Re: fsck vs zvol

2019-06-11 Thread O'Connor, Daniel 



> On 12 Jun 2019, at 14:40, Eugene Grosbein  wrote:
> 
> 12.06.2019 8:04, O'Connor, Daniel wrote:
> 
>> Hi,
>> I have a small UFS partition that is the sysvol for Samba 4 (otherwise it 
>> doesn't work due to ACL issues).
>> 
>> I found that I usually have to manually fsck it on a bad reboot, even if I 
>> have fsck_y_enable so I added a hack to /etc/rc.d/fsck to fsck -y that FS 
>> before the normal fsck runs.
>> 
>> Logging looks like..
>> 
>> Jun 12 09:45:41 moo kernel: Setting hostuuid: 
>> ----ac1f6b01103a.
>> Jun 12 09:45:41 moo kernel: Setting hostid: 0x6cbfec5d.
>> Jun 12 09:45:41 moo kernel: Starting file system checks:
>> Jun 12 09:45:41 moo kernel: Mounting local filesystems:mount: 
>> /dev/zvol/zroot/samba4sysvol: R/W mount of /var/db/samba4/sysvol denied. 
>> Filesystem is not clean - run fsck.: Operation not permitted
>> Jun 12 09:45:41 moo kernel: Mounting /etc/fstab filesystems failed, will 
>> retry after root mount hold release
>> Jun 12 09:45:41 moo kernel: mount: /dev/zvol/zroot/samba4sysvol: R/W mount 
>> of /var/db/samba4/sysvol denied. Filesystem is not clean - run fsck.: 
>> Operation not permitted
>> Jun 12 09:45:41 moo kernel: .
>> Jun 12 09:45:41 moo kernel: Mounting /etc/fstab filesystems failed, startup 
>> aborted
>> Jun 12 09:45:41 moo kernel: ERROR: ABORTING BOOT (sending SIGTERM to parent)!
>> 
>> I'm not sure if marking this FS as 'late' would help (and/or break something 
>> else..)
>> 
>> Has anyone else tried something like this?
>> 
>> The Samba 4 ZFS is from https://wiki.freebsd.org/Samba4ZFS (which I wrote..)
> 
> Please show your /etc/fstab line for this UFS-inside-ZVOL and your changes to 
> rc.d/fsck.
> Your logs do not show that fsck is started so I presume some mistake in the 
> /etc/fstab.
> Maybe you forgot that it needs non-zero sixth field.

The fsck line is..
/dev/zvol/zroot/samba4sysvol /var/db/samba4/sysvol ufs   rw,acls 0   0

The diff is just..
--- fsck.orig   2019-06-12 14:43:03.279407000 +0930
+++ fsck2019-06-12 14:42:51.094142000 +0930
@@ -24,6 +24,10 @@
# During fsck ignore SIGQUIT
trap : 3

+   echo "Samba FS hack"
+   fsck -y /dev/zvol/zroot/samba4sysvol
+   err=$?
+
check_startmsgs && echo "Starting file system checks:"
if checkyesno background_fsck; then
fsck -F -p

Oh I see for the passno field.. It must be non-zero it fsck won't check it at 
all!

Today I Learnt... :)

Thanks Eugene!

--
Daniel O'Connor
"The nice thing about standards is that there
are so many of them to choose from."
 -- Andrew Tanenbaum


___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: fsck vs zvol

2019-06-11 Thread Eugene Grosbein 
12.06.2019 8:04, O'Connor, Daniel wrote:

