We've branched, so let's start the weekly blocker status email yayyyyy
Action summary ====================
Accepted blockers ----------------- 1. libreport — abrt-server errors when processing zstd compressed core dumps produced by systemd-246~rc1-1.fc33 — POST ACTION: abrt maintainers to make a new release containing the fix
2. resteasy — FreeIPA deployment fails in current Rawhide due to various issues with Java 11 — NEW ACTION: resteasy maintainers to diagnose issue and fix or pass the buck as appropriate
3. sddm — login stuck when changing users repeatedly (log out, log in a different one) — NEW ACTION: sddm maintainers to diagnose issue and fix or pass the buck as appropriate
4. selinux-policy — SELinux is preventing systemd-machine from 'create' accesses on the sock_file io.systemd.Machine. — POST ACTION: selinux-policy maintainers to make a new release containing the fixes
5. tomcat — FreeIPA server deployment fails in Fedora-Rawhide-20200714.n.0 — MODIFIED ACTION: tomcat maintainers to investigate failed upgrade test
Proposed blockers -----------------
1. gnome-shell — Network manager started stuck when I tries connecting to VPN (openconnect) after upgrade gnome-shell to 3.37.1-1.fc33 version — NEW ACTION: gnome-shell maintainers to diagnose issue
2. systemd — systemd-resolved.service not work with DNS server placed behind VPN (openconnect) — NEW ACTION: systemd maintainers to diagnose issue
3. udisks2 — zram-setup@zram0.service: Failed to load configuration: No such file or directory — NEW ACTION: Someone! to figure out why udisks2-zram gets pulled in
Bug-by-bug detail =============
Accepted blockers ----------------- 1. libreport — https://bugzilla.redhat.com/show_bug.cgi?id=1860616 — POST abrt-server errors when processing zstd compressed core dumps produced by systemd-246~rc1-1.fc33
abrt doesn't support core dumps compressed with zstd, which is now used by systemd. Merged upstream PR uses libarchive to add support for zstd (and other compression formats): https://github.com/abrt/libreport/pull/656.
2. resteasy — https://bugzilla.redhat.com/show_bug.cgi?id=1866570 — NEW FreeIPA deployment fails in current Rawhide due to various issues with Java 11
Deploying FreeIPA fails with a traceback in pki-tomcat. This appears to be a dependency chain issue that (currently) ends at resteasy via dogtag-pki. This may not be the last layer in the Java 11 onion.
3. sddm — https://bugzilla.redhat.com/show_bug.cgi?id=1861700 — NEW login stuck when changing users repeatedly (log out, log in a different one)
This is now a blocker based on a recent approval-in-principle of a release criterion related to logout and user switching. User processes linger after logout which blocks logging in when another user has logged in between the two sessions. Or when the second user logs back in. Or when a single user logs in repeatedly. Using KillUserProcesses=Yes helps the first problem, but raises on of its own. It appears to be a race condition of some kind as the behavior is not fully consistent.
4. selinux-policy — https://bugzilla.redhat.com/show_bug.cgi?id=1862686 — POST SELinux is preventing systemd-machine from 'create' accesses on the sock_file io.systemd.Machine.
librvirt VMs and systemd-machined service fail with AVC denials. Merged upstream PRs should fix the denials mentioned in this bug: https://github.com/fedora-selinux/selinux-policy/pull/405 and https://github.com/fedora-selinux/selinux-policy/pull/407
5. tomcat — https://bugzilla.redhat.com/show_bug.cgi?id=1857043 — MODIFIED FreeIPA server deployment fails in Fedora-Rawhide-20200714.n.0 due to pki-tomcat failing to run with "java.lang.ClassNotFoundException: org.apache.tomcat.util.modeler.modules.MbeansDescriptorsIntrospectionSource"
An upstream commit resulted in tomcat-coyote.jar missing packages. A previous fix (FEDORA-2020-f897a68801) moved the goalposts. tomcat-9.0.37-3 should fix this. Verification is blocked on BZ1866570, but Adam's most recent upgrade test still fails with a ClassNotFoundException for com.sun.xml.internal.bind.v2.ContextFactory.
Proposed blockers -----------------
1. gnome-shell — https://bugzilla.redhat.com/show_bug.cgi?id=1830343 — NEW Network manager started stuck when I tries connecting to VPN (openconnect) after upgrade gnome-shell to 3.37.1-1.fc33 version
Connecting to an OpenConnect VPN using 2FA hangs after the second factor is entered. The journal contains an array.toString() message which may or may not be a red herring (it looks like a deprecation warning).
2. systemd — https://bugzilla.redhat.com/show_bug.cgi?id=1863041 — NEW systemd-resolved.service not work with DNS server placed behind VPN (openconnect)
When connected to an OpenConnect VPN, host name resolution fails. Manually disabling the systemd-resolved service works around this issue.
3. udisks2 — https://bugzilla.redhat.com/show_bug.cgi?id=1861463 — NEW zram-setup@zram0.service: Failed to load configuration: No such file or directory
Installation on armhfp and aarch64 single-board computers fails with zram-setup errors. It appears to be a conflict betwen udisks2-zram and zram-generator-defaults. Update FEDORA-2020-e143a283fb renamse the service unit file to make its origin more obvious. Removing udisks2-zram fixes the issue, but it's pulled in as a dependency of something, so that needs to be identified.
Action summary ============
Accepted blockers ----------------- 1. libreport — abrt-server errors when processing zstd compressed core dumps produced by systemd-246~rc1-1.fc33 — POST ACTION: abrt maintainers to make a new release containing the fix
2. resteasy — FreeIPA deployment fails in current Rawhide due to various issues with Java 11 — NEW ACTION: resteasy maintainers to diagnose issue and fix or pass the buck as appropriate
3. sddm — login stuck when changing users repeatedly (log out, log in a different one) — NEW ACTION: sddm maintainers to diagnose issue and fix or pass the buck as appropriate
4. selinux-policy — SELinux is preventing systemd-machine from 'create' accesses on the sock_file io.systemd.Machine. — POST ACTION: selinux-policy maintainers to make a new release containing the fixes
5. tomcat — FreeIPA server deployment fails in Fedora-Rawhide-20200714.n.0 — MODIFIED ACTION: tomcat maintainers to investigate failed upgrade test
6. distribution — Workstation live x86_64 image exceeds maximum size — NEW ACTION: Workstation WG to determine whether or not to remove packages or increase size limit.
