Re: geom_label, fstab without device names & swap partition?

[Eugene Grosbein]

On 13.01.2011 02:11, Mark Saad wrote:

 Use glabel(8) to label the device:

 # glabel label swap ada0s1b
>>>
>>> On a side note there is not a simple way to glabel mounted filesystem
>>
>> True, but for a swap partition, it's easy to disable swap, label the
>> partition, and re-enable swap.  :)
>>
>> --
>> Freddie Cash
>> fjwc...@gmail.com
>>
> 
> Well the thing is I like use glabel in place of a fs label as it makes
> my devices all names something similar
> /dev/label/thingie rather then have /dev/ufs/rootfs and
> /dev/label/swap  , I'll have /dev/label/rootfs /dev/label/swap.
> Also I dont think you can easily change a live label either .

"tunefs -L" changes UFS label (/dev/ufs/label) and does not use extra 512 bytes 
of space.

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

Re: i386/153947: Make buildworld fails in hastd/token.c

[Eugene Grosbein]

This also breaks source upgrade path from RELENG_7.
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: kern/138341: [nanobsd] [patch] 8.0-BETA3: nanobsd build broken due to sysipc kernel module

[Eugene Grosbein]

'make MODULES_WITH_WORLD=yes buildworld' is still broken for 8.2-PRERELEASE.

Here is a patch for RELENG_8 sources updated today:

--- sys/modules/cryptodev/Makefile.orig 2010-08-23 12:13:44.0 +0700
+++ sys/modules/cryptodev/Makefile  2010-08-23 12:13:52.0 +0700
@@ -3,6 +3,6 @@
 .PATH: ${.CURDIR}/../../opencrypto
 KMOD   = cryptodev
 SRCS   = cryptodev.c
-SRCS   += bus_if.h device_if.h
+SRCS   += bus_if.h device_if.h opt_compat.h
 
 .include 
--- sys/modules/dtrace/lockstat/Makefile.orig   2009-09-16 23:05:25.0 
+0800
+++ sys/modules/dtrace/lockstat/Makefile2009-09-16 23:05:45.0 
+0800
@@ -5,7 +5,7 @@
 KMOD=  lockstat
 SRCS=  lockstat.c 
 
-SRCS+= vnode_if.h
+SRCS+= vnode_if.h opt_kdtrace.h
 
 CFLAGS+=   -I${.CURDIR}/../../../cddl/compat/opensolaris \
-I${.CURDIR}/../../../cddl/contrib/opensolaris/uts/common \
--- sys/modules/mqueue/Makefile.orig2010-04-24 17:47:03.0 +0700
+++ sys/modules/mqueue/Makefile 2010-04-24 17:47:14.0 +0700
@@ -5,6 +5,6 @@
 KMOD=  mqueuefs
 SRCS=  uipc_mqueue.c \
vnode_if.h \
-   opt_posix.h
+   opt_posix.h opt_compat.h
 
 .include 
--- sys/modules/sysvipc/sysvmsg/Makefile.orig   2009-08-30 19:12:16.0 
+0800
+++ sys/modules/sysvipc/sysvmsg/Makefile2009-09-19 01:12:18.0 
+0800
@@ -3,6 +3,6 @@
 .PATH: ${.CURDIR}/../../../kern
 
 KMOD=  sysvmsg
-SRCS=  sysv_msg.c opt_sysvipc.h
+SRCS=  sysv_msg.c opt_sysvipc.h opt_compat.h
 
 .include 
--- sys/modules/sysvipc/sysvsem/Makefile.orig   2009-08-30 19:52:13.0 
+0800
+++ sys/modules/sysvipc/sysvsem/Makefile2009-08-30 19:52:33.0 
+0800
@@ -3,6 +3,6 @@
 .PATH: ${.CURDIR}/../../../kern
 
 KMOD=  sysvsem
-SRCS=  sysv_sem.c opt_sysvipc.h
+SRCS=  sysv_sem.c opt_sysvipc.h opt_compat.h
 
 .include 
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: 8.2-PRE: hastd: Compile Error

[Eugene Grosbein]

On 18.01.2011 04:38, Larry Rosenman wrote:
> What can I do to fix this?
> 
> csup'd around 2pm Central (US) time from cvsup5.
> 
> cc -O2 -pipe  -I/usr/src/sbin/hastd -DINET -DINET6 -DYY_NO_UNPUT 
> -DYY_NO_INPUT -DHAVE_CRYPTO -std=gnu99 -fstack-protector -Wsystem-headers 
> -Werror -Wall -Wno-format-y2k -W -Wno-unused-parameter -Wstrict-prototypes 
> -Wmissing-prototypes -Wpointer-arith -Wreturn-type -Wcast-qual 
> -Wwrite-strings -Wswitch -Wshadow -Wcast-align -Wunused-parameter 
> -Wchar-subscripts -Winline -Wnested-externs -Wredundant-decls 
> -Wno-pointer-sign -c token.c
> cc1: warnings being treated as errors
> /usr/src/sbin/hastd/token.l:456: warning: function declaration isn't a 
> prototype
> *** Error code 1
> 
> Stop in /usr/src/sbin/hastd.
> *** Error code 1
> 
> Stop in /usr/src/sbin.
> *** Error code 1
> 
> Stop in /usr/src.
> *** Error code 1
> 
> Stop in /usr/src.
> *** Error code 1
> 
> Stop in /usr/src.
> # $

Someone have to fix hastd's Makefile and change WARNS level there.
This problem breaks source upgrade.

Eugene Grosbein



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

Re: 8.2-PRE: hastd: Compile Error

[Eugene Grosbein]

On 18.01.2011 13:31, Larry Rosenman wrote:

>> Someone have to fix hastd's Makefile and change WARNS level there.
>> This problem breaks source upgrade.
>>
>> Eugene Grosbein
> 
> Is anyone working on it?  I'm trying to get a source upgrade from 8.0 to 8.2 
> done. 

It seems, no. Also, there is PR for that:
http://www.freebsd.org/cgi/query-pr.cgi?pr=153947

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

Re: 8.2-PRE: hastd: Compile Error

[Eugene Grosbein]

On 18.01.2011 13:34, Lystopad Aleksandr wrote:

> There are solution from Vsevolod Lobko:
>
>   : just do:
>   : 
>   : cd /usr/src/usr.bin/lex
>   : make obj
>   : make
>   : make install
>   : 
>   : and buildworld works properly
> 
> in http://www.freebsd.org/cgi/query-pr.cgi?pr=153947
> 
> I make buildworld without problems after that yesturday.

That's not a solution for 7->8 upgrade as new lex won't run under 7.x kernel.
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: 8.2-PRE: hastd: Compile Error

[Eugene Grosbein]

On 18.01.2011 14:18, Olivier Certner wrote:
>> I make buildworld without problems after that yesturday.
> 
>   Same here. Perhaps someone could add a small entry about that in 
> UPDATING? It 
> seems that this problem is likely to be reported over and over without such a 
> notice.

