Today, on attempting resuming after suspending my system before boarding my plane. My system hung.
So I powered cycled, and since I have not done an update in over a week, instead of logging into my graphical window, I switched to Term2, logged in as root and ran "dnf update".
I have done this a lot of times when I hung like this. Don't log into graphical window and then have all the apps start and still should do an update.
So ~87 updates and a new kernel. Big deal. But it hangs on a scriptlet that looked like selinux-policy.
So I switch to Term3, log in as myself and run top. I see restorecon eating up 96% cpu. I wait 30min. Still not done.
I go to my Win10 system and google Fedora and restorecon eating up time. Found bug 1832327 on F35! The poster killed restorecon, so I did too.
Obviously update finished. I rebooted and logged in.
Is there something I should do/check since I killed it?
thanks
On Fri, May 2, 2025 at 1:16 PM Robert Moskowitz rgm@htt-consult.com wrote:
Today, on attempting resuming after suspending my system before boarding my plane. My system hung.
So I powered cycled, and since I have not done an update in over a week, instead of logging into my graphical window, I switched to Term2, logged in as root and ran "dnf update".
I have done this a lot of times when I hung like this. Don't log into graphical window and then have all the apps start and still should do an update.
So ~87 updates and a new kernel. Big deal. But it hangs on a scriptlet that looked like selinux-policy.
On recent updates here, the selinux update took quite a long time with no signs of life -- patience was rewarded with a completed update.
So I switch to Term3, log in as myself and run top. I see restorecon eating up 96% cpu. I wait 30min. Still not done.
I go to my Win10 system and google Fedora and restorecon eating up time. Found bug 1832327 on F35! The poster killed restorecon, so I did too.
Obviously update finished. I rebooted and logged in.
Is there something I should do/check since I killed it?
"restorecon" has a "-n" (passive check) option
On 5/3/25 10:05 AM, George N. White III wrote:
On Fri, May 2, 2025 at 1:16 PM Robert Moskowitz rgm@htt-consult.com wrote:
Today, on attempting resuming after suspending my system before boarding my plane. My system hung. So I powered cycled, and since I have not done an update in over a week, instead of logging into my graphical window, I switched to Term2, logged in as root and ran "dnf update". I have done this a lot of times when I hung like this. Don't log into graphical window and then have all the apps start and still should do an update. So ~87 updates and a new kernel. Big deal. But it hangs on a scriptlet that looked like selinux-policy.On recent updates here, the selinux update took quite a long time with no signs of life -- patience was rewarded with a completed update.
I felt 45min was a bit much of waiting for Gedot.
So I switch to Term3, log in as myself and run top. I see restorecon eating up 96% cpu. I wait 30min. Still not done. I go to my Win10 system and google Fedora and restorecon eating up time. Found bug 1832327 on F35! The poster killed restorecon, so I did too. Obviously update finished. I rebooted and logged in. Is there something I should do/check since I killed it?"restorecon" has a "-n" (passive check) option
Just tried that and system came back immediately. I checked /var/log/messages and nothing reported there from running it.
On 5 May 2025, at 10:41, Robert Moskowitz rgm@htt-consult.com wrote:
Just tried that and system came back immediately.
See the man page. Without a path to check the command does nothing. I am not sure what is an appropriate command line to use. Any one else know?
Barry
On Mon, 2025-05-05 at 17:23 +0100, Barry wrote:
On 5 May 2025, at 10:41, Robert Moskowitz rgm@htt-consult.com wrote:
Just tried that and system came back immediately.
See the man page. Without a path to check the command does nothing. I am not sure what is an appropriate command line to use. Any one else know?
Without some context I can't tell what either of you are talking about. Is there a gap in the thread?
poc
On Mon, May 5, 2025 at 6:15 PM Patrick O'Callaghan pocallaghan@gmail.com wrote:
On Mon, 2025-05-05 at 17:23 +0100, Barry wrote:
On 5 May 2025, at 10:41, Robert Moskowitz rgm@htt-consult.com wrote:
Just tried that and system came back immediately.
See the man page. Without a path to check the command does nothing. I am not sure what is an appropriate command line to use. Any one else know?
Without some context I can't tell what either of you are talking about. Is there a gap in the thread?
I suggested using "restorecon -r" to verify that killing the process didn't leave some bad contexts. Robert ran it without providing the required pathname and Barry suggested reading the man page.
There is a small example at < https://www.oreilly.com/library/view/centos-quick-start/9781789344875/cc8fda... that may help
On Tue, 2025-05-06 at 08:07 -0300, George N. White III wrote:
On Mon, May 5, 2025 at 6:15 PM Patrick O'Callaghan pocallaghan@gmail.com wrote:
On Mon, 2025-05-05 at 17:23 +0100, Barry wrote:
On 5 May 2025, at 10:41, Robert Moskowitz rgm@htt-consult.com wrote:
Just tried that and system came back immediately.
See the man page. Without a path to check the command does nothing. I am not sure what is an appropriate command line to use. Any one else know?
Without some context I can't tell what either of you are talking about. Is there a gap in the thread?
I suggested using "restorecon -r" to verify that killing the process didn't leave some bad contexts. Robert ran it without providing the required pathname and Barry suggested reading the man page.
There is a small example at < https://www.oreilly.com/library/view/centos-quick-start/9781789344875/cc8fda... that may help
Thanks.
poc
I am pretty sure I saw one of the rpm installs doing a long running restorecon.
I am pretty sure I saw this as the command: "restorecon -r /" (from ps) and that command from the command line takes a while to run.
On Tue, May 6, 2025 at 7:15 AM Patrick O'Callaghan pocallaghan@gmail.com wrote:
On Tue, 2025-05-06 at 08:07 -0300, George N. White III wrote:
On Mon, May 5, 2025 at 6:15 PM Patrick O'Callaghan pocallaghan@gmail.com wrote:
On Mon, 2025-05-05 at 17:23 +0100, Barry wrote:
On 5 May 2025, at 10:41, Robert Moskowitz rgm@htt-consult.com wrote:
Just tried that and system came back immediately.
See the man page. Without a path to check the command does nothing. I am not sure what is an appropriate command line to use. Any one else know?
Without some context I can't tell what either of you are talking about. Is there a gap in the thread?
I suggested using "restorecon -r" to verify that killing the process didn't leave some bad contexts. Robert ran it without providing the required pathname and Barry suggested reading the man page.
There is a small example at < https://www.oreilly.com/library/view/centos-quick-start/9781789344875/cc8fda... that may help
Thanks.
poc
users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-leave@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
On 05/06/2025 12:45 PM, Roger Heflin wrote:
I am pretty sure I saw this as the command: "restorecon -r /" (from ps) and that command from the command line takes a while to run.
I would expect it to run for quite some time. Reading the man page tells me that this command runs in / and continues recursively, meaning that it runs through your entire file system. HTH, HAND.
-
Barry -
George N. White III -
Joe Zeff -
Patrick O'Callaghan -
Robert Moskowitz -
Roger Heflin