7. distribution — Everything boot x86_64 image exceeds maximum size — ASSIGNED ACTION: bcotton to get someone to determine whether or not to remove packages or increase size limit.
8. distribution — Server boot x86_64 image exceeds maximum size — NEW ACTION: Server WG to determine whether or not to remove packages or increase size limit.
9. distribution — Release-blocking Fedora 33 images have Fedora 32 backgrounds — NEW ACTION: kde-settings maintainers to update package to use new backgrounds
Proposed blockers -----------------
1. systemd — systemd-resolved.service not work with DNS server placed behind VPN (openconnect) — NEW ACTION: systemd maintainers to diagnose issue
Bug-by-bug detail =============
Accepted blockers ----------------- 1. libreport — https://bugzilla.redhat.com/show_bug.cgi?id=1860616 — POST abrt-server errors when processing zstd compressed core dumps produced by systemd-246~rc1-1.fc33
abrt doesn't support core dumps compressed with zstd, which is now used by systemd. Merged upstream PR uses libarchive to add support for zstd (and other compression formats): https://github.com/abrt/libreport/pull/656.
2. resteasy — https://bugzilla.redhat.com/show_bug.cgi?id=1866570 — NEW FreeIPA deployment fails in current Rawhide due to various issues with Java 11
Deploying FreeIPA fails with a traceback in pki-tomcat. This appears to be a dependency chain issue that (currently) ends at resteasy via dogtag-pki. This may not be the last layer in the Java 11 onion.
3. sddm — https://bugzilla.redhat.com/show_bug.cgi?id=1861700 — NEW login stuck when changing users repeatedly (log out, log in a different one)
This is now a blocker based on a recent approval-in-principle of a release criterion related to logout and user switching. User processes linger after logout which blocks logging in when another user has logged in between the two sessions. Or when the second user logs back in. Or when a single user logs in repeatedly. Using KillUserProcesses=Yes helps the first problem, but raises on of its own. It appears to be a race condition of some kind as the behavior is not fully consistent.
4. selinux-policy — https://bugzilla.redhat.com/show_bug.cgi?id=1862686 — POST SELinux is preventing systemd-machine from 'create' accesses on the sock_file io.systemd.Machine.
librvirt VMs and systemd-machined service fail with AVC denials. Merged upstream PRs should fix the denials mentioned in this bug: https://github.com/fedora-selinux/selinux-policy/pull/405 and https://github.com/fedora-selinux/selinux-policy/pull/407
5. tomcat — https://bugzilla.redhat.com/show_bug.cgi?id=1857043 — MODIFIED FreeIPA server deployment fails in Fedora-Rawhide-20200714.n.0 due to pki-tomcat failing to run with "java.lang.ClassNotFoundException: org.apache.tomcat.util.modeler.modules.MbeansDescriptorsIntrospectionSource"
An upstream commit resulted in tomcat-coyote.jar missing packages. A previous fix (FEDORA-2020-f897a68801) moved the goalposts. tomcat-9.0.37-3 should fix this. Verification is blocked on BZ1866570, but Adam's most recent upgrade test still fails with a ClassNotFoundException for com.sun.xml.internal.bind.v2.ContextFactory.
6. distribtution — https://bugzilla.redhat.com/show_bug.cgi?id=1827915 — NEW Workstation live x86_64 image exceeds maximum size
Workstation WG is evaluating whether they can reduce the size and if not will bump the threshold.
7. distribtution — https://bugzilla.redhat.com/show_bug.cgi?id=1849430 — ASSIGNED Everything boot x86_64 image exceeds maximum size
Who owns this anyway? I'll try to get someone to figure out how we can save ~14 MB.
8. distribtution — https://bugzilla.redhat.com/show_bug.cgi?id=1849431 — ASSIGNED Server boot x86_64 image exceeds maximum size
We need to slim it down by ~14 MB.
9. distribution — https://bugzilla.redhat.com/show_bug.cgi?id=1869892 — ASSIGNED Release-blocking Fedora 33 images have Fedora 32 backgrounds
Backgrounds package has been created, but KDE Plasma needs to be explicitly updated to use it.
Proposed blockers -----------------
1. systemd — https://bugzilla.redhat.com/show_bug.cgi?id=1863041 — NEW systemd-resolved.service not work with DNS server placed behind VPN (openconnect)
When connected to an OpenConnect VPN, host name resolution fails. Manually disabling the systemd-resolved service works around this issue.
Action summary ====================
Accepted blockers ----------------- 1. libreport — abrt-server errors when processing zstd compressed core dumps produced by systemd-246~rc1-1.fc33 — POST ACTION:
2. sddm — login stuck when changing users repeatedly (log out, log in a different one) — NEW ACTION: sddm maintainers to diagnose issue and fix or pass the buck as appropriate NEEDINFO: mbriza, rdieter
3. distribution — Everything boot x86_64 image exceeds maximum size — ASSIGNED ACTION: none. Fixing Server boot should fix this, too
4. distribution — Server boot x86_64 image exceeds maximum size — NEW ACTION: Server WG to determine whether or not to remove packages or increase size limit.