IMO the problem should be fixed, not documented :-)
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: 8.2-PRE: hastd: Compile Error

[Eugene Grosbein]

On 18.01.2011 14:47, Olivier Certner wrote:
>> That's not a solution for 7->8 upgrade as new lex won't run under 7.x
>> kernel.
> 
>   I can confirm that compiling and installing 'lex' in advance produces a 
> working 'lex' on 7.x. 
> 
>   My current understanding is that the procedure mentioned in a previous 
> mail 
> uses the 8.x source but the 7.x tools and headers, so 'lex' then runs 
> correctly, since it also doesn't depend on exotic kernel features found only 
> in 8.x.

Got it, thanks. This is still workaround until real fix made, not solution.
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: 8-STABLE/amd64 semi-regular crash with "kernel trap 12 with interrupts disabled" in "process 12 (swi4: clock)"

[Eugene Grosbein]

On 19.01.2011 03:12, Lev Serebryakov wrote:
> Hello, Freebsd-stable.
> 
> 
>   One of my servers crashes about once a week, with always same
> diagnostics: "kernel trap 12 with interrupts disabled" and in same
> process: "swi4: clock"
> 
>   It doesn't look as memory failure, as memtest86+ can not find any
> errors in 8 passes.
> 
>   Also, after this crash server refuse to auto-reboot, last message on
> console is "cpu_reset: Stopping other CPUs", and it hangs.
> 
>   Kernel config, booting dmesg & results of "savecore" are attached
> (bzipped).

You have not mentioned what tasks does it perform.

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

Re: RFC: Upgrade BIND version in RELENG_7 to BIND 9.6.x

[Eugene Grosbein]

On 18.12.2010 15:16, Eugene Grosbein wrote:
> On 18.12.2010 11:41, Doug Barton wrote:
> 
>> I am particularly interested in feedback from users with significant DNS
>> usage that are still using 9.4, especially if you're using the version
>> in the base. I would appreciate it if you could install 9.6 from the
>> ports and at minimum run /usr/local/sbin/named-checkconf to see if any
>> errors are generated. Of course it would be that much more helpful if
>> you could also evaluate BIND 9.6 in operation in your environment.
>>
>> Your feedback on the issue of upgrading BIND in RELENG_7 is welcome.
>> Sooner is better. :)
> 
> I run serveral DNS servers based on FreeBSD+bind.
> One of them is FreeBSD 7.3-STABLE and bind-9.4.
> It's public primary server for two zones (one forward and one reverse)
> and public secondary for several others; plus it's recursive caching DNS 
> server
> for users, low-loaded.
> 
> I've just updated ports tree and installed dns/bind96.
> /usr/local/sbin/named-checkconf shows no messages at all.
> 
> I've restarted bind using ports version. All seems OK. I'll write back if 
> problems arise.
> About upgrading bind version in RELENG_7, I'm for it.

Full month has passed, no problems noticed :-)
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: 8-STABLE/amd64 semi-regular crash with "kernel trap 12 with interrupts disabled" in "process 12 (swi4: clock)"

[Eugene Grosbein]

On 19.01.2011 03:12, Lev Serebryakov wrote:
> Hello, Freebsd-stable.
> 
> 
>   One of my servers crashes about once a week, with always same
> diagnostics: "kernel trap 12 with interrupts disabled" and in same
> process: "swi4: clock"
> 
>   It doesn't look as memory failure, as memtest86+ can not find any
> errors in 8 passes.
> 
>   Also, after this crash server refuse to auto-reboot, last message on
> console is "cpu_reset: Stopping other CPUs", and it hangs.
> 
>   Kernel config, booting dmesg & results of "savecore" are attached
> (bzipped).

There is known instability in em(4) driver in 8.2-RELEASE,
it may panic due to some lack of NULL pointer checks.
You should update to RELENG_8 containting fix and retest.

Eugene Grosbein

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

Re: 8-STABLE/amd64 semi-regular crash with "kernel trap 12 with interrupts disabled" in "process 12 (swi4: clock)"

[Eugene Grosbein]

On 19.01.2011 15:00, Lev Serebryakov wrote:
> Hello, Eugene.
> You wrote 19 января 2011 г., 11:44:01:
> 
>> There is known instability in em(4) driver in 8.2-RELEASE,
>> it may panic due to some lack of NULL pointer checks.
>> You should update to RELENG_8 containting fix and retest.
>  uname -v
>  FreeBSD 8.2-PRERELEASE #5: Sat Jan  8 14:38:46 MSK 2011
> 
>  It is built about hour after cvsup.

Yes, I've missed it's PRERELEASE already.
Backtrace points to the problem in em_local_timer() fixed in CURRENT
7 days ago, take a look:

http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/dev/e1000/if_em.c#rev1.65

I run my servers with this commit backported manually as
it has not been MFC'd yet.

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

Living on gmirror: need to reincarnate /etc/rc.early

[Eugene Grosbein]

Hi!

In RELENG_8, gmirror is good enough to keep whole HDD pair withing the mirror.
Its performance, stability any pretty ease of maintainance allows
to use it widely.

With wide deployment of gmirror in production I've faced inability
of RELENG_8 to store kernel crashdumps out-of-the-box.
gmirror manual page documents a way to setup FreeBSD so that
it would store crashdumps again but that way involves /etc/rc.early
removed from RELENG_8. I've read about intentions - it was unsafe etc.
But we still need working crashdump support.

Easiest way is to reincarnate /etc/rc.d/early support making it better and safer
and it should support gmirror's mechanics for crashdumps out-of-the-box.

Comments?

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

Re: em0: Watchdog timeout -- resetting

[Eugene Grosbein]

On 01.02.2011 13:58, Lev Serebryakov wrote:
> Hello, Freebsd-stable.
> You wrote 1 февраля 2011 г., 10:24:16:
> 
>>   And all connections are reset. Before latest commits to driver
>> this system paniced in swi_clock. Now it works without panics, but
>> seems, that problem is not fixed completely.
>   I forgot to give one last pice of information: POLLING is in action.
> Without it single thread copy from this server via SMB eats one core
> of CPU completely.
> 

You could give a try to netisr parallelism of RELENG_8 instead of POLLING
(and tune interrupt throttling) if your box does not have lots of dynamic
interfaces like when using mpd.

In /etc/sysctl.conf:

net.isr.direct=0
net.isr.direct_force=0

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

Re: em0: Watchdog timeout -- resetting

[Eugene Grosbein]

On 01.02.2011 18:38, Lev Serebryakov wrote:

> = INTR - ISR.DIRECT=1
> Real speed (accroding to Windows'7 report) ~101MiB/s.
> 
>   I've re-created file to flush caches on both sides between trys.
> 

netisr queues help to deal with lots of incoming traffic.
If you bother about outgoing traffic only, it won't help.

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

System does not boot without console

[Eugene Grosbein]

Hi!

