November 2022 - openSUSE Bugs - openSUSE Mailing Lists

[Bug 1205139] FTBFS: openSUSE:Factory/{gromacs,lammps,votca} fails to build (killed after 90 minutes without output)
by bugzilla_noreply＠suse.com 08 Nov '22

08 Nov '22

https://bugzilla.suse.com/show_bug.cgi?id=1205139 https://bugzilla.suse.com/show_bug.cgi?id=1205139#c9 Nicolas Morey-Chaisemartin <nmoreychaisemartin(a)suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|CONFIRMED |IN_PROGRESS --- Comment #9 from Nicolas Morey-Chaisemartin <nmoreychaisemartin(a)suse.com> --- SR sent: https://build.opensuse.org/request/show/1034518 I've confirmed it fixes the issue at least for * python-pytest-mpi (without the ob1 change) * python-mpi4py (there was no second bug, some tests are just much longer than the others) * sundials I'm waiting on votca's results but seeing the 3 other tests, I'm pretty confident -- You are receiving this mail because: You are on the CC list for the bug.

1 0

[Bug 1202141] New: YaST samba-client - AD joined VM - pam stack issues / winbind auth / console login / ssh login ...
by bugzilla_noreply＠suse.com 08 Nov '22

08 Nov '22

http://bugzilla.opensuse.org/show_bug.cgi?id=1202141 Bug ID: 1202141 Summary: YaST samba-client - AD joined VM - pam stack issues / winbind auth / console login / ssh login ... Classification: openSUSE Product: openSUSE Tumbleweed Version: Current Hardware: x86-64 OS: openSUSE Tumbleweed Status: NEW Severity: Normal Priority: P5 - None Component: YaST2 Assignee: yast2-maintainers(a)suse.de Reporter: rs.opensuse(a)spitzenpfeil.org QA Contact: jsrain(a)suse.com Found By: --- Blocker: --- Since 'samba-client' doesn't crash anymore on TW, I've managed to join our AD, enabled winbind auth, pam_mount ... (all of this works on Leap 15.2 / 15.4 with samba packages coming from "home:markusd:samba-fresh" 4.16.4-lp154.1.1). After the motions with YaST samba-client I could not log on text consoles anymore, ssh login did not work anymore ... sddm login still works. I then copied the whole contents of /etc/pam.d/ from a known good Leap 15.2 installation to my TW VM and everything works again. When I have some more time, I'll diff the auto-created files that YaST spits out and maybe find something. Related: https://bugzilla.opensuse.org/show_bug.cgi?id=1199734 -- You are receiving this mail because: You are on the CC list for the bug.

1 15

[Bug 1205139] FTBFS: openSUSE:Factory/{gromacs,lammps,votca} fails to build (killed after 90 minutes without output)
by bugzilla_noreply＠suse.com 08 Nov '22

08 Nov '22

https://bugzilla.suse.com/show_bug.cgi?id=1205139 https://bugzilla.suse.com/show_bug.cgi?id=1205139#c8 Nicolas Morey-Chaisemartin <nmoreychaisemartin(a)suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |CONFIRMED Assignee|screening-team-bugs(a)suse.de |nmoreychaisemartin(a)suse.com Flags|needinfo?(nmoreychaisemarti | |n(a)suse.com) | --- Comment #8 from Nicolas Morey-Chaisemartin <nmoreychaisemartin(a)suse.com> --- There are at least 2 bugs. The 'cm' pml deadlocks when starting without any network interface available. I'm tracking the culprit down. It's calling the ofi mtl which fails to nit (as expected as there are no netdev apart from lo) but when trying to clean up, we end up in a deadlock for some reason. Not sure if it's openmpi of libfabric's fault This one explains different behaviour when building in chroot (which usually has netdevs) vs kvm (which doesn't) For python-mpi4py, there is a later deadlock which I haven't had time to look into yet. Might be a separate issue from the other packages(In reply to Atri Bhattacharya from comment #7) > (In reply to Daniel Garcia from comment #5) > Will this be done for the openmpi packages so that mpiexec sets this > directly, or should we do this for each package in its specific specfile? No. The cm pml is important and ob1 usage is quite restricted (bad with RDMA hardware). This just hides the first bug above. -- You are receiving this mail because: You are on the CC list for the bug.

