| Summary: | sys-boot/os-prober - WARNING: Failed to connect to lvmetad. | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | moshe2go |
| Component: | Current packages | Assignee: | No maintainer - Look at https://wiki.gentoo.org/wiki/Project:Proxy_Maintainers if you want to take care of it <maintainer-needed> |
| Status: | RESOLVED TEST-REQUEST | ||
| Severity: | normal | CC: | bes.internal, floppym, jcallen, jstein |
| Priority: | Normal | ||
| Version: | unspecified | ||
| Hardware: | AMD64 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
|
Description
moshe2go
2016-02-14 09:45:54 UTC
*** Bug 574676 has been marked as a duplicate of this bug. *** I updated from sys-fs/lvm2-2.02.116 to -r4 and have the same warning with `grub-mkconfig`. I have it with `rc-update -u` too, so it's more general: # rc-update -u * Caching service dependencies ... File descriptor 3 (pipe:[5196152]) leaked on lvm invocation. Parent PID 17753: /bin/sh /run/lvm/lvmetad.socket: connect failed: No such file or directory WARNING: Failed to connect to lvmetad. Falling back to internal scanning. I'm not using LVM directly. cryptsetup depends on it through Portage (for my LUKS partitions), but from what I understand, it's not even starting lvm/lvmetad initscripts, which also are not in my `rc-update show`. I have not enabled LVM in my genkernel.conf file either. I have not tested a reboot yet, to check if it's just an harmless warning... An elog message in lvm2 ebuild mentions restarting udev, but it changes nothing. If I downgrade back to lvm2-2.02-116, there is no more warning message with `rc-update -u`. ... but the warning message is still here with `grub-mkconfig -o /boot/grub/grub.cfg`... I did update from sys-boot/grub-2.02_beta2-r8 to -r9 too at the same time as lvm2... ... but if I downgrade back to -r8, I still have the warning... I'm pretty sure I ran `grub-mkconfig` without this warning a few months back though... Maybe I was still using -r7 at that time, but it's not in Portage anymore to easily test downgrading... I had lvm2-2.02.109 at that time too... [ebuild R] sys-fs/lvm2-2.02.116-r4::gentoo USE="-device-mapper-only -lvm1 -lvm2create_initrd readline -static static-libs -systemd thin udev" [ebuild R] sys-boot/grub-2.02_beta2-r9:2/2.02_beta2-r9::gentoo [2.02_beta2-r8:2/2.02_beta2-r8::gentoo] USE="-debug -device-mapper doc -efiemu fonts -mount -multislot nls -sdl -static themes truetype" GRUB_PLATFORMS="efi-64" That warning is coming from os-prober, which gets called by /etc/grub.d/30_os-prober. (In reply to Kanito Tasoga from comment #2) > I updated from sys-fs/lvm2-2.02.116 to -r4 and have the same warning with > `grub-mkconfig`. > > > I have it with `rc-update -u` too, so it's more general: > > # rc-update -u > * Caching service dependencies ... > File descriptor 3 (pipe:[5196152]) leaked on lvm invocation. Parent PID > 17753: /bin/sh > /run/lvm/lvmetad.socket: connect failed: No such file or directory > WARNING: Failed to connect to lvmetad. Falling back to internal scanning. > On the next boot the problem will go away. rc-update -u just updates the dependencies for the NEXT boot. It doesn't do anything with your current state. Your LVM setup is currently wrong and you need a clean boot. I think you just need to change use_lvmetad to 0 in /etc/lvm/lvm.conf to make this warning go away. This is lvm related bug. In clean lvm setup init.d/lvmetad don't create /run/lvm/ dir for /run/lvm/lvmetad.socket. `mkdir /run/lvm/` solved the problem We have to fix it in init.d/lvmetad or send to upstream? (In reply to Vladimir Varlamov from comment #6) Sounds like a separate bug report to me. What is the status with 1.71? |