I have ASUS P7H55 (no integrated video) motherboard that boots
FreeBSD 8.2-PRERELEASE just fine even when I do not plug any video card
in its PCIEx16 slot. System boots and works just fine without video
and keyboard but with default /boot/device.hints only.

It does NOT boot if I make the change:

--- /boot/device.hints.orig   2011-02-15 15:58:11.0 +0600
+++ /boot/device.hints2011-02-15 15:58:17.0 +0600
@@ -25,7 +25,7 @@
 hint.apm.0.flags="0x20"
 hint.uart.0.at="isa"
 hint.uart.0.port="0x3F8"
-hint.uart.0.flags="0x10"
+hint.uart.0.flags="0x0"
 hint.uart.0.irq="4"
 hint.uart.1.at="isa"
 hint.uart.1.port="0x2F8"

If I power the box down thereafter, insert video card and boot it again,
it boots and I see it did not mount root file system, it's clean.
Normally it takes a munute to boot this system and I waited for hours to test.

It does not boot without video card if I additionally write
"-m" (without quotes) to /boot.config and "console=nullconsole" to 
/boot/loader.conf

Is it supposed behavour?

Motivation: I need the only PCIEx16 slot for Intel dualport gigabit ethernet 
network adapter
and want to have common NanoBSD image for boxes like this and other boxes 
having IPMI card
and want to have hint.uart.2.flags="0x10" in /boot/loader.conf

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

Re: GELI speed

[Eugene Grosbein]

On 29.03.2011 00:57, Clayton Milos wrote:
> Hi Pawel
> 
> I'm running GELI encryption on a hardware RAID5 (8x1TB), formatted to 6.3T 
> usable space using i386 FreeBSD 7.3-PRERELEASE. Write speeds to the drive 
> are just over 40MB/s which is not too bad but I would like to improve it as 
> the drive gets 270MB/s without encryption.
> 
> I've run the following tests:
> dd if=/dev/gzero.eli of=/dev/null bs=1m count=4096
> 4294967296 bytes transferred in 99.763266 secs (43051591 bytes/sec)
> and then
> dd if=/dev/zero of=test.file bs=1m count=1024
> 1073741824 bytes transferred in 26.118988 secs (41109626 bytes/sec)
> 
> There is not much speed lost due to the filesystem throughput or the RAID 
> card.

You can also try increase reading performance by using geom_cache
with sufficiently large cache. It won't impact on linear reads
but for general operations like directory lookups, many short reads etc.
it can decrease number of decrypt requests.

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

Re: Filesystem strangeness

[Eugene Grosbein]

On Tue, Apr 05, 2011 at 09:22:27AM +0200, Kurt Jaeger wrote:

> On a FreeBSD 8.0-RELEASE i386 (GENERIC kernel):
> 
> # cd /usr/src/sys/boot/efi/libefi/
> # ls -l
> ls: efinet.c: Bad file descriptor

Your file system it broken and needs fsck.

> total 38
> -rw-r--r--  1 root  wheel   461 Oct 25  2009 Makefile
> -rw-r--r--  1 root  wheel  1777 Oct 25  2009 delay.c
> -rw-r--r--  1 root  wheel  2682 Oct 25  2009 efi_console.c
> -rw-r--r--  1 root  wheel  9957 Oct 25  2009 efifs.c
> -rw-r--r--  1 root  wheel  2269 Oct 25  2009 errno.c
> -rw-r--r--  1 root  wheel  2536 Oct 25  2009 handles.c
> -rw-r--r--  1 root  wheel  5712 Oct 25  2009 libefi.c
> -rw-r--r--  1 root  wheel  6087 Oct 25  2009 time.c
> # df /usr
> Filesystem  1K-blocksUsedAvail Capacity  
> Mounted on
> /dev/ad4s1e 101554150 3143488157741142862 -3143488157647713044 3364544879817% 
>/usr
> gk-hwkrt-new# 
> 
> The last reboot did not do any fsck, smartctl does not complain.
> So, what can I do to fix this ?

Go to single user mode, unmount /usr and run "fsck -y /usr"

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

Builworld is broken for RELENG_8 when CPUTYPE?=core2

[Eugene Grosbein]

Hi!

Yesterday I've updated sources of my RELENG_8 box using csup
and ran "make -j5 buildworld" while having CPUTYPE?=core2 in /etc/make.conf

It failed:

===> kerberos5/tools/asn1_compile (depend)
cd /home/src/kerberos5/tools/asn1_compile/../make-print-version && make
cd /home/src/kerberos5/tools/asn1_compile/../make-roken && make
lex -t  
/home/src/kerberos5/tools/asn1_compile/../../../crypto/heimdal/lib/asn1/lex.l > 
lex.c
sed -e '96s/"/"#ifdef __PARSE_UNITS_H__\\n/;' -e '96s/",/\\n#endif\\n",/' 
/home/src/kerberos5/tools/asn1_compile/../../../crypto/heimdal/lib/asn1/gen_glue.c
 > gen_glue-fixed.c
yacc -d -o parse.c 
/home/src/kerberos5/tools/asn1_compile/../../../crypto/heimdal/lib/asn1/parse.y
yacc: 4 shift/reduce conflicts
cc -O2 -pipe -march=core2 -DHAVE_CONFIG_H 
-I/home/src/kerberos5/tools/make-print-version/../../include -std=gnu99   -c 
/home/src/kerberos5/tools/make-print-version/../../../crypto/heimdal/lib/vers/make-print-version.c
cc -O2 -pipe -march=core2 -DHAVE_CONFIG_H 
-I/home/src/kerberos5/tools/make-roken/../../include -std=gnu99   -c 
make-roken.c
/home/src/kerberos5/tools/make-print-version/../../../crypto/heimdal/lib/vers/make-print-version.c:1:
 error: bad value (core2) for -march= switch
/home/src/kerberos5/tools/make-print-version/../../../crypto/heimdal/lib/vers/make-print-version.c:1:
 error: bad value (core2) for -mtune= switch
make-roken.c:1: error: bad value (core2) for -march= switch
make-roken.c:1: error: bad value (core2) for -mtune= switch
*** Error code 1
1 error
*** Error code 2
*** Error code 1
1 error
*** Error code 2
2 errors
*** Error code 2
1 error
*** Error code 2
1 error
*** Error code 2
1 error

Please take a look.

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

running newsyslog fiveminly

[Eugene Grosbein]

Hi!

Suppose, there is a machine which writes two kinds of log files through syslogd:
quickly-growing that need to be rotated based on their size (hourly is too 
seldom)
and other that should be rotated once a day, at midnight only.

For first kind of logs we have to run newsyslog once every 5 minutes using cron:

*/5 *   *   *   *   rootnewsyslog

For second kind of logs we have lines in newsyslog.conf such as following:

/var/log/mpd.log 640 16 * @T  JC

This must ensure that /var/log/mpd.log is rotated and compressed at midnigt 
only.
Note, that compressing the file takes 8 minutes.