5. distribution — Release-blocking Fedora 33 images have Fedora 32 backgrounds — NEW ACTION: kde-settings maintainers to update package to use new backgrounds (BZ 1872054)
6. pki-core — FreeIPA server upgrade from Fedora 32 or Fedora 31 fails with java.lang.UnsupportedClassVersionError — ASSIGNED ACTION: pki-core maintainers to package an update which includes the upstream PR
7. libpreport — bugs can't be reported: "No matching actions found for this event" — NEW ACTION: abrt maintainers to diagnose and fix issue
8. openconnect — systemd-resolved.service not work with DNS server placed behind VPN (openconnect) — ASSIGNED ACTION: openconnect maintainers to add integration for systemd-resolved. NEEDINFO: dwmw2
9. anaconda — Booting Server DVD then selecting a mirrorlist repository does not work — NEW ACTION: anaconda maintainers to diagnose and fix issue
Bug-by-bug detail =============
Accepted blockers ----------------- 1. libreport — https://bugzilla.redhat.com/show_bug.cgi?id=1860616 — POST abrt-server errors when processing zstd compressed core dumps produced by systemd-246~rc1-1.fc33
FEDORA-2020-59e144acee contains a potential fix, but introduces a new blocker (BZ 1873029). It may be moot until the retrace server is brought back online.
2. sddm — https://bugzilla.redhat.com/show_bug.cgi?id=1861700 — NEW login stuck when changing users repeatedly (log out, log in a different one)
User processes linger after logout which blocks logging in when another user has logged in between the two sessions. Or when the second user logs back in. Or when a single user logs in repeatedly. Using KillUserProcesses=Yes helps the first problem, but raises on of its own. It appears to be a race condition of some kind as the behavior is not fully consistent.
3. distribtution — https://bugzilla.redhat.com/show_bug.cgi?id=1849430 — ASSIGNED Everything boot x86_64 image exceeds maximum size
Latest compose does not significantly shrink the image size. Given the similar nature of the overages, I believe fixing the Server image size, which sgallagh is working on, will fix this as well.
4. distribtution — https://bugzilla.redhat.com/show_bug.cgi?id=1849431 — ASSIGNED Server boot x86_64 image exceeds maximum size
We need to slim it down by ~14 MB.
5. distribution — https://bugzilla.redhat.com/show_bug.cgi?id=1869892 — ASSIGNED Release-blocking Fedora 33 images have Fedora 32 backgrounds
Backgrounds package has been created, but KDE Plasma needs to be explicitly updated to use it. BZ 1872054 open for this.
6. pki-core — https://bugzilla.redhat.com/show_bug.cgi?id=1871990 — ASSIGNED FreeIPA server upgrade from Fedora 32 or Fedora 31 fails with java.lang.UnsupportedClassVersionError
/etc/sysconfig/tomcat.conf needs to be changed to point to JDK 11 on upgrade. Upstream PR contains a potential fix: https://github.com/dogtagpki/pki/pull/532
7. libreport — https://bugzilla.redhat.com/show_bug.cgi?id=1873029 — NEW bugs can't be reported: "No matching actions found for this event"
No crashes can be reported. abrt shows "no matching event" error message. This appears to be introduced by the fix for 1860616.
8. openconnect — https://bugzilla.redhat.com/show_bug.cgi?id=1863041 — ASSIGNED systemd-resolved.service not work with DNS server placed behind VPN (openconnect)
When connected to an OpenConnect VPN, host name resolution fails. Manually disabling the systemd-resolved service works around this issue. Opencoonect needs to push server information to systemd-resolved.
9. anaconda — https://bugzilla.redhat.com/show_bug.cgi?id=1872056 — NEW Booting Server DVD then selecting a mirrorlist repository does not work
The Server DVD openQA installation tests fail when using a mirrorlist URL. They do not fail with a direct repository URL.
On Mon, 2020-08-31 at 15:53 -0400, Ben Cotton wrote:
Accepted blockers
- libreport — https://bugzilla.redhat.com/show_bug.cgi?id=1860616 — POST
abrt-server errors when processing zstd compressed core dumps produced by systemd-246~rc1-1.fc33
FEDORA-2020-59e144acee contains a potential fix, but introduces a new blocker (BZ 1873029). It may be moot until the retrace server is brought back online.
I'll just note the same team is responsible for this whole complex - the retrace and FAF servers, and the abrt/libreport tool cluster - so ultimately the ask here of that team is: get the whole kaboodle working so we can actually report crashes in F33.
On Mon, Aug 31, 2020 at 02:35:29PM -0700, Adam Williamson wrote:
On Mon, 2020-08-31 at 15:53 -0400, Ben Cotton wrote:
Accepted blockers
- libreport — https://bugzilla.redhat.com/show_bug.cgi?id=1860616 — POST
abrt-server errors when processing zstd compressed core dumps produced by systemd-246~rc1-1.fc33
FEDORA-2020-59e144acee contains a potential fix, but introduces a new blocker (BZ 1873029). It may be moot until the retrace server is brought back online.
I'll just note the same team is responsible for this whole complex - the retrace and FAF servers, and the abrt/libreport tool cluster - so ultimately the ask here of that team is: get the whole kaboodle working so we can actually report crashes in F33.
Just to make sure folks know, the retrace server being down is not this teams fault, it's ours (infrastructure). We planned to just have it down for a week or less when moving it to RDU, but it turned out that datacenter move was not at all what we hoped for and it's been down much longer than intended.
That said, we have a machine up now there and it's just needing configuration/setup to get the service back up and running. :) So, hopefully soon it will again be online.
kevin
On Tue, 2020-09-01 at 10:29 -0700, kevin wrote:
On Mon, Aug 31, 2020 at 02:35:29PM -0700, Adam Williamson wrote:
On Mon, 2020-08-31 at 15:53 -0400, Ben Cotton wrote:
Accepted blockers
- libreport — https://bugzilla.redhat.com/show_bug.cgi?id=1860616 — POST
abrt-server errors when processing zstd compressed core dumps produced by systemd-246~rc1-1.fc33
FEDORA-2020-59e144acee contains a potential fix, but introduces a new blocker (BZ 1873029). It may be moot until the retrace server is brought back online.
I'll just note the same team is responsible for this whole complex - the retrace and FAF servers, and the abrt/libreport tool cluster - so ultimately the ask here of that team is: get the whole kaboodle working so we can actually report crashes in F33.