> Hi,
> I have a small UFS partition that is the sysvol for Samba 4 (otherwise it 
> doesn't work due to ACL issues).
> 
> I found that I usually have to manually fsck it on a bad reboot, even if I 
> have fsck_y_enable so I added a hack to /etc/rc.d/fsck to fsck -y that FS 
> before the normal fsck runs.
> 
> Logging looks like..
> 
> Jun 12 09:45:41 moo kernel: Setting hostuuid: 
> ----ac1f6b01103a.
> Jun 12 09:45:41 moo kernel: Setting hostid: 0x6cbfec5d.
> Jun 12 09:45:41 moo kernel: Starting file system checks:
> Jun 12 09:45:41 moo kernel: Mounting local filesystems:mount: 
> /dev/zvol/zroot/samba4sysvol: R/W mount of /var/db/samba4/sysvol denied. 
> Filesystem is not clean - run fsck.: Operation not permitted
> Jun 12 09:45:41 moo kernel: Mounting /etc/fstab filesystems failed, will 
> retry after root mount hold release
> Jun 12 09:45:41 moo kernel: mount: /dev/zvol/zroot/samba4sysvol: R/W mount of 
> /var/db/samba4/sysvol denied. Filesystem is not clean - run fsck.: Operation 
> not permitted
> Jun 12 09:45:41 moo kernel: .
> Jun 12 09:45:41 moo kernel: Mounting /etc/fstab filesystems failed, startup 
> aborted
> Jun 12 09:45:41 moo kernel: ERROR: ABORTING BOOT (sending SIGTERM to parent)!
> 
> I'm not sure if marking this FS as 'late' would help (and/or break something 
> else..)
> 
> Has anyone else tried something like this?
> 
> The Samba 4 ZFS is from https://wiki.freebsd.org/Samba4ZFS (which I wrote..)

Please show your /etc/fstab line for this UFS-inside-ZVOL and your changes to 
rc.d/fsck.
Your logs do not show that fsck is started so I presume some mistake in the 
/etc/fstab.
Maybe you forgot that it needs non-zero sixth field.



___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

fsck vs zvol

2019-06-11 Thread O'Connor, Daniel 
Hi,
I have a small UFS partition that is the sysvol for Samba 4 (otherwise it 
doesn't work due to ACL issues).

I found that I usually have to manually fsck it on a bad reboot, even if I have 
fsck_y_enable so I added a hack to /etc/rc.d/fsck to fsck -y that FS before the 
normal fsck runs.

Logging looks like..

Jun 12 09:45:41 moo kernel: Setting hostuuid: 
----ac1f6b01103a.
Jun 12 09:45:41 moo kernel: Setting hostid: 0x6cbfec5d.
Jun 12 09:45:41 moo kernel: Starting file system checks:
Jun 12 09:45:41 moo kernel: Mounting local filesystems:mount: 
/dev/zvol/zroot/samba4sysvol: R/W mount of /var/db/samba4/sysvol denied. 
Filesystem is not clean - run fsck.: Operation not permitted
Jun 12 09:45:41 moo kernel: Mounting /etc/fstab filesystems failed, will retry 
after root mount hold release
Jun 12 09:45:41 moo kernel: mount: /dev/zvol/zroot/samba4sysvol: R/W mount of 
/var/db/samba4/sysvol denied. Filesystem is not clean - run fsck.: Operation 
not permitted
Jun 12 09:45:41 moo kernel: .
Jun 12 09:45:41 moo kernel: Mounting /etc/fstab filesystems failed, startup 
aborted
Jun 12 09:45:41 moo kernel: ERROR: ABORTING BOOT (sending SIGTERM to parent)!

I'm not sure if marking this FS as 'late' would help (and/or break something 
else..)

Has anyone else tried something like this?

The Samba 4 ZFS is from https://wiki.freebsd.org/Samba4ZFS (which I wrote..)

Thanks.

--
Daniel O'Connor
"The nice thing about standards is that there
are so many of them to choose from."
 -- Andrew Tanenbaum


___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

RE: Update from r347952 to r348848

2019-06-11 Thread driesm.michiels 
Yep I haven’t tested with the new revision of the source tree but indeed I was 
using an outdated package which didn’t get deleted and replaced by the new name 
scheme.

I’m using Synth and did not add graphics/drm-kmod to the list to build as the 
module is required to be rebuild with every kernel build anyway.

This begs me to question does PORTS_MODULES= also update the port in question 
when a kernel rebuild is done?

