During boot of an UML with 0.12.3 (x86 or ~x86) I do get sometimes : * Wiping /tmp directory ... [ ok ] * Setting up dm-crypt mappings ... * crypt-swap using: -c aes -h sha1 -d /dev/urandom create crypt-swap /dev/ubdb ... [ ok ] * pre_mount: mkswap ${dev} ... [ ok ] * Setting hostname to trinity ... [ ok ] /lib/rc/sh/rc-cgroup.sh: line 87: /sys/fs/cgroup/openrc/loopback/tasks: No such file or directory * Bringing up network interface lo ... [ ok ]
FWIW it is an arbitrarily happen issue, happens with different system, not always, eg today I got : * Bringing up network interface lo ... [ ok ] /lib/rc/sh/rc-cgroup.sh: line 87: /sys/fs/cgroup/openrc/net.lo/tasks: No such file or directory * Bringing up interface lo Probably related to UML, I dunno.
It's known race in cgroups, I'm going to address this issue on one of the following weekends. The problem is that issue is not reproducible on my systems.
(In reply to Alexander Vershilov from comment #2) > It's known race in cgroups, I'm going to address this issue on one of the > following weekends. The problem is that issue is not reproducible on my > systems. Ah - thx. Don't hesitate to contact me at irc or here via b.g.o - I do have 2 UML images to play/test with it.
FWIW here is probably a dup if this : https://bugs.gentoo.org/show_bug.cgi?id=485638 and again - something I can help to address/solve this issue ?
I can confirm this with sys-apps/openrc-0.12.4 running on a Alix Board. (This is a slow system, maybe thats a hint?) --- [ ok ] * Remounting filesystems ... [ ok ] * Updating /etc/mtab ... [ ok ] * Activating swap devices ... [ ok ] * Mounting local filesystems ... [ ok ] * Configuring kernel parameters ... [ ok ] * Creating user login records ... [ ok ] * Wiping /tmp directory ... [ ok ] /lib/rc/sh/rc-cgroup.sh: line 87: /sys/fs/cgroup/openrc/hostname/tasks: No such file or directory * Setting hostname to alix ... [ ok ] * Bringing up network interface lo ... [ ok ] /lib/rc/sh/rc-cgroup.sh: line 87: echo: write error: No such device * Bringing up interface lo * 127.0.0.1/8 ... [ ok ] * Adding routes * 127.0.0.0/8 via 127.0.0.1 ... [ ok ] * Activating additional swap space ... [ ok ] * setting up tmpfiles.d entries ... [ ok ] * Initializing random number generator ... [ ok ] INIT: Entering runlevel: 3 * Starting metalog ... [ ok ] /lib/rc/sh/rc-cgroup.sh: line 87: /sys/fs/cgroup/openrc/net.br0/tasks: No such file or directory * Bringing up interface br0 * Creating bridge br0 ... * Adding ports to br0 * lan0 ... [ ok ] * lan1 ... [ ok ] * lan2 ... [ ok ] * 192.168.0.254 ... [ ok ] * Bringing up interface lan0 [ ok ] * Bringing up interface lan1 [ ok ] * Bringing up interface lan2 [ ok ] ---
Set rc_controller_cgroups="NO" in /etc/rc.conf as in #485638 does not solve it.
(In reply to slick from comment #5) > (This is a slow system, maybe thats a hint?) Hhm, often I do start the UML guest with "nice ..."
I do observe this even with 0.12.4 under a KVM machine (x86 Gentoo guest) with current kernel 3.16.1 and 3.17-rc1.
(In reply to Toralf Förster from comment #8) > I do observe this even with 0.12.4 under a KVM machine (x86 Gentoo guest) > with current kernel 3.16.1 and 3.17-rc1. Is it still happening with 0.13?
(In reply to William Hubbs from comment #9) > (In reply to Toralf Förster from comment #8) > > I do observe this even with 0.12.4 under a KVM machine (x86 Gentoo guest) > > with current kernel 3.16.1 and 3.17-rc1. > > Is it still happening with 0.13? did not observed this till now with 0.13, rebooted few times both my user mode linux systems and teh KVM's
*** Bug 485638 has been marked as a duplicate of this bug. ***
(In reply to William Hubbs from comment #9) > (In reply to Toralf Förster from comment #8) > > I do observe this even with 0.12.4 under a KVM machine (x86 Gentoo guest) > > with current kernel 3.16.1 and 3.17-rc1. > > Is it still happening with 0.13? oh - bad news, happened on a x86 user mdoe linux w/ unstable openrc-0.13.1 : * Starting DHCP Client Daemon ... [ ok ] * Starting haveged ... [ ok ] * Starting sshd ... [ ok ] /lib/rc/sh/rc-cgroup.sh: line 87: echo: write error: No such device * Starting local ... [ ok ]
openrc-0.13.9 on a desktop Gentoo machine... ... [ ok ] /lib64/rc/sh/rc-cgroup.sh: line 88: /sys/fs/cgroup/openrc/ntp-client/tasks: No such file or directory * Setting clock via the NTP client 'ntpdate' ... ... [ ok ] /lib64/rc/sh/rc-cgroup.sh: line 88: /sys/fs/cgroup/openrc/apache2/tasks: No such file or directory * Starting apache2 ... ...
I have been unable to reproduce this issue with any services on my system. Also, there was a cgroups fix applied to OpenRC-0.18.3 which took care of several cgroups-related issues. Can you test with OpenRC-0.18.3 and re-open this bug if this still occurs?
I'm seeing this issue with openrc-0.20.5 on gentoo-sources-4.4.2, in regular hardware (not a VM) both at startup and shutdown with several random services each time.
(In reply to mg05182-gentoo from comment #15) > I'm seeing this issue with openrc-0.20.5 on gentoo-sources-4.4.2, in regular > hardware (not a VM) both at startup and shutdown with several random > services each time. me to, regular Desktop, openrc-1.19.1, vanilla kernel 4.4.7 and sys-fs/udev 225
I am also having this issue on my new install. I used stage3-amd64-20160623.tar.bz2 with kernel linux-4.4.6-gentoo
This is on a desktop, by the way. A real PC....no VM.
Interesting, if I boot up with kernel 3.14.14-gentoo with the same root fs and openRc, this is not a issue.
repoening per comment 15
Can I just let it complain about the missing directory during boot up without having any side effects or breaking anything?
All, is this ever happening for a service that starts a daemon, or is it just for one-off tasks?
Reproduced in regular machine using Archlinux [see https://bbs.archlinux.org/viewtopic.php?pid=1650059]