Just to make sure folks know, the retrace server being down is not this teams fault, it's ours (infrastructure). We planned to just have it down for a week or less when moving it to RDU, but it turned out that datacenter move was not at all what we hoped for and it's been down much longer than intended.
That said, we have a machine up now there and it's just needing configuration/setup to get the service back up and running. :) So, hopefully soon it will again be online.
Sorry about that, thanks for the correction.
I think for Beta at least it would be acceptable if we can submit reports via local backtrace generation - i.e. if the tools correctly handled the servers being down and fell back.
Dne 31. 08. 20 v 23:35 Adam Williamson napsal(a):
On Mon, 2020-08-31 at 15:53 -0400, Ben Cotton wrote:
Accepted blockers
- libreport — https://bugzilla.redhat.com/show_bug.cgi?id=1860616 — POST
abrt-server errors when processing zstd compressed core dumps produced by systemd-246~rc1-1.fc33
FEDORA-2020-59e144acee contains a potential fix, but introduces a new blocker (BZ 1873029). It may be moot until the retrace server is brought back online.
I'll just note the same team is responsible for this whole complex - the retrace and FAF servers, and the abrt/libreport tool cluster - so ultimately the ask here of that team is: get the whole kaboodle working so we can actually report crashes in F33.
The status of the HW: * it is racked in RDU2 * there has been issue in networking * with Smoodge (and bunch of others) help it get some temporary network setup, so we can ssh-access it now. * I hope that this week it will be configured and I hope that by next week it will be available under the original address.
The Go/No-Go meeting is Thursday!
Action summary ====================
Accepted blockers ----------------- 1. libreport — abrt-server errors when processing zstd compressed core dumps produced by systemd-246~rc1-1.fc33 — POST ACTION: msuchy to get retrace server back in service
2. sddm — login stuck when changing users repeatedly (log out, log in a different one) — NEW ACTION: sddm maintainers to diagnose issue and fix or pass the buck as appropriate NEEDINFO: mbriza, rdieter
3. distribution — Everything boot x86_64 image exceeds maximum size — ASSIGNED ACTION: none. Fixing Server boot should fix this, too
4. distribution — Server boot x86_64 image exceeds maximum size — NEW ACTION: Server WG to determine whether or not to remove packages or increase size limit.
5. distribution — Release-blocking Fedora 33 images have Fedora 32 backgrounds — NEW ACTION: kde-settings maintainers to update package to use new backgrounds (BZ 1872054)
6. pki-core — FreeIPA server upgrade from Fedora 32 or Fedora 31 fails with java.lang.UnsupportedClassVersionError — ASSIGNED ACTION: pki-core maintainers to package an update which includes the upstream PR
7. libpreport — bugs can't be reported: "No matching actions found for this event" — NEW ACTION: abrt maintainers to diagnose and fix issue
8. NetworkManager-openconnect — systemd-resolved.service not work with DNS server placed behind VPN (openconnect) — ASSIGNED ACTION: NetworkManager-openconnect maintainers to add integration for systemd-resolved.
9. anaconda — Booting Server DVD then selecting a mirrorlist repository does not work — MODIFIED ACTION: anaconda maintainers to package anaconda-33.25.3 update
10. kernel — 5.8.3-300.fc33.aarch64 kernel panic on boot (X-Gene PMU) — VERIFIED ACTION: none!
Proposed blockers -----------------
1. gnome-initial-setup — g-i-s fails to completely launch following clean install, opens in a tiny non-resizeable window instead — NEW ACTION: QA to test on bare metal to see if scope is limited to VMs while upstream hopefully gets us a fix
2. kernel — dasbus.error.DBusError: cannot initialize a disk that has partitions — NEW ACTION: kernel maintainers to diagnose issue, QA to try to find a reliable reproducer
Bug-by-bug detail =============
Accepted blockers ----------------- 1. libreport — https://bugzilla.redhat.com/show_bug.cgi?id=1860616 — POST abrt-server errors when processing zstd compressed core dumps produced by systemd-246~rc1-1.fc33
FEDORA-2020-59e144acee contains a potential fix, but appears to introduce a new blocker (BZ 1873029). It may be moot until the retrace server is brought back online. The infra team has provisioned a basic instance, which msuchy is working to get ready for use.
2. sddm — https://bugzilla.redhat.com/show_bug.cgi?id=1861700 — NEW login stuck when changing users repeatedly (log out, log in a different one)
User processes linger after logout which blocks logging in when another user has logged in between the two sessions. Or when the second user logs back in. Or when a single user logs in repeatedly. Using KillUserProcesses=Yes helps the first problem, but raises on of its own. It appears to be a race condition of some kind as the behavior is not fully consistent.
3. distribtution — https://bugzilla.redhat.com/show_bug.cgi?id=1849430 — ASSIGNED Everything boot x86_64 image exceeds maximum size
Latest compose does not significantly shrink the image size. Given the similar nature of the overages, I believe fixing the Server image size, which sgallagh is working on, will fix this as well.
4. distribtution — https://bugzilla.redhat.com/show_bug.cgi?id=1849431 — ASSIGNED Server boot x86_64 image exceeds maximum size
We need to slim it down by ~14 MB.
5. distribution — https://bugzilla.redhat.com/show_bug.cgi?id=1869892 — ASSIGNED Release-blocking Fedora 33 images have Fedora 32 backgrounds
Backgrounds package has been created, but KDE Plasma needs to be explicitly updated to use it. BZ 1872054 open for this.
6. pki-core — https://bugzilla.redhat.com/show_bug.cgi?id=1871990 — ASSIGNED FreeIPA server upgrade from Fedora 32 or Fedora 31 fails with java.lang.UnsupportedClassVersionError
/etc/sysconfig/tomcat.conf needs to be changed to point to JDK 11 on upgrade. Upstream PR contains a potential fix: https://github.com/dogtagpki/pki/pull/532
7. libreport — https://bugzilla.redhat.com/show_bug.cgi?id=1873029 — NEW bugs can't be reported: "No matching actions found for this event"
No crashes can be reported. abrt shows "no matching event" error message. This appears to be introduced by the fix for 1860616.