1 0

[Bug 1189172] USB CDC-ACM device keeps disconnecting and reconnecting
by bugzilla_noreply＠suse.com 08 Nov '22

08 Nov '22

https://bugzilla.suse.com/show_bug.cgi?id=1189172 https://bugzilla.suse.com/show_bug.cgi?id=1189172#c8 --- Comment #8 from Marc Balmer <marc(a)msys.ch> --- The scanner is operation as long as the machine is in the grub bootloader. As soon as the kernel boots, the scanner goes into a non-operational mode. There must be a bug in the cdc_acm driver/subsystem, that was introduced with openSUSE Leap 15.3 that that persists till today. atm, the machine is running the 5.14.21-150400.24.28-default kernel. -- You are receiving this mail because: You are on the CC list for the bug.

1 0

[Bug 1205043] Installation of libpoppler-devel
by bugzilla_noreply＠suse.com 08 Nov '22

08 Nov '22

https://bugzilla.suse.com/show_bug.cgi?id=1205043 https://bugzilla.suse.com/show_bug.cgi?id=1205043#c9 --- Comment #9 from Stefan Hundhammer <shundhammer(a)suse.com> --- Adam, thanks for that explanation. That makes perfect sense. But I wonder if we can't reflect the fact that the other libboost packages are not meant for normal production use clearer to users; maybe by adding something to their names (since they already contain the version number). I think it's perfectly understandable that a user, much more so a developer, installs the highest version number that is available, and that's probably how the reporter got those packages installed. As for myself, I just added the "C++ development" pattern when I installed and didn't look twice, so I got the default ("system") boost packages. But had I looked in more detail, I might also have been tempted to install the higher numbers, and I might also have run into similar trouble. IIRC we have some "notice pop-up" mechanism to display messages when a user selects a package for installation, but I am not sure if they are only displayed when a user uses YaST, or also if using a zypper command line. But a package name like "libboost_xy_1_75_experimental" or "_compat" or whatever for sure would have made me think twice. Maybe that would be a way to avoid this situation. Just my two cents. -- You are receiving this mail because: You are on the CC list for the bug.

1 0

[Bug 1189172] USB CDC-ACM device keeps disconnecting and reconnecting
by bugzilla_noreply＠suse.com 08 Nov '22

08 Nov '22

https://bugzilla.suse.com/show_bug.cgi?id=1189172 https://bugzilla.suse.com/show_bug.cgi?id=1189172#c7 --- Comment #7 from Marc Balmer <marc(a)msys.ch> --- Setting the ID_MM_DEVICE_IGNORE flag does not change anything. In fact, I even deinstalled ModemManager, no change either. -- You are receiving this mail because: You are on the CC list for the bug.

1 0

[Bug 1189172] USB CDC-ACM device keeps disconnecting and reconnecting
by bugzilla_noreply＠suse.com 08 Nov '22

08 Nov '22

https://bugzilla.suse.com/show_bug.cgi?id=1189172 Marc Balmer <marc(a)msys.ch> changed: What |Removed |Added ---------------------------------------------------------------------------- Version|Leap 15.3 |Leap 15.4 OS|openSUSE Leap 15.3 |openSUSE Leap 15.4 -- You are receiving this mail because: You are on the CC list for the bug.