However, every night at 00:05 I get an error:

bzip2: I/O or other error, bailing out.  Possible reason follows.
bzip2: No such file or directory
Input file = /var/log/mpd.log.0, output file = /var/log/mpd.log.0.bz2
newsyslog: `bzip2 -f /var/log/mpd.log.0' terminated with a non-zero status (1)

It seems, newsyslog still wants to process my file at 00:05 despite @T
time specification. Is it broken?

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

Re: running newsyslog fiveminly

[Eugene Grosbein]

31.07.2011 23:57, Freddie Cash пишет:
> Simplest solution is to have two separate config files for newsyslog

My question is about possibly broken newsyslog, not how about "real task",
it is described for clarification only.

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

Re: running newsyslog fiveminly

[Eugene Grosbein]

01.08.2011 00:31, Jeremy Chadwick writes:

>> For second kind of logs we have lines in newsyslog.conf such as following:
>>
>> /var/log/mpd.log 640 16 * @T  JC
>>
>> This must ensure that /var/log/mpd.log is rotated and compressed at midnigt 
>> only.
>> Note, that compressing the file takes 8 minutes.
   ^^^

> I have three things to say on the matter, all of which are somewhat

Five things really :-)

> independent of one another so please keep that in mind.  I imagine #1
> below is your problem.
> 
> 1) The newsyslog.conf(5) man page has this clause in it, for the "when"
> field (in your case, @T):
> 
>  when...  If the when field contains an asterisk (`*'), log rotation
>  will solely depend on the contents of the size field.  Otherwise,
>  the when field consists of an optional interval in hours, usually
>  followed by an `@'-sign and a time in restricted ISO 8601 format.
> 
>  If a time is specified, the log file will only be trimmed if
>  newsyslog(8) is run within one hour of the specified time.  If an
>  interval is specified, the log file will be trimmed if that many
>  hours have passed since the last rotation. ...
> 
> You might think that "one hour of the specified time" value/clause
> correlates with the interval that newsyslog is run at via cron, but that
> would be wrong.  newsyslog REALLY DOES have hard-coded values for 3600
> seconds (1 hour) in it (grep -r 3600 /usr/src/usr.sbin/newsyslog).  I
> have not looked at the code, but the fact of the matter is, 1 hour
> appears to be a "special" value.  I would heed that as a warning.
> 
> 2) Are you absolutely sure mpd.log is being rotated AND compressed within
> the 5 minute window?  If mpd.log is extremely large and your disks are
> slow, this could take a long time.  If possible, try (temporarily)
> removing bzip2 from the picture (remove J flag).

I've noted (see above) that compression takes 8 minutes.
I just think newsyslog should not deal with the file at 00:05.

> 3) mpd(8) logs via syslog(3).  When newsyslog(8), are you aware that it
> sends a SIGHUP to syslogd(8)?  As such, are you absolutely certain when
> this happen (every 5 minutes!) that the new log files are getting
> created correctly and promptly?

I see no other problems.

> 4) To debug this, you're probably going to need to run some cronjobs or
> daemons that keep a very close eye on /var/log/mpd.log* when the log
> rotation runs, in combination with running syslogd(8) in debug mode
> and/or verbose mode.

syslogd or newsyslo needs debug mode?

> 5) Why do you need to rotate logs every 5 minutes?  Why do you need such
> extreme levels of granularity in your rotated logs?  Just how much data
> are you logging via syslog?  If a lot, why so much?  It might be more
> effective to consider expanding your logging infrastructure to multiple
> machines if this the case.

Most of my boxes are diskless NanoBSD installations having /var in memory
and I need very detailed debug logs that grow quickly. These logs
can easily overflow /var partition in case of network problems (storms etc.)
so newsyslog have to check them often.

And I have another router that has an HDD to keep daily log and I'd like
to have their crontabs unified.

Eugene Grosbein

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

Re: running newsyslog fiveminly

[Eugene Grosbein]

01.08.2011 09:39, Jason Hellenthal пишет:
> 
> What line of the newsyslog.conf file is your line inserted and can you
> move that to a higher line number. FIFO

I don't get why position of line in newsyslog.conf can have any meaning.

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

Re: 32GB limit per swap device?

[Eugene Grosbein]

10.08.2011 17:42, Daniel Kalchev writes:

> I believe the gmirror bug might exist in smaller partitions as well, but 
> haven't tested it yet (have few such systems that never duped core). It 
> does not matter if I do full dump or minidump: on gmirrored 64GB 
> partittion savecore does not find anything.

You should read gmirror(8) manual page about "Doing kernel dumps to gmirror 
providers".

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

Re: 8-stable panics in igb(8)

[Eugene Grosbein]

25.08.2011 16:41, Maxim Sobolev пишет:
> Hi,
> 
> We have observed what appears to be a crash in the interrupt thread for 
> the igb(8) driver on one of our systems. This machine runs May 22's 
> 8.2-STABLE. For some reason the IP of the crash (0x806d6399) 
> points to the key_sendup0() function, however we never had IPsec 
> configured on that box, so I suspect it is not accurate.
> 
> http://sobomax.sippysoft.com/ScreenShot833.png
> 
> Any ideas of what could be wrong there are appreciated.

It would be nice to see KDB backtrace with symbols:

options KDB
options KDB_TRACE
options KDB_UNATTENDED
options DDB
options DDB_NUMSYM

and "options KDB_UNATTENDED" if you wish.

For next crash, of course :-)

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

Unfixable UFS2 corruption

[Eugene Grosbein]

Hi!

Long story short: my /usr/local UFS2 filesystem somehow got corrupted
and "fsck -y" in single user mode does not fix it.

Explanation:

# ls -al /usr/local/obj/usr/local/src/secure/lib/libssh
ls: : No such file or directory
total 8
drwxr-xr-x  2 root  wheel  4608 Aug 30 01:28 .
drwxr-xr-x  3 root  wheel   512 Aug 30 01:28 ..

# rm -rf /usr/local/obj/usr/local/src/secure/lib/libssh
rm: /usr/local/obj/usr/local/src/secure/lib/libssh: Directory not empty

As I've said, I cold booted this FreeBSD 8.2-STABLE system to single user mode
where all file systems are not mounted (except root) and ran fsck -y /usr/local
It found no errors and said it is CLEAN. The problem still persists.

I've written small program and it said me this directory contains third file
(besides <.> and <..> entries) having zero file length.