8. NetworkManager-openconnect — https://bugzilla.redhat.com/show_bug.cgi?id=1863041 — NEW systemd-resolved.service not work with DNS server placed behind VPN (openconnect)
When connected to an OpenConnect VPN, host name resolution fails. Openconnect needs to push server information to systemd-resolved. Upstream vpnc-script added support for systemd-resolved in 2016, so it's not clear what is missing. This may need to be reassigned to the vpnc-script component.
9. anaconda — https://bugzilla.redhat.com/show_bug.cgi?id=1872056 — MODIFIED Booting Server DVD then selecting a mirrorlist repository does not work
The Server DVD openQA installation tests fail when using a mirrorlist URL. They do not fail with a direct repository URL. An upstream fix has been merged and will be in anaconda-33.25.3.
10. kernel — https://bugzilla.redhat.com/show_bug.cgi?id=1874117) — VERIFIED 5.8.3-300.fc33.aarch64 kernel panic on boot (X-Gene PMU)
Uninitialized variable existed xgene PMU driver. This bug has been squashed (and patch sent upstream).
Proposed blockers -----------------
1. gnome-initial-setup — https://bugzilla.redhat.com/show_bug.cgi?id=1875140 — NEW g-i-s fails to completely launch following clean install, opens in a tiny non-resizeable window instead
Some fraction of installs end up with gnome-initial-setup in a tiny window. So far, this has only been observed in vritual machines and is functional when resized to a usable size. Issue filed upstream: https://gitlab.gnome.org/GNOME/gnome-initial-setup/-/issues/110
2. kernel — https://bugzilla.redhat.com/show_bug.cgi?id=1828188 — NEW dasbus.error.DBusError: cannot initialize a disk that has partitions
It is possible that partitions are not discovered by kernel after boot (meaning no nodes in /dev and no records in sysfs) so anaconda doesn't add partitions to devicetree during reset. But after running some parted checks the partitions magically appears, that's why device.format.partitions works. Reproducibility currently unclear. It may be related to mdraid.
Fedora 33 Beta was no-go by default due to outstanding blockers. Let's try again this week!
Action summary ====================
Accepted blockers ----------------- 1. libreport — abrt-server errors when processing zstd compressed core dumps produced by systemd-246~rc1-1.fc33 — ON_QA ACTION: testers needed to see if BZ 1873029 is reproducible
2. dbus-broker — login stuck when changing users repeatedly (log out, log in a different one) — NEW ACTION: Someone (anyone!) figure out what causes this behavior.
3. distribution — Everything boot x86_64 image exceeds maximum size — ASSIGNED ACTION: none. Fixing Server boot should fix this, too
4. distribution — Server boot x86_64 image exceeds maximum size — NEW ACTION: Server WG to determine whether or not to remove packages or increase size limit.
5. libpreport — bugs can't be reported: "No matching actions found for this event" — NEW ACTION: abrt maintainers to diagnose and fix issue
6. anaconda — Booting Server DVD then selecting a mirrorlist repository does not work — VERIFIED ACTION: None!
7. systemd — resolv.conf misconfigured on fresh install — ASSIGNED ACTION: systemd maintainers to add a removal of existing /etc/resolv.conf when not an upgrade
Bug-by-bug detail =============
Accepted blockers ----------------- 1. libreport — https://bugzilla.redhat.com/show_bug.cgi?id=1860616 — ON_QA abrt-server errors when processing zstd compressed core dumps produced by systemd-246~rc1-1.fc33
FEDORA-2020-59e144acee contains a potential fix, but appears to introduce a new blocker (BZ 1873029). It may be moot until the retrace server is brought back online. The infra team has provisioned a basic instance, which msuchy is working to get ready for use.
2. dbus-broker — https://bugzilla.redhat.com/show_bug.cgi?id=1861700 — NEW login stuck when changing users repeatedly (log out, log in a different one)
User processes linger after logout which blocks logging in when another user has logged in between the two sessions. Or when the second user logs back in. Or when a single user logs in repeatedly. Using KillUserProcesses=Yes helps the first problem, but raises on of its own. It appears to be a race condition of some kind as the behavior is not fully consistent. Nobody really knows what the problem is.
3. distribtution — https://bugzilla.redhat.com/show_bug.cgi?id=1849430 — ASSIGNED Everything boot x86_64 image exceeds maximum size
Latest compose does not significantly shrink the image size. Given the similar nature of the overages, I believe fixing the Server image size, which sgallagh is working on, will fix this as well.
4. distribtution — https://bugzilla.redhat.com/show_bug.cgi?id=1849431 — ASSIGNED Server boot x86_64 image exceeds maximum size
We need to slim it down by ~14 MB.
5. libreport — https://bugzilla.redhat.com/show_bug.cgi?id=1873029 — NEW bugs can't be reported: "No matching actions found for this event"
No crashes can be reported. abrt shows "no matching event" error message. This appears to be introduced by the fix for 1860616.
6. anaconda — https://bugzilla.redhat.com/show_bug.cgi?id=1872056 — VERIFIED Booting Server DVD then selecting a mirrorlist repository does not work
The Server DVD openQA installation tests fail when using a mirrorlist URL. They do not fail with a direct repository URL. FEDORA-2020-daac40f3d3 fixes this.
7. systemd — https://bugzilla.redhat.com/show_bug.cgi?id=1873856 — ASSIGNED resolv.conf misconfigured on fresh install
/etc/resolv.conf is a file, not a symlink, which causes DNS problems in some cases. FEDORA-2020-2255b438a2 fixed it in Worskation, but the behavior apparently still exists in the netinstall. Removing the existing file created by NetworkManager when not an upgrade should fix this.
