Fedora Account System
Red Hat Associate
Red Hat Customer
Since we switched to webui for the ostree installer images I've noticed intermittent openQA failures where the install seems to just get stuck at "Writing objects". I just hit this in a local VM install as well, but on further investigation, the logs seem to indicate the install actually completed. For instance, packaging.log shows "Thread Done: AnaTaskThread-ConfigureBootloader-1" then "DEBUG:anaconda.modules.common.base.base:Generating kickstart...", which are also the last two lines in that log on a complete install. So I think the install is actually completing, but the UI is getting stuck somehow. Perhaps a progress signal is getting missed or something?
AGREED AcceptedBetaFreezeException Discussed at the 2026-08-24 (blocker / freeze exception) review meeting: This is accepted as it's obviously a major installer issue that can't be remedied with an update. For now we only know for sure it affects non-blocking images, we will consider it as a blocker if it turns out to affect IoT. https://meetbot-raw.fedoraproject.org//blocker-review_matrix_fedoraproject-org/2026-08-24/f45-blocker-review.2026-08-24-15.59.log.txt
FEDORA-2026-5ac781d0d4 (anaconda-webui-82-1.fc46) has been submitted as an update to Fedora 46. https://bodhi.fedoraproject.org/updates/FEDORA-2026-5ac781d0d4
FEDORA-2026-5ac781d0d4 (anaconda-webui-82-1.fc46) has been pushed to the Fedora 46 stable repository. If problem still persists, please make note of it in this bug report.
I don't think closing the bug was intended. The changes in the build are only listed as "Relevant" to this bug, not as "Resolves".
I can still reproduce this from an ISO with https://github.com/rhinstaller/anaconda-webui/pull/1434 and https://github.com/rhinstaller/anaconda/pull/7270 applied. Attaching anaconda-webui.log and dbus.log. Yell if anything else is needed.
Created attachment 2155848 [details] anaconda-webui.log
oh, dbus.log is tiny and boring: dbus-daemon[2246]: SELinux support is enabled dbus-daemon[2246]: Activating service name='org.fedoraproject.Anaconda.Boss' requested by ':1.0' (uid=0 pid=2175 comm="/usr/bin/python3 /usr/bin/anaconda" label="system_u:system_r:kernel_t:s0") dbus-daemon[2246]: Successfully activated service 'org.fedoraproject.Anaconda.Boss' dbus-daemon[2246]: Activating service name='org.fedoraproject.Anaconda.Modules.Runtime' requested by ':1.1' (uid=0 pid=2250 comm="python3 -m pyanaconda.modules.boss" label="system_u:system_r:kernel_t:s0") dbus-daemon[2246]: Activating service name='org.fedoraproject.Anaconda.Modules.Users' requested by ':1.1' (uid=0 pid=2250 comm="python3 -m pyanaconda.modules.boss" label="system_u:system_r:kernel_t:s0") dbus-daemon[2246]: Activating service name='org.fedoraproject.Anaconda.Modules.Payloads' requested by ':1.1' (uid=0 pid=2250 comm="python3 -m pyanaconda.modules.boss" label="system_u:system_r:kernel_t:s0") dbus-daemon[2246]: Activating service name='org.fedoraproject.Anaconda.Modules.Storage' requested by ':1.1' (uid=0 pid=2250 comm="python3 -m pyanaconda.modules.boss" label="system_u:system_r:kernel_t:s0") dbus-daemon[2246]: Activating service name='org.fedoraproject.Anaconda.Modules.Localization' requested by ':1.1' (uid=0 pid=2250 comm="python3 -m pyanaconda.modules.boss" label="system_u:system_r:kernel_t:s0") dbus-daemon[2246]: Activating service name='org.fedoraproject.Anaconda.Modules.Timezone' requested by ':1.1' (uid=0 pid=2250 comm="python3 -m pyanaconda.modules.boss" label="system_u:system_r:kernel_t:s0") dbus-daemon[2246]: Activating service name='org.fedoraproject.Anaconda.Modules.Network' requested by ':1.1' (uid=0 pid=2250 comm="python3 -m pyanaconda.modules.boss" label="system_u:system_r:kernel_t:s0") dbus-daemon[2246]: Successfully activated service 'org.fedoraproject.Anaconda.Modules.Users' dbus-daemon[2246]: Successfully activated service 'org.fedoraproject.Anaconda.Modules.Runtime' dbus-daemon[2246]: Successfully activated service 'org.fedoraproject.Anaconda.Modules.Payloads' dbus-daemon[2246]: Successfully activated service 'org.fedoraproject.Anaconda.Modules.Network' dbus-daemon[2246]: Successfully activated service 'org.fedoraproject.Anaconda.Modules.Timezone' dbus-daemon[2246]: Successfully activated service 'org.fedoraproject.Anaconda.Modules.Localization' dbus-daemon[2246]: Successfully activated service 'org.fedoraproject.Anaconda.Modules.Storage'
Created attachment 2155850 [details] journal.log Notable extracts: Aug 27 19:27:38 fedora systemd-resolved[1917]: Under memory pressure, flushing caches. Aug 27 19:27:38 fedora systemd-resolved[1917]: Flushed all caches. Aug 27 20:40:38 fedora org.fedoraproject.Anaconda.Modules.Payloads[2274]: INFO:anaconda.core.threads:Thread Done: AnaTaskThread-DeployOSTreeTask-1 (140044201498304) Aug 27 20:40:38 fedora org.fedoraproject.Anaconda.Modules.Payloads[2274]: INFO:anaconda.core.threads:Running Thread: AnaTaskThread-SetSystemRootTask-1 (140044201498304) Aug 27 20:40:39 fedora org.fedoraproject.Anaconda.Modules.Payloads[2274]: INFO:anaconda.core.threads:Thread Done: AnaTaskThread-SetSystemRootTask-1 (140044201498304) Aug 27 20:40:39 fedora org.fedoraproject.Anaconda.Modules.Payloads[2274]: INFO:anaconda.core.threads:Running Thread: AnaTaskThread-CopyBootloaderDataTask-1 (140044201498304) ...and so on - you can see the installer proceeding through further stages, but webui.log just seems to give up somewhere around 19:27:39. Is it possible the memory pressure that causes resolved to flush its caches at 19:27:38 is also involved somehow? I can't see any other related log messages, though. I don't see any OOM kills or anything.
Adam Kankovsky has a potential fix for this, but can't reproduce it. openQA is still hitting this fairly frequently so I can run it through an openQA test cannon today and check it that way.
FEDORA-2026-8576419da0 (anaconda-webui-83-1.fc46) has been submitted as an update to Fedora 46. https://bodhi.fedoraproject.org/updates/FEDORA-2026-8576419da0
FEDORA-2026-83e09bc1a6 (anaconda-webui-83-1.fc45) has been submitted as an update to Fedora 45. https://bodhi.fedoraproject.org/updates/FEDORA-2026-83e09bc1a6
FEDORA-2026-8576419da0 (anaconda-webui-83-1.fc46) has been pushed to the Fedora 46 stable repository. If problem still persists, please make note of it in this bug report.
Re-opening as this is filed against F45, and we need to verify the fix.
FEDORA-2026-83e09bc1a6 has been pushed to the Fedora 45 testing repository. Soon you'll be able to install the update with the following command: `sudo dnf upgrade --enablerepo=updates-testing --refresh --advisory=FEDORA-2026-83e09bc1a6` You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2026-83e09bc1a6 See also https://fedoraproject.org/wiki/QA:Updates_Testing for more information on how to test updates.
And I have been attempting 3rd time with different Silverblue 45 iso's for i18n testday event and it keep stuck at "Writing objects" step and then blaming for my newly installed Fedora CSB system for not supporting Silverblue installation!