I got contents of the directory to plain file with
"cat /usr/local/obj/usr/local/src/secure/lib/libssh > /tmp/libssh and put it 
online:
http://www.grosbein.net/crash/corruption/libssh

Please help. The program and its output follow:

#include 
#include 
#include 
#include 

int main(int argc, char* argv[])
{

  DIR   *dirp;
  struct dirent *dp;
  unsigned  i;

  if (argc<2)
return 1;

  if ( (dirp = opendir(argv[1])) == NULL )
err (1, "opendir");

  i = 0;
  while ((dp = readdir(dirp)) != NULL) {
i++;
printf("Entry %u:\n"
   "d_fileno=%u\n"
   "d_reclen=%u\n"
   "d_type=%u\n"
   "d_namlen=%u\n"
   "d_name=<%s>\n\n",
   i, (unsigned) dp->d_fileno, (unsigned) dp->d_reclen,
   (unsigned) dp->d_type, (unsigned) dp->d_namlen,
   (char *) dp->d_name);
  }
  return closedir(dirp);
}

# # ./readdir /usr/local/obj/usr/local/src/secure/lib/libssh
Entry 1:
d_fileno=1531227
d_reclen=12
d_type=4
d_namlen=1
d_name=<.>

Entry 2:
d_fileno=1389650
d_reclen=500
d_type=4
d_namlen=2
d_name=<..>

Entry 3:
d_fileno=24
d_reclen=512
d_type=8
d_namlen=0
d_name=<>
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: Unfixable UFS2 corruption

[Eugene Grosbein]

31.08.2011 21:35, Eugene Grosbein пишет:

> # ls -al /usr/local/obj/usr/local/src/secure/lib/libssh
> ls: : No such file or directory
> total 8
> drwxr-xr-x  2 root  wheel  4608 Aug 30 01:28 .
> drwxr-xr-x  3 root  wheel   512 Aug 30 01:28 ..
> 
> # rm -rf /usr/local/obj/usr/local/src/secure/lib/libssh
> rm: /usr/local/obj/usr/local/src/secure/lib/libssh: Directory not empty
> 
> As I've said, I cold booted this FreeBSD 8.2-STABLE system to single user mode
> where all file systems are not mounted (except root) and ran fsck -y 
> /usr/local
> It found no errors and said it is CLEAN. The problem still persists.
> 
> I've written small program and it said me this directory contains third file
> (besides <.> and <..> entries) having zero file length.

Not file but file name length is zero. I've just found that
dircheck() function in src/sbin/fsck_ffs/dir.c simply does not check
if d_namlen is zero as it should, shouldn't it?

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

Re: Unfixable UFS2 corruption

[Eugene Grosbein]

31.08.2011 23:02, Adam Vande More пишет:

> Long story short: my /usr/local UFS2 filesystem somehow got corrupted
> and "fsck -y" in single user mode does not fix it.
> 
> Not sure if this helps or not but on rare occasion I've had to run fsck twice 
> consecutively to fix a FS.

Not this time - fsck does NOT find any problems in this file system.

Now I think fsck_ffs needs a patch:

--- sbin/fsck_ffs/dir.c.orig2011-08-31 22:54:23.0 +0700
+++ sbin/fsck_ffs/dir.c 2011-08-31 22:54:48.0 +0700
@@ -225,7 +225,7 @@
type = dp->d_type;
if (dp->d_reclen < size ||
idesc->id_filesize < size ||
-   namlen > MAXNAMLEN ||
+   namlen == 0 || namlen > MAXNAMLEN ||
type > 15)
goto bad;
for (cp = dp->d_name, size = 0; size < namlen; size++)


Comments?

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

Re: Unfixable UFS2 corruption

[Eugene Grosbein]

31.08.2011 23:13, Eugene Grosbein пишет:
> 31.08.2011 23:02, Adam Vande More пишет:
> 
>> Long story short: my /usr/local UFS2 filesystem somehow got corrupted
>> and "fsck -y" in single user mode does not fix it.
>>
>> Not sure if this helps or not but on rare occasion I've had to run fsck 
>> twice consecutively to fix a FS.
> 
> Not this time - fsck does NOT find any problems in this file system.
> 
> Now I think fsck_ffs needs a patch:
> 
> --- sbin/fsck_ffs/dir.c.orig  2011-08-31 22:54:23.0 +0700
> +++ sbin/fsck_ffs/dir.c   2011-08-31 22:54:48.0 +0700
> @@ -225,7 +225,7 @@
>   type = dp->d_type;
>   if (dp->d_reclen < size ||
>   idesc->id_filesize < size ||
> - namlen > MAXNAMLEN ||
> + namlen == 0 || namlen > MAXNAMLEN ||
>   type > 15)
>   goto bad;
>   for (cp = dp->d_name, size = 0; size < namlen; size++)
> 
> 
> Comments?

With this patch applied, my FS has finally been fixed by fsck:

** Last Mounted on /usr/local
** Phase 1 - Check Blocks and Sizes
** Phase 2 - Check Pathnames
DIRECTORY CORRUPTED  I=1531227  OWNER=root MODE=40755
SIZE=4608 MTIME=Aug 30 01:28 2011 
DIR=/obj/usr/local/src/secure/lib/libssh

SALVAGE? [yn] 

** Phase 3 - Check Connectivity
** Phase 4 - Check Reference Counts
LINK COUNT FILE I=24  OWNER=root MODE=100644
SIZE=892 MTIME=Sep 17 11:10 2010  COUNT 2 SHOULD BE 1
ADJUST? [yn] 

** Phase 5 - Check Cyl groups
459580 files, 7411823 used, 7819495 free (105503 frags, 964249 blocks, 0.7% 
fragmentation)

* FILE SYSTEM IS CLEAN *

* FILE SYSTEM WAS MODIFIED *

Should I fill PR?

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

Re: Unfixable UFS2 corruption

[Eugene Grosbein]

31.08.2011 23:34, Adrian Chadd пишет:
> Have you created a PR for this?

http://www.freebsd.org/cgi/query-pr.cgi?pr=160339

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

Re: stop scheduler on panic patches updated

[Eugene Grosbein]

02.09.2011 23:22, Andriy Gapon пишет:
> 
> The patches can be found at the same locations:
> head - http://people.freebsd.org/~avg/stop_scheduler_on_panic.diff
> stable/8 - http://people.freebsd.org/~avg/stop_scheduler_on_panic.8.x.diff
> 
> Additionally, if you use a USB keyboard, then the following patch is required 
> for
> proper operation in post-panic environment:
> http://people.freebsd.org/~avg/stop_scheduler_on_panic.usb.diff
> The patch is the same for both head and stable/8.
> It shouldn't hurt if you don't use USB devices or use other USB devices.
> 

Thanks, I'm going to try it just now because my dummynet panics are back
with start of new school year :-)

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

Re: running newsyslog fiveminly

[Eugene Grosbein]