On Fri, 2020-09-11 at 15:50 -0400, Ben Cotton wrote:
Accepted blockers
- libreport — https://bugzilla.redhat.com/show_bug.cgi?id=1860616 — ON_QA
abrt-server errors when processing zstd compressed core dumps produced by systemd-246~rc1-1.fc33
FEDORA-2020-59e144acee contains a potential fix, but appears to introduce a new blocker (BZ 1873029). It may be moot until the retrace server is brought back online. The infra team has provisioned a basic instance, which msuchy is working to get ready for use.
So yeah: it would really help if other folks could try installing the update and see if they are able to successfully file a crash bug or not. Please report your findings to the bug report(s).
Reporting bugs *is* supposed to work even if using the retrace server doesn't. In that case abrt should fall back on installing debuginfo locally and generating the backtrace locally (after confirming with the user that this is OK). If that doesn't happen currently, that also could be considered a bug.
My basic thought here is that for Beta release it would be acceptable if we can file crash bugs successfully, even if the retrace server isn't up. Obviously the faster we can get the retrace server up the better.
- distribtution — https://bugzilla.redhat.com/show_bug.cgi?id=1849430
— ASSIGNED Everything boot x86_64 image exceeds maximum size
Latest compose does not significantly shrink the image size. Given the similar nature of the overages, I believe fixing the Server image size, which sgallagh is working on, will fix this as well.
- distribtution — https://bugzilla.redhat.com/show_bug.cgi?id=1849431
— ASSIGNED Server boot x86_64 image exceeds maximum size
We need to slim it down by ~14 MB.
I've been working on this. One PR was merged to Rawhide already: https://github.com/weldr/lorax/pull/1070 but not yet to the F33 branch. I have an equivalent for F33 pending: https://github.com/weldr/lorax/pull/1073 Plus a couple of further changes currently proposed for Rawhide but which I will also submit for F33: https://github.com/weldr/lorax/pull/1075 https://github.com/weldr/lorax/pull/1076 I have one more to go on top of 1075 when/if it's merged. If we take all these changes, the images should be 20-30MB or so under the limit. We can also take any combination of whichever people think are the safest to try and get under the limit.
We're getting close! I'm going to make Adam request an RC next week whether he likes it or not. :-)
Action summary ====================
Accepted blockers ----------------- 1. libreport — abrt-server errors when processing zstd compressed core dumps produced by systemd-246~rc1-1.fc33 — VERIFIED ACTION: none
2. dbus-broker — login stuck when changing users repeatedly (log out, log in a different one) — ON_QA ACTION: QA to test if FEDORA-2020-5b486e191b improves the behavior
3. libpreport — bugs can't be reported: "No matching actions found for this event" — VERIFIED ACTION: none
4. abrt — Can't report a crash (even with local processing) due to "Could not resolve host: retrace.fedoraproject.org" — NEW ACTION: abrt maintainers to diagnose and fix issue ACTION: msuchy to finish bringing retrace server back online
Proposed blockers -----------------
1. anaconda — a metalink repository can't be added using UI — ON_QA ACTION: QA to verify FEDORA-2020-e1d834c1ef fixes the issue
2. NetworkManager — systemd-resolved.service not work with DNS server placed behind VPN (openconnect) — NEW ACTION: NetworkManager maintainers to determine future behavior of default routes when VPN server pushes some routes
3. pki-core — Running ipa-server-install fails in step [1/30]: configuring certificate server instance — ON_QA ACTION: reporter to verify that this is happening in F33
Bug-by-bug detail =============
Accepted blockers ----------------- 1. libreport — https://bugzilla.redhat.com/show_bug.cgi?id=1860616 — VERIFIED abrt-server errors when processing zstd compressed core dumps produced by systemd-246~rc1-1.fc33
This appears to be fixed with FEDORA-2020-444a3363f0, which causes BZ 1878317. That may block getting this fix in.
2. dbus-broker — https://bugzilla.redhat.com/show_bug.cgi?id=1861700 — ON_QA login stuck when changing users repeatedly (log out, log in a different one)
User processes linger after logout which blocks logging in when another user has logged in between the two sessions. Or when the second user logs back in. Or when a single user logs in repeatedly. An update to kf5-baloo (FEDORA-2020-5b486e191b) address part of the issue, but it's likely that this is a many-fix problem.
3. libreport — https://bugzilla.redhat.com/show_bug.cgi?id=1873029 — VERIFIED bugs can't be reported: "No matching actions found for this event"
This appears to be fixed with FEDORA-2020-444a3363f0, which causes BZ 1878317. That may block getting this fix in.
4. abrt https://bugzilla.redhat.com/show_bug.cgi?id=1878317 — NEW Can't report a crash (even with local processing) due to "Could not resolve host: retrace.fedoraproject.org"
Since the retrace server is still offline, abrt should fall back to local processing. It does not. As an unprivileged user, skipping the report_uReport step results in chmod errors and behavior like BZ 1873029. As root, it core dumps. We're in general agreement that if either online or local processing works, that's good enough for beta. Ticket for retrace server is https://pagure.io/fedora-infrastructure/issue/9060
Proposed blockers -----------------
1. anaconda — https://bugzilla.redhat.com/show_bug.cgi?id=1879127 — ON_QA a metalink repository can't be added using UI
Adding a metalink repository to the installer incorrectly modifies the URL. FEDORA-2020-e1d834c1ef contains a potential fix (Adam has a ISO at https://www.happyassassin.net/temp/00669088-FEDORA-2020-e1d834c1ef-netinst-x...). This bug is accepted as a freeze exception even if it doesn't get enough votes for blocker status.
2. NetworkManager — https://bugzilla.redhat.com/show_bug.cgi?id=1863041 — NEW systemd-resolved.service not work with DNS server placed behind VPN (openconnect)
The scope of this bug has narrowed to the point where a revote indicates it's probably no longer a blocker. We may consider using only the static "Use this connection..." property instead of that *and* pushed routes to determine if the VPN is full-tunnel. Currently if the server pushed any routes, the openconnect plugin tells NetworkManager not to add a default route.