FYI, I tried Fedora-Silverblue-Installer-45-20260906.n.0.x86_64.iso, Fedora-Silverblue-Installer-45-20260907.n.0.x86_64.iso and Fedora-Silverblue-Installer-45_Beta-1.3.x86_64.iso
Parag https://bodhi.fedoraproject.org/updates/FEDORA-2026-83e09bc1a6 is still not pushed to stable - so it's expected that you can't use the fix yes in the ISOs listed above.
(In reply to Katerina Koukiou from comment #17) > Parag https://bodhi.fedoraproject.org/updates/FEDORA-2026-83e09bc1a6 is > still not pushed to stable - so it's expected that you can't use the fix yes > in the ISOs listed above. I see but I thought at least Fedora-Silverblue-Installer-45_Beta-1.3.x86_64.iso contains anaconda-webui-83-1.fc46 as per https://forge.fedoraproject.org/releng/tickets/issues/13517#issuecomment-1353778 and I can also see its included in https://kojipkgs.fedoraproject.org/compose/45/Fedora-45-20260905.0/compose/Everything/x86_64/os/Packages/a/
You are right. To unblock you try increasing the resources of the test VM. I was not able to reproduce with 4vcpus / 6GiB VM. If you would like to help with testing this bug, can you check if indeed the installation had finished for your case (check Journal for 'Installation successfully finished.') and it's just the UI stuck?
Created attachment 2157288 [details] journalctl command output1
Created attachment 2157289 [details] journalctl command output2
(In reply to Katerina Koukiou from comment #19) > You are right. To unblock you try increasing the resources of the test VM. I > was not able to reproduce with 4vcpus / 6GiB VM. If you would like to help > with testing this bug, can you check if indeed the installation had finished > for your case (check Journal for 'Installation successfully finished.') and > it's just the UI stuck? Its still stuck at "Writing objects" screen. Attached about some snapshot of journalctl command. 1st one where last anaconda commands logged. 2nd one is the last lines of journalctl command. Installation is not completed I think from the logs?
Katerina, Interesting it just finished. This was my 4th different F45 iso installation but now with more resources. Previous installation I wait like 2 hrs and nothing happened. I reboot already and system boot fine. I am curious to find this message "Installation successfully finished" on installed system logs now. I tried to look for that message in /var/log/anaconda/journal.log but could not find such message. So now memory requirement has been increased/changed for VM installation of Silverblue 45? Thanks for your help.
I tested Fedora 45 Beta 1.3 Silverblue in virt-manager with 2 vCPUs and 4 GiB of RAM. The installation appeared to get stuck at “Writing objects”, but I confirmed that the installation actually completed successfully. I will also test again with 16 vCPUs and 32 GiB of RAM to see if the available CPU/memory resources make any difference, and I’ll report the results here.
I also tested with 16 vCPUs and 32 GiB of RAM. The installation completed successfully in just 1 minute.
(In reply to Efraim Kaov from comment #24) > I tested Fedora 45 Beta 1.3 Silverblue in virt-manager with 2 vCPUs and 4 > GiB of RAM. > > The installation appeared to get stuck at “Writing objects”, but I confirmed > that the installation actually completed successfully. Does this 2 vCPU and 4 GiB RAM installation worked for you? If yes how did you verified that installation completed successfully with UI stuck? Currently I am running this installation, its over 1 hour now still installation not completed and showing "Writing objects". > > I will also test again with 16 vCPUs and 32 GiB of RAM to see if the > available CPU/memory resources make any difference, and I’ll report the > results here.
(In reply to Efraim Kaov from comment #24) > I tested Fedora 45 Beta 1.3 Silverblue in virt-manager with 2 vCPUs and 4 > GiB of RAM. > > The installation appeared to get stuck at “Writing objects”, but I confirmed > that the installation actually completed successfully. > Just now its 2 hrs still installation is not finished. UI stuck at "Writing objects". Checked journalctl output nothing there for last 1 hr log. I don't think with 2 vCPU and 4 GiB RAM installation is working in KVM.