 

Maybe as a general question: what is the best way to maintain kernel modules 
from ports on a system by using a package builder like poudriere or synth?

 

Thanks for helping out !

 

From: Johannes Lundberg  
Sent: maandag 10 juni 2019 22:09
To: Dries Michiels ; freebsd-stable@freebsd.org
Subject: Re: Update from r347952 to r348848

 

drm-next-kmod has been replaced by several packages. Not sure why 'pkg install 
drm-kmod' doesn't replace it with the new one. 

Try 'pkg delete drm-kmod drm-next-kmod && pkg install drm-kmod'

You should get drm-fbsd12.0-kmod-4.16.g20190430.

 

On 6/10/19 1:05 PM, Dries Michiels wrote:

pkg info | grep drm

drm-kmod-g20181126 Metaport of DRM modules for the linuxkpi-based 
KMS components

drm-next-kmod-4.11.g20181027   DRM modules for the linuxkpi-based KMS components

libdrm-2.4.96,1Userspace interface to kernel Direct Rendering 
Module services

Running 12-STABLE on my system.

Get Outlook for Android  

 

  _  

From: Johannes Lundberg   
Sent: Monday, June 10, 2019 9:59:36 PM
To: driesm.michi...@gmail.com  ; 
freebsd-stable@freebsd.org  
Subject: Re: Update from r347952 to r348848 

 

drm-kmod is a meta port. Show me 'pkg info | grep drm' to see what is
actually installed.

On 6/10/19 12:33 PM, driesm.michi...@gmail.com 
  wrote:
>
>> From: Johannes Lundberg   
>>
>> Hi
>>
>> What graphics driver do you have installed?
> I have drm-kmod installed and rebuild/reinstall every time I install my 
> kernel by setting PORTS_MODULES = graphics/drm-kmod in make.conf
> I have an Intel(R) Xeon(R) CPU E3-1245 v5 @ 3.50GHz with skylake graphics so 
> using the i915 kms driver.
>
>> This error should not happen if you have the latest drm-kmod package.
>> Try 'pkg update && pkg install drm-kmod'.
>>
>> On 6/10/19 5:26 AM, driesm.michi...@gmail.com 
>>   wrote:
>>> Hi stable mailing list!
>>>
>>>
>>>
>>> Today I tried updating my machine to r348848 from r347952. It didn't
>>> go as smooth as I would've hoped (normally its just fine).
>>>
>>> After doing the upgrade using the standard upgrade from source
>>> procedure I went ahead and booted my machine.
>>>
>>> During booting I got black screened (after the messages below) and
>>> booting got stuck from there.
>>>
>>> *   Loading kernel modules:
>>> *   Anon_inodefs_init:
>>> *   Sysctl_warn_reuse: can't re-use a leaf (compat.linuxkpi.debug)!
>>> *   Black screen and booting stops
>>>
>>>
>>>
>>> I went ahead and reverted to a ZFS snapshot preupdate so all is fine again.
>>>
>>> Any pointers as to where the regression might be? I did a clean build
>>> to update.
>>>
>>>
>>>
>>> Regards,
>>>
>>> Dries
>>>
>>>
>>>
>>> ___
>>> freebsd-stable@freebsd.org   mailing list
>>> https://lists.freebsd.org/mailman/listinfo/freebsd-stable
>>> To unsubscribe, send any mail to "freebsd-stable-
>> unsubscr...@freebsd.org  "

___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: Kernel panic on 12-STABLE-r348203 amd64

2019-06-11 Thread Mark Saad 
All
  I am going to try to reset the box to factory defaults and try to make it 
crash again today . I’ll update you with my outcome . 

---
Mark Saad | nones...@longcount.org

> On Jun 11, 2019, at 1:15 AM, Kubilay Kocak  wrote:
> 
>> On 6/06/2019 5:04 am, Mark Saad wrote:
>>> On Wed, Jun 5, 2019 at 2:42 PM Mark Saad  wrote:
>>> 
 On Wed, Jun 5, 2019 at 12:29 PM Mark Saad  wrote:
 
 All
  I was wondering if anyone could shed some light on this boot panic I
 saw yesterday. This is on a Dell R630 with Bios 2.9.1  booting
 12.0-STABLE-r348203 amd64.
 I reverted this back to 12.0-RELEASE-p4 and its fine .
 