3. pki-core — https://bugzilla.redhat.com/show_bug.cgi?id=1878616 — ON_QA Running ipa-server-install fails in step [1/30]: configuring certificate server instance
Installation fails in Rawhide (only?). The package updates that appear to be causing the problem never made it into F33.
On Fri, 2020-09-18 at 16:10 -0400, Ben Cotton wrote:
Can't report a crash (even with local processing) due to "Could not resolve host: retrace.fedoraproject.org"
Since the retrace server is still offline, abrt should fall back to local processing. It does not.
This is not quite an accurate description. If you get to the actual backtrace generation action, fallback from the remote server to local processing *does* work. The problem is that report_uReport tries to run *before you ever reach that stage*, and report_uReport can only work if retrace.fedoraproject.org is up and accepting reports. There's no fallback for report_uReport.
Arguably, it failing should be non-fatal, but it seems like it's fatal.
As an unprivileged user, skipping the report_uReport step results in chmod errors and behavior like BZ 1873029. As root, it core dumps.
For me, anyway. Would be good if others could confirm.
Beta is out! Time to focus on Final blockers.
Action summary ====================
Accepted blockers ----------------- 1. firefox — Firefox not using langpacks for localization — ASSIGNED ACTION: firefox maintainers to fix issue
2. sddm — login stuck when changing users repeatedly (log out, log in a different one) — ASSIGNED ACTION: kf5-kglobalaccel maintainers to provide systemd service file
3. gnome-control-center — Select Printer Driver hangs — POST ACTION: none (waiting for coming upstream release)
Proposed blockers -----------------
1. mesa — Massive memory leak on AMD cards — NEW ACTION: reporter to test with older kernel packages
2. cheese — cheese creates invalid and extremely long video files, each app restart making them longer — NEW ACTION: cheese maintainers to diagnose and fix issue ACTION: everyone to watch kparal's puppet show
3. distribution-gpg-keys — gnome-software 3.38.0 does not list all software in Add-ons — NEW ACTION: distribution-gpg-keys maintainer to add rpmfusion 33 keys
4. gnome-remote-desktop — Can't login if the session is locked — NEW ACTION: none
5. shim — Secure Boot fails to boot F33 Beta image — NEW NEEDINFO: aathomas
Bug-by-bug detail =============
Accepted blockers ----------------- 1. firefox — https://bugzilla.redhat.com/show_bug.cgi?id=1816547 — ASSIGNED Firefox not using langpacks for localization
Behavior was previously fixed as an F32 blocker, but it has returned. Regardless of the default locale, Firefox launches in English.
2. sddm — https://bugzilla.redhat.com/show_bug.cgi?id=1861700 — ASSIGNED login stuck when changing users repeatedly (log out, log in a different one)
Tricky bug that may need F34 to fix fully. Short term suggestion is to provide service files of type=dbus. Request for kf5-kglobalaccel filed as BZ 1884696.
3. gnome-control-center — https://bugzilla.redhat.com/show_bug.cgi?id=1868141 — POST Select Printer Driver hangs
Fixed in upstream 3-38 branch. 3.38.1 release in the next few days should contain this.
Proposed blockers -----------------
1. mesa — https://bugzilla.redhat.com/show_bug.cgi?id=1880833 — NEW Massive memory leak on AMD cards
Athlon 200ge, integrated Radeon Vega 3 GPU, using amdgpu driver runs out of memory overnight. It is unclear if this is limited to certain card models or if it is in mesa, the kernel or somewhere else.
2. cheese — https://bugzilla.redhat.com/show_bug.cgi?id=1884260 — NEW cheese creates invalid and extremely long video files, each app restart making them longer
cheese seems to insert empty length before subsequent video captures. See Kamil's excellent puppet show in comment 3 as an example.
3. distribution-gpg-keys https://bugzilla.redhat.com/show_bug.cgi?id=1882863 NEW gnome-software 3.38.0 does not list all software in Add-ons
NVIDIA rpmfusion repo does not display in GNOME Software, but does in dnf. It appears to be because the gpg key for the rpmfusion F33 repo is not in the distribution-gpg-keys package.
4. gnome-remote-desktop — https://bugzilla.redhat.com/show_bug.cgi?id=1882718 — NEW Can't login if the session is locked
When the main session is locked, a VNC connection cannot login. An accompanying crash has been fixed upstream, but overall the behavior is intended for security purposes. An upstream issue is open to provide the desired behavior: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/3212 (note that the behavior is new in PipeWire screen casts. org.gnome.Shell.Screencast previously behaved this way)
5. shim — https://bugzilla.redhat.com/show_bug.cgi?id=1883609 — NEW Secure Boot fails to boot F33 Beta image
Trying to boot F33 Beta on some machines with Secure Boot on results in an access denied error. Disabling Secure Boot doesn't boot. Booting previous releases and then rebooting into 33 seems to work. Note that this is also being discussed in a FESCo ticket https://pagure.io/fesco/issue/2479
The Go/No-Go meeting is Thursday. Here we go!
