheino+ub/ubuntu/+source/bash/commit/?h=bugfix/1960366
[23:31]  * nonix4 actually meant that for #ubuntu-bugs...