01.08.2011 00:31, Jeremy Chadwick пишет:
> On Sun, Jul 31, 2011 at 11:51:40PM +0700, Eugene Grosbein wrote:
>> Hi!
>>
>> Suppose, there is a machine which writes two kinds of log files through 
>> syslogd:
>> quickly-growing that need to be rotated based on their size (hourly is too 
>> seldom)
>> and other that should be rotated once a day, at midnight only.
>>
>> For first kind of logs we have to run newsyslog once every 5 minutes using 
>> cron:
>>
>> */5 *   *   *   *   rootnewsyslog
>>
>> For second kind of logs we have lines in newsyslog.conf such as following:
>>
>> /var/log/mpd.log 640 16 * @T  JC
>>
>> This must ensure that /var/log/mpd.log is rotated and compressed at midnigt 
>> only.
>> Note, that compressing the file takes 8 minutes.
>>
>> However, every night at 00:05 I get an error:
>>
>> bzip2: I/O or other error, bailing out.  Possible reason follows.
>> bzip2: No such file or directory
>>  Input file = /var/log/mpd.log.0, output file = /var/log/mpd.log.0.bz2
>> newsyslog: `bzip2 -f /var/log/mpd.log.0' terminated with a non-zero status 
>> (1)
>>
>> It seems, newsyslog still wants to process my file at 00:05 despite @T
>> time specification. Is it broken?
> 
> I have three things to say on the matter, all of which are somewhat
> independent of one another so please keep that in mind.  I imagine #1
> below is your problem.
> 
> 1) The newsyslog.conf(5) man page has this clause in it, for the "when"
> field (in your case, @T):
> 
>  when...  If the when field contains an asterisk (`*'), log rotation
>  will solely depend on the contents of the size field.  Otherwise,
>  the when field consists of an optional interval in hours, usually
>  followed by an `@'-sign and a time in restricted ISO 8601 format.
> 
>  If a time is specified, the log file will only be trimmed if
>  newsyslog(8) is run within one hour of the specified time.  If an
>  interval is specified, the log file will be trimmed if that many
>  hours have passed since the last rotation. ...
> 
> You might think that "one hour of the specified time" value/clause
> correlates with the interval that newsyslog is run at via cron, but that
> would be wrong.  newsyslog REALLY DOES have hard-coded values for 3600
> seconds (1 hour) in it (grep -r 3600 /usr/src/usr.sbin/newsyslog).  I
> have not looked at the code, but the fact of the matter is, 1 hour
> appears to be a "special" value.  I would heed that as a warning.

After reading newsyslog code, now it's obvious it just ignores minutes and 
seconds
while making decision if a file should be rotated. It looks at hours only.
That's sad.

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

Re: running newsyslog fiveminly

[Eugene Grosbein]

07.09.2011 17:08, Jeremy Chadwick writes:

>> After reading newsyslog code, now it's obvious it just ignores minutes and 
>> seconds
>> while making decision if a file should be rotated. It looks at hours only.
>> That's sad.
> 
> I imagine this "design limitation" is due to the fact that newsyslog is
> called from cron, which only supports minute-level granularity.
> 
> The newsyslog.conf man page even hints at this while describing the
> "when" column:
> 
> There is no provision for the specification of a timezone.  There
> is little point in specifying an explicit minutes or seconds com-
> ponent in the current implementation, since the only comparison
> is ``within the hour''.
> 
> Given this, I would say the "special" 3600-second value within the
> source code makes sense.
> 
> I'm not sure what you could use for an alternate method of log rotation
> for syslog-logged data.

I have just followed some of past advices and split my newsyslog.conf in two,
moving mpd-like logs with size-based rotation only to /etc/newsyslog-quick.conf.
And made another cron job for fiveminly running newsyslog -f 
/etc/newsyslog-quick.conf

> I think what the rest of the world might tell you is something to the
> effect of "you can't have your cake and eat it too".  You've got
> diskless systems that aren't syslogging via network (e.g. to a pool of
> syslog servers, or a single syslog server) but instead to a
> memory-backed filesystem, in addition to enabling debug-level logging in
> mpd by default.
> 
> A memory-backed filesystem means you don't have much disk space, and you
> know this based on the need to rotate logs every 5 minutes, right?  So
> I'm confused why one would need debug logging.  I imagine that the
> newsyslog.conf on these machines has a very small number for the "count"
> column for /var/log/mpd.log.  So chances are, by the time you noticed a
> problem, the logs would have been rotated and removed, no?  So why the
> debug logging?
> 
> If debug logging really is something you absolutely need, no argument
> about it, then honestly it sounds like you need some sort of
> "centralised" logging infrastructure for all of these diskless machines.
> Most diskless machines I've used utilise some form of centralised
> "something" -- whether it be a centralised DHCP/PXE server (which you
> obviously have in some form), or an NFS-mounted root or /home, etc...
> You get the idea.  Could you deploy similar infrastructure for syslog
> and simply use a remote syslog server in syslog.conf?

In fact, I do have centralized syslogd server that collects logs
from diskless servers. But, I need also local copies of individual
server's logs in the MFS. I was in hope to make it with one cron job
and one newsyslog.conf but as it seems impossible, I will use two cron jobs :-)

Local (compressed) logs residing on the MFS give me convinience to manage and 
debug
a server within one ssh session without need to consult with remote syslog 
archives.
In general, I do have enough MFS space to keep needed backlog
but in case of network PPPoE PADI broadcast storms I need quick rotation to 
prevent
MFS overflows. I think I'll get all of this now.

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

gmirror+gjournal often makes inconsistens file systems

[Eugene Grosbein]

Hi!

For long time I experience same UFS2 filesystem problems with several 8.2 
systems
running on gmirror+gjournal+async. In case of unclean shutdown, kernel panic or 
power failure
gjournal makes fsck skip its checks and that's why I use it.

But quite often my /var partition (and sometimes others) still has severe 
damage in it
and running with such /var mounted read-write leads to another panics or hangs 
and so on.

For example, I have such 8.2-STABLE system with ad4 and ad6 drives combined to 
/dev/mirror/gm0.
I have just removed ad6 from the mirror, ran fsck -y manually for all its 
filesystems,
shut down this machine again cleanly and booted it next time from ad6
while keeping mirror with ad4 not mounted nor checked.

Then, I ran fsck -y /dev/mirror/gm0.journals1e (/var on the mirrored drive)
and got LOTS of bad errors on presumably clean file system.
Of course, I've seen the same errors while checking ad6 after it was removed 
from running mirror.
I have auto-sync gmirror feature turned ON. I've tried to turn it OFF but that 
just
increase frequency of such damages not fixed after reboot.

It seems that gjournal cannot handle system crashes reliably, can it?
I basically run in without any manual tuning. I've also tried to tune it - 
without luck,
it works nice when there are no unclean shutdowns but it's here to deal with 
them in the first place.

# fsck -t ffs -y /dev/mirror/gm0.journals1e
** /dev/mirror/gm0.journals1e
** Last Mounted on /var
** Phase 1 - Check Blocks and Sizes
3955872 DUP I=989242
3955873 DUP I=989242
3955874 DUP I=989242
3955875 DUP I=989242
3955876 DUP I=989242
3955877 DUP I=989242
3955878 DUP I=989242
3955879 DUP I=989242
3955880 DUP I=989242
3955881 DUP I=989242
3955882 DUP I=989242
EXCESSIVE DUP BLKS I=989242
CONTINUE? yes

INCORRECT BLOCK COUNT I=989242 (448 should be 424)
CORRECT? yes

3955888 DUP I=989289
3955889 DUP I=989289
3955890 DUP I=989289
3955891 DUP I=989289
3955892 DUP I=989289
3955893 DUP I=989289
3955894 DUP I=989289
3955895 DUP I=989289
** Phase 1b - Rescan For More DUPS
3955872 DUP I=989242
3955873 DUP I=989242
3955874 DUP I=989242
3955875 DUP I=989242
3955876 DUP I=989242
3955877 DUP I=989242
3955878 DUP I=989242
3955879 DUP I=989242
3955880 DUP I=989242
3955881 DUP I=989242
3955888 DUP I=989242
3955889 DUP I=989242
3955890 DUP I=989242
3955891 DUP I=989242
3955892 DUP I=989242
3955893 DUP I=989242
3955894 DUP I=989242
3955895 DUP I=989242
** Phase 2 - Check Pathnames
DUP/BAD  I=989289  OWNER=root MODE=100640
SIZE=14367 MTIME=Sep  9 11:30 2011 
FILE=/log/kernel.log

REMOVE? yes

DUP/BAD  I=989242  OWNER=root MODE=100640
SIZE=202631 MTIME=Sep  8 19:52 2011 
FILE=/log/mpd.log.0

REMOVE? yes

** Phase 3 - Check Connectivity
** Phase 4 - Check Reference Counts
UNREF FILE I=376866  OWNER=root MODE=140666
SIZE=0 MTIME=Sep  5 12:27 2011 
CLEAR? yes

UNREF FILE I=376868  OWNER=root MODE=140666

UNREF FILE I=376868  OWNER=root MODE=140666
SIZE=0 MTIME=Sep  7 20:30 2011
CLEAR? yes

UNREF FILE I=376869  OWNER=root MODE=140666
SIZE=0 MTIME=Sep  8 11:17 2011
CLEAR? yes

UNREF FILE I=376870  OWNER=root MODE=140666
SIZE=0 MTIME=Sep  8 12:11 2011
CLEAR? yes

BAD/DUP FILE I=989242  OWNER=root MODE=100640
SIZE=202631 MTIME=Sep  8 19:52 2011
CLEAR? yes

UNREF FILE  I=989259  OWNER=root MODE=100640
SIZE=648 MTIME=Aug 27 00:00 2011
RECONNECT? yes

BAD/DUP FILE I=989289  OWNER=root MODE=100640
SIZE=14367 MTIME=Sep  9 11:30 2011
CLEAR? yes
LINK COUNT FILE I=989293  OWNER=root MODE=100640
SIZE=961 MTIME=Sep  9 11:26 2011  COUNT 1 SHOULD BE 2
ADJUST? yes

UNREF FILE  I=989327  OWNER=root MODE=100640
SIZE=114 MTIME=Aug 27 00:00 2011
RECONNECT? yes

** Phase 5 - Check Cyl groups
FREE BLK COUNT(S) WRONG IN SUPERBLK
SALVAGE? yes

SUMMARY INFORMATION BAD
SALVAGE? yes

BLK(S) MISSING IN BIT MAPS
SALVAGE? yes

1188 files, 90007 used, 4987072 free (360 frags, 623339 blocks, 0.0%
fragmentation)

* FILE SYSTEM IS CLEAN *

* FILE SYSTEM WAS MODIFIED *
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: gmirror+gjournal often makes inconsistens file systems

[Eugene Grosbein]

09.09.2011 15:21, Lev Serebryakov пишет:
> Hello, Eugene.
> You wrote 9 сентября 2011 г., 9:17:06:
> 
>> # fsck -t ffs -y /dev/mirror/gm0.journals1e
>   I may be wrong, but I've encountered strong advice not
>  to gjournal whole disk, but make gjournal on per-FS basis, many times.
>  And it seems, that your first create big journal, and splice/partition/newfs
>  it for several FSes.

Yes, I did. Shoud not this kind of partitioning work too?

Eugene Grosbein


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

Re: gmirror+gjournal often makes inconsistens file systems

[Eugene Grosbein]

Dear Pawel Jakub,

09.09.2011 12:17, Eugene Grosbein writes:
> Hi!
> 
> For long time I experience same UFS2 filesystem problems with several 8.2 
> systems
> running on gmirror+gjournal+async. In case of unclean shutdown, kernel panic 
> or power failure
> gjournal makes fsck skip its checks and that's why I use it.
> 
> But quite often my /var partition (and sometimes others) still has severe 
> damage in it
> and running with such /var mounted read-write leads to another panics or 
> hangs and so on.
> 
> For example, I have such 8.2-STABLE system with ad4 and ad6 drives combined 
> to /dev/mirror/gm0.
> I have just removed ad6 from the mirror, ran fsck -y manually for all its 
> filesystems,
> shut down this machine again cleanly and booted it next time from ad6
> while keeping mirror with ad4 not mounted nor checked.
> 
> Then, I ran fsck -y /dev/mirror/gm0.journals1e (/var on the mirrored drive)
> and got LOTS of bad errors on presumably clean file system.
> Of course, I've seen the same errors while checking ad6 after it was removed 
> from running mirror.
> I have auto-sync gmirror feature turned ON. I've tried to turn it OFF but 
> that just
> increase frequency of such damages not fixed after reboot.
> 
> It seems that gjournal cannot handle system crashes reliably, can it?
> I basically run in without any manual tuning. I've also tried to tune it - 
> without luck,
> it works nice when there are no unclean shutdowns but it's here to deal with 
> them in the first place.
> 
> # fsck -t ffs -y /dev/mirror/gm0.journals1e
> ** /dev/mirror/gm0.journals1e
> ** Last Mounted on /var
> ** Phase 1 - Check Blocks and Sizes
> 3955872 DUP I=989242
> 3955873 DUP I=989242
> 3955874 DUP I=989242
> 3955875 DUP I=989242
> 3955876 DUP I=989242
> 3955877 DUP I=989242
> 3955878 DUP I=989242
> 3955879 DUP I=989242
> 3955880 DUP I=989242
> 3955881 DUP I=989242
> 3955882 DUP I=989242
> EXCESSIVE DUP BLKS I=989242
> CONTINUE? yes
> 
> INCORRECT BLOCK COUNT I=989242 (448 should be 424)
> CORRECT? yes
> 
> 3955888 DUP I=989289
> 3955889 DUP I=989289
> 3955890 DUP I=989289
> 3955891 DUP I=989289
> 3955892 DUP I=989289
> 3955893 DUP I=989289
> 3955894 DUP I=989289
> 3955895 DUP I=989289
> ** Phase 1b - Rescan For More DUPS
> 3955872 DUP I=989242
> 3955873 DUP I=989242
> 3955874 DUP I=989242
> 3955875 DUP I=989242
> 3955876 DUP I=989242
> 3955877 DUP I=989242
> 3955878 DUP I=989242
> 3955879 DUP I=989242
> 3955880 DUP I=989242
> 3955881 DUP I=989242
> 3955888 DUP I=989242
> 3955889 DUP I=989242
> 3955890 DUP I=989242
> 3955891 DUP I=989242
> 3955892 DUP I=989242
> 3955893 DUP I=989242
> 3955894 DUP I=989242
> 3955895 DUP I=989242
> ** Phase 2 - Check Pathnames
> DUP/BAD  I=989289  OWNER=root MODE=100640
> SIZE=14367 MTIME=Sep  9 11:30 2011 
> FILE=/log/kernel.log
> 
> REMOVE? yes
> 
> DUP/BAD  I=989242  OWNER=root MODE=100640
> SIZE=202631 MTIME=Sep  8 19:52 2011 
> FILE=/log/mpd.log.0
> 
> REMOVE? yes
> 
> ** Phase 3 - Check Connectivity
> ** Phase 4 - Check Reference Counts
> UNREF FILE I=376866  OWNER=root MODE=140666
> SIZE=0 MTIME=Sep  5 12:27 2011 
> CLEAR? yes
> 
> UNREF FILE I=376868  OWNER=root MODE=140666
> 
> UNREF FILE I=376868  OWNER=root MODE=140666
> SIZE=0 MTIME=Sep  7 20:30 2011
> CLEAR? yes
> 
> UNREF FILE I=376869  OWNER=root MODE=140666
> SIZE=0 MTIME=Sep  8 11:17 2011
> CLEAR? yes
> 
> UNREF FILE I=376870  OWNER=root MODE=140666
> SIZE=0 MTIME=Sep  8 12:11 2011
> CLEAR? yes
> 
> BAD/DUP FILE I=989242  OWNER=root MODE=100640
> SIZE=202631 MTIME=Sep  8 19:52 2011
> CLEAR? yes
> 
> UNREF FILE  I=989259  OWNER=root MODE=100640
> SIZE=648 MTIME=Aug 27 00:00 2011
> RECONNECT? yes
> 
> BAD/DUP FILE I=989289  OWNER=root MODE=100640
> SIZE=14367 MTIME=Sep  9 11:30 2011
> CLEAR? yes
> LINK COUNT FILE I=989293  OWNER=root MODE=100640
> SIZE=961 MTIME=Sep  9 11:26 2011  COUNT 1 SHOULD BE 2
> ADJUST? yes
> 
> UNREF FILE  I=989327  OWNER=root MODE=100640
> SIZE=114 MTIME=Aug 27 00:00 2011
> RECONNECT? yes
> 
> ** Phase 5 - Check Cyl groups
> FREE BLK COUNT(S) WRONG IN SUPERBLK
> SALVAGE? yes
> 
> SUMMARY INFORMATION BAD
> SALVAGE? yes
> 
> BLK(S) MISSING IN BIT MAPS
> SALVAGE? yes
> 
> 1188 files, 90007 used, 4987072 free (360 frags, 623339 blocks, 0.0%
> fragmentation)
> 
> * FILE SYSTEM IS CLEAN *
> 
> * FILE SYSTEM WAS MODIFIED *

Please explain if such partitioning is supported?
physical drive - geom_mirror - geom_journal - geom_part_mbr - geom_part_bsd - 
journalled UFS2

If not, mounting such UFS2 should warn us, shouldn't it?
No warnings now.

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

busdma MFC broke ipfw fwd for RELENG_6

[Eugene Grosbein]

Hi!

I understand that it is "a bit" late for RELENG_6 reports as 6.4-RELEASE was 
out in 2008
but the breakage had happened due to MFC so it's possible same problem exists 
in newer branches.

Long story short: I've updated my old 6.4-STABLE system for recent zoneinfo 
updates
and found the update broke 'ipfw fwd' feature: forwarded packets get corrupted,
routed packets go just fine.

The commit in question has been performed in 2010/08/06:
http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/i386/i386/Attic/busdma_machdep.c.diff?r1=1.74.2.6;r2=1.74.2.7

I've rolled it back using recent RELENG_6 sources and packet corruption have 
disappeared.

Full commit can be seen with the following command:

cvs -q diff -u -j "RELENG_6:2010/08/06 00:00:00 GMT" -j "RELENG_6:2010/08/07 
00:00:00 GMT"

There was only one commit to RELENG_6 that day.
I do not understand how and why it broke "ipfw fwd" but that's the fact.

Some details: I have FreeBSD 6.4-STABLE router with LAN interface (rl0)
and WAN interface (re0) and L2TP tunnel (ng1) built with mpd5.
I forward some packets from LAN to ng0 with "ipfw fwd" and before the MFC
that setup worked just fine.

After update to recent RELENG_6, forwarded packets arrive corrupted to other 
side of L2TP tunnel
or get lost in between (due to bad headers, I guess). At sending side, tcpdump 
shows no problems
not for ng1 interface nor for re0 where it shows l2tp traffic with IP packets 
inside but really
ougouing L2TP packets go damaged (e.g. payload with zero bytes only), that can 
be seen
at receiving side of tunnel.

If I replace "ipfw fwd" with static route into the tunnel, packets go through 
just fine again,
so there is no problem with intermediate network.

But I need policy based routing and rollback of mentioned MFC has revived my 
setup again.
The guilty commit was found with simple "time dichotomy" (csup/build 
kernel/reboot/test/again etc.)

Just for archives.

Eugene Grosbein

P.S. Most of my production routers are still 6.4-STABLE
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: busdma MFC broke ipfw fwd for RELENG_6

[Eugene Grosbein]

16.09.2011 00:14, John Baldwin пишет:
> On Thursday, September 15, 2011 12:44:32 pm Eugene Grosbein wrote:
>> Hi!
>>
>> I understand that it is "a bit" late for RELENG_6 reports as 6.4-RELEASE was 
>> out in 2008
>> but the breakage had happened due to MFC so it's possible same problem 
>> exists in newer branches.
>>
>> Long story short: I've updated my old 6.4-STABLE system for recent zoneinfo 
>> updates
>> and found the update broke 'ipfw fwd' feature: forwarded packets get 
>> corrupted,
>> routed packets go just fine.
>>
>> The commit in question has been performed in 2010/08/06:
>> http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/i386/i386/Attic/busdma_machdep.c.diff?r1=1.74.2.6;r2=1.74.2.7
>>
>> I've rolled it back using recent RELENG_6 sources and packet corruption have 
>> disappeared.
> 
> It may be a bug in the rl(4) driver?  Perhaps this is a candidate to try?
> 
> http://svnweb.freebsd.org/base?view=revision&revision=184240
> 

My system is i386 with 512MB RAM only and outgoing packets go through re(4), 
not rl(4).

Nevertheless, I've just applied revision 184240 to RELENG_6 manualy (it did not 
apply cleanly)
and reapplied MFC. The problem has returned.

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

Re: busdma MFC broke ipfw fwd for RELENG_6

[Eugene Grosbein]

16.09.2011 01:15, YongHyeon PYUN пишет:

> I remember re(4) in 6.x also have a couple of bus_dma(9) bugs. How
> about applying the following revision?