I did found a workaround, but iam still convinced that we have a kind of bug. I think i should explain our typical systemsetup, for better understandig.
Typical field of application are failoversystems. overall we use very few software and systems have minimal load. we have 2 servers in a cluster realised with pacemaker/corosync. they manage a reource haproxy and a floating ip. We do this with different ubuntu OS, from 18.04 over 20.04 to 22.04 Our systems are bound to Windows Active Director with SSSD (System Security Services Daemon) (https://schroeffu.ch/2019/09/linux-active- directory-ldap-ssh-login-mit-sssd-und-realmd/) so it is possible to Login with our AD Credentials. last component is altiris server management suite agent (former symantec now broadcom) wich is running with root privileges and helps to manage our computerlandscape. And this is where i located the problem. every evening the agent runs a bash script which was wrote by me 3 years ago. it is a small script with 90 lines, it collects some data, mounts a windows fileshare and finally uploads some small files before unmounting the share. nothing special, it takes around 5 seconds to complete, but here seems to be the problem. As i can see every affected server shows in syslog this lines about kernel bug i uploaded on 2023-09-11 (#4) . In some cases there happens something unexpected and triggers this bug. this happens since 5.15.0-83-generic the system gets unstable, high load without running processes, every command takes forever to complete. mostly we had to do a vmware hardstop, because even "reboot -f" failed. i uploaded already some logs. I deactivated this job and the problems disappeared. i was not able to trigger this problem by manual run of the script. as the job was active, every morning we had a bunch of servers in this state between life and death. So i can not confirm a change on our site, i still think about a newly introduced kind of bug. I would like to hear from you, please tell me your opinion to this case. strange that i report a bug with documented kernel error and no one gets back to me. Thanks, Hajo -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux in Ubuntu. https://bugs.launchpad.net/bugs/2034701 Title: unexpected system behaviour after kernel update Status in linux package in Ubuntu: Confirmed Bug description: Hello, since last kernel updates on ubuntu 20.04 and 22.04 we see lines like this in syslog after boot: Sep 7 13:10:53 myhost kernel: [ 1.202296] pci 0000:00:15.4: BAR 13: no space for [io size 0x1000] Sep 7 13:10:53 myhost kernel: [ 1.202387] pci 0000:00:15.4: BAR 13: failed to assign [io size 0x1000] We see this on Ubuntu 20.04 after installing kernel 5.4.0-162-generic and on ubuntu 22.04 after installing kernel 5.15.0-83-generic Since this time we notice some strange server behaviour. - unexpected reboots during workhours - systems hanging/freezing which requires a hardstop of the machine - systems with high load but no significant number of processes, even on non productive machines with usually zero load. We use most of our ubuntu servers as virtual machines in a VMWare Environment with VSphere and ESXi 7.0.3 There were no VMWare updates. I can say this with certainty because iam the responsible person for VMWare too and i did not install any ESXi Updates last days. Other OS not affected as i can see. Thank you, Hans p.s. i wanted to help and wanted to choose fitting package from dropdown " In what package did you find this bug?" above this Inputfield, but it seems that packages like linux-image-5.15.0-83-generic (or similiar i tried somethging...) are unknown and search respondes with improper suggestions. To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2034701/+subscriptions -- Mailing list: https://launchpad.net/~kernel-packages Post to : kernel-packages@lists.launchpad.net Unsubscribe : https://launchpad.net/~kernel-packages More help : https://help.launchpad.net/ListHelp