 The only custom options I had were in loader.conf
 
 kern.geom.label.gptid.enable="0"
 ipmi_load="YES"
 boot_multicons="YES"
 boot_serial="YES"
 console="comconsole,vidconsole"
 net.inet.tcp.tso="0"
 cc_htcp_load="YES"
 autoboot_delay="5"
 hw.mfi.mrsas_enable="1"
 hw.usb.no_pf="1"# Disable USB packet filtering
 hw.usb.no_shutdown_wait="1"
 hw.vga.textmode="1" # Text mode
 machdep.hyperthreading_allowed="0"
 
 Any ideas ?
 
 Screen shot here
 https://imgur.com/a/nGvHtIs
 
 --
 mark saad | nones...@longcount.org
>>> 
>>> Plain text version of the crash
>>> 
>>> Loading kernel...
>>> /boot/kernel/kernel text=0x168d811 data=0x1cf968+0x768c80
>>> syms=[0x8+0x1778e8+0x8   /
>>> +0x194f1d]
>>> Loading configured modules...
>>> /boot/kernel/ipmi.ko size 0x11e10 at 0x2645000
>>> loading required module 'smbus'
>>> /boot/kernel/smbus.ko size 0x2ef0 at 0x2657000
>>> /boot/entropy size=0x1000
>>> /boot/kernel/cc_httcp.ko size 0x2330 at 0x265b000
>>> ---<>---c_hmodule 'smbus'
>>> Copyright (c) 1992-2019 The FreeBSD Project.
>>> Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
>>> The Regents of the University of California. All rights reserved.
>>> FreeBSD is a registered trademark of The FreeBSD Foundation.
>>> FreeBSD 12.0-STABLE r348693 GENERIC amd64
>>> FreeBSD clang version 8.0.0 (tags/RELEASE_800/final 356365) (based on
>>> LLVM 8.0.0)
>>> panic: UMA zone "UMA Zones": Increase vm.boot_pages
>>> cpuid = 0
>>> time = 1
>>> KDB: stack backtrace:
>>> #0 0x80c16df7 at ??+0
>>> #1 0x80bcaccd at ??+0
>>> #2 0x80bcab23 at ??+0
>>> #3 0x80f0b03c at ??+0
>>> #4 0x80f08d8d at ??+0
>>> #5 0x80f0bb3d at ??+0
>>> #6 0x80f0b301 at ??+0
>>> #7 0x80f0b3d1 at ??+0
>>> #8 0x80f066c4 at ??+0
>>> #9 0x80f0543f at ??+0
>>> #10 0x80f23aef at ??+0
>>> #11 0x80f1133b at ??+0
>>> #12 0x80b619c8 at ??+0
>>> #13 0x8036a02c at ??+0
>>> Uptime: 1s
>>> 
>>> 
>>> Also increasing the vm.boot_pages to 128 in the loader works. Anyone
>>> know why ? This box has 64G ram.
>>> 
>>> --
>>> mark saad | nones...@longcount.org
>> So after some poking in the bios this has to do with how the Dell NUMA
>> options are set. If the system is set Cluster On Die mode, you get a
>> kernel panic
>> Home Snoop or Early Snoop no issue.
> 
> Hi Mark,
> 
> Could you report this bug (Bugzilla) if you haven't already, providing:
> 
> - exact freebsd version(s) reproducible with
> - panic/backtrace output as an attachment. Ideally with a debug kernel
> - /var/run/dmesg.boot output (as an attachment) in a verbose boot
> - if you can test a current snapshot, that would be great
> - any other system information you believe might be helpful in isolating root 
> cause(s) or potential fixes
> 
> Thanks!
> Feel free to CC me on it
___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"