Action summary ====================
Accepted blockers ----------------- 1. sddm — login stuck when changing users repeatedly (log out, log in a different one) — ASSIGNED ACTION: kf5-kglobalaccel maintainers to provide systemd service file NEEDINFO: rdieter
2. cheese — cheese creates invalid and extremely long video files, each app restart making them longer — ASSIGNED ACTION: cheese maintainers to diagnose and fix issue ACTION: everyone to watch kparal's puppet show
3. libvncserver — Server crashes after one connection attempt, all subsequent attempts will fail — NEW ACTION: libvncserver maintainers to package upstream fix
4. kernel — 5.8.3-300.fc33.aarch64 kernel panic on boot (X-Gene PMU) — ON_QA ACTION: QA to verify FEDORA-2020-9664e2f1d2
5. shim — Secure Boot fails to boot F33 Beta image — NEW ACTION: Do whatever pjones says we need to do
Proposed blockers -----------------
1. freeipa — FreeIPA server upgrade from F32 to F33 doesn't work any more because F32's FreeIPA is newer — MODIFIED ACTION: Releng to push FEDORA-2020-e9e815177e stable if it's not replaced by a fixier version
2. gnome-calendar — gnome-calendar: on_calendar_monitor_completed_cb(): gnome-calendar killed by SIGABRT — NEW ACTION: gnome-calendar maintainers to diagnose and fix issue
3. gnome-online-accounts — Google account: Credentials have expired — ASSIGNED ACTION: gnome-online-accounts maintainers to diagnose and fix issue NEEDINFO: chrismurphy
4. gnome-software — Gnome software unable to rate review — NEW ACTION: gnome-software maintainers to diagnose and fix issue
5. uboot-tools uboot-tools-2020.10 is available ON_QA ACTION: QA to verify FEDORA-2020-8de85c0b4f
Bug-by-bug detail =============
Accepted blockers ----------------- 1. sddm — https://bugzilla.redhat.com/show_bug.cgi?id=1861700 — ASSIGNED login stuck when changing users repeatedly (log out, log in a different one)
Tricky bug that may need F34 to fix fully. Short term suggestion is to provide service files of type=dbus. Request for kf5-kglobalaccel filed as BZ 1884696.
2. cheese — https://bugzilla.redhat.com/show_bug.cgi?id=1884260 — NEW cheese creates invalid and extremely long video files, each app restart making them longer
cheese seems to insert empty length before subsequent video captures. See Kamil's excellent puppet show in comment 3 as an example.
3. libvncserver — https://bugzilla.redhat.com/show_bug.cgi?id=1882718 — NEW Server crashes after one connection attempt, all subsequent attempts will fail
An inital VNC session succeeds, but the subsquent attempt crashes. There is an upstream fix available: https://github.com/LibVNC/libvncserver/pull/444
4. kernel — https://bugzilla.redhat.com/show_bug.cgi?id=1874117 — ON_QA 5.8.3-300.fc33.aarch64 kernel panic on boot (X-Gene PMU)
5. shim https://bugzilla.redhat.com/show_bug.cgi?id=1883609 NEW Secure Boot fails to boot F33 Beta image
It appears that Ubuntu has pushed a Secure Boot revocation list early, so people who have previously installed Ubtunu will not be able to boot Fedora. This is accepted as a FESCo blocker.
Proposed blockers -----------------
1. freeipa — https://bugzilla.redhat.com/show_bug.cgi?id=1886205 — MODIFIED FreeIPA server upgrade from F32 to F33 doesn't work any more because F32's FreeIPA is newer
What it says on the tin. Update FEDORA-2020-e9e815177e is in updates-testing but does not fix systemd-resolved integration as it's intended to. We may need to push the update and let a future update fix the systemd-resolved integration.
2. gnome-calendar — https://bugzilla.redhat.com/show_bug.cgi?id=1883681 — NEW gnome-calendar: on_calendar_monitor_completed_cb(): gnome-calendar killed by SIGABRT
The second launch of gnome-calendar fails. This may or may not be related to an issue with a potential upstream fix: https://gitlab.gnome.org/GNOME/gnome-calendar/-/merge_requests/149
3. gnome-online-accounts — https://bugzilla.redhat.com/show_bug.cgi?id=1885479 — ASSIGNED Google account: Credentials have expired
Immediately after authenticating with 2FA enabled, Google account shows that credentials have expired. It appears to be related to the token not being written to gnome-keyring.
4. gnome-software — https://bugzilla.redhat.com/show_bug.cgi?id=1886406 — NEW Gnome software unable to rate review
It is possible to add a review for an application, but not to rate it (useful/not useful). This appears to be headed for RejectedBlocker status as it does not fall under "basic functionality".
5. uboot-tools https://bugzilla.redhat.com/show_bug.cgi?id=1885316 ON_QA uboot-tools-2020.10 is available
This is a bump from the RC currently in Fedora 33. Historically, we get a lot of questions when we ship with an RC of this component on the media.
With luck, this is the last time I'll send this until Fedora 34 Beta! The Go/No-Go is Thursday.
Action summary ====================
Accepted blockers ----------------- 1. sddm — login stuck when changing users repeatedly (log out, log in a different one) — ASSIGNED ACTION: QA to test FEDORA-2020-f5be904258
2. shim — Secure Boot fails to boot F33 Beta image — NEW ACTION: (none. waiting for new shim to be signed)
Proposed blockers -----------------
1. bluez — bluetooth mouse does not autoconnect after sleep — NEW ACTION: bluez maintainers to diagnose and fix issue
Bug-by-bug detail =============
Accepted blockers ----------------- 1. sddm — https://bugzilla.redhat.com/show_bug.cgi?id=1861700 — ASSIGNED login stuck when changing users repeatedly (log out, log in a different one)
Tricky bug that may need F34 to fix fully. Short term suggestion is to provide service files of type=dbus. rdieter has implemented a workaround to restart dbus.service after an 8-second wait. This is update FEDORA-2020-f5be904258.
2. shim https://bugzilla.redhat.com/show_bug.cgi?id=1883609 NEW Secure Boot fails to boot F33 Beta image
It appears that Ubuntu has pushed a Secure Boot revocation list early, so people who have previously installed Ubtunu will not be able to boot Fedora. This is accepted as a FESCo blocker. A new shim was intended for submission with an expectation of getting it signed in "not more than a couple of days".
Proposed blockers -----------------
1. bluez — https://bugzilla.redhat.com/show_bug.cgi?id=1874082 — NEW bluetooth mouse does not autoconnect after sleep
Some Bluetooth devices do not reconnect after resuming from sleep without manual intervention. From the discussion on the test mailing list, it's not universal, but it does not appear to be limited to a few devices, either. Update FEDORA-2020-00bf5ac290 contained a possible fix, but did not change the behavior.