1 0

[Bug 1204604] [Build 20221022] wicked: openQA test fails in t08_setup_second_card
by bugzilla_noreply＠suse.com 08 Nov '22

08 Nov '22

https://bugzilla.suse.com/show_bug.cgi?id=1204604 https://bugzilla.suse.com/show_bug.cgi?id=1204604#c1 Clemens Famulla-Conrad <cfamullaconrad(a)suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |cfamullaconrad(a)suse.com --- Comment #1 from Clemens Famulla-Conrad <cfamullaconrad(a)suse.com> --- Looks like a network issue in OpenQA. In the PCAP of the SUT [1] you can find the DHCP REQUEST but these requests do never reach the DHCP-SERVER(REF) [2] I think the reason why this issue didn't appear before is because of this change https://github.com/openSUSE/wicked/pull/934 where we reuse an existing LEASE once the REQUEST times out. So we should make sure that the connection is working with the DHCP-Server (e.g. wait Or use DHCLIENT_USE_LAST_LEASE=no . Compared to the last good run [3], there is the working DHCP-REQUEST after 22s. All this looks like some bridge setup issues with RSTP depends on the openvswitch configuration, which takes 15s to be ready. [1] https://openqa.opensuse.org/tests/2859798/file/t08_setup_second_card_tcpdum… [2] https://openqa.opensuse.org/tests/2859596/file/t08_setup_second_card_tcpdum… [3] https://openqa.opensuse.org/tests/2826895/file/t08_setup_second_card_tcpdum… -- You are receiving this mail because: You are on the CC list for the bug.

1 0

[Bug 1205043] Installation of libpoppler-devel
by bugzilla_noreply＠suse.com 08 Nov '22

08 Nov '22

https://bugzilla.suse.com/show_bug.cgi?id=1205043 https://bugzilla.suse.com/show_bug.cgi?id=1205043#c8 --- Comment #8 from Adam Majer <amajer(a)suse.com> --- TL;DR: You need to install Boost 1.66 (default system boost) to link with libpoppler-devel. Other boost versions *may* not work. They have different ABI from the ones used to link libpoppler runtime and you'll *may* get segfaults. For extra details, (In reply to Arvin Schnell from comment #7) > So as I just found out there are three boost versions in Leap 15.4: > 1.66 (boost), 1.71 (boost171) and 1.75 (boost.175). > > AFAIU the newer two were just added as workarounds and should likely not > be normally installed on any system (esp. not on Leap, maybe on SLE). At least one of these was added for HPC module only but somehow "leaked" out :-) The naming is also consequence of going through Factory. After adding boost-defaults package that provides real package with libboost-headers-devel, the Provides() were renamed to add *-impl so the names don't conflict. The idea at the time was to be able to upgrade boost-devel packages in SLES Service Packs. This turns out is impossible due to unversioned boost symbols -- they would clash at runtime if different libraries imported different boost libraries into one executable. Another reason for removing Provides() and using real packages is that people were not seeing libboost-headers-devel and related in zypper search results and that was a complaint. And yet another reason to have the package with versioned RPM names is that OBS images don't have versions in the filename. So boost 1.66 and boost 1.75 would overwrite themselves on the OBS side if they produced a binary libboost-headers-devel with different versions. But, regardless, the problem here is that libpoppler-devel requires system liboost_headers-devel but the Requires: is somewhat misleading with the >= instead of = So, unless you really know what you are doing, on Leap or SLES, use system boost 1.66 for devel. On TW, there is only one Boost so it's easier to decide. -- You are receiving this mail because: You are on the CC list for the bug.

1 0

[Bug 1204860] [SELinux] Installed selinux (after fresh install): no labeling, got locked out
by bugzilla_noreply＠suse.com 08 Nov '22

08 Nov '22

https://bugzilla.suse.com/show_bug.cgi?id=1204860 https://bugzilla.suse.com/show_bug.cgi?id=1204860#c5 Thorsten Kukuk <kukuk(a)suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution|--- |INVALID --- Comment #5 from Thorsten Kukuk <kukuk(a)suse.com> --- No idea why this bug was assigned without any comment to me, I have nothing to do with this package. And no idea why somebody installs random packages without any meaningful description on his system and is wondering afterwards, that his system is broken. That's expected. This package should never be installed on a traditional Tumbleweed installation and will never be installed there by our patterns if somebody installs SELinux, only if he uses transactional-update. We have patterns to install SELinux, and this patterns do it correct. -- You are receiving this mail because: You are on the CC list for the bug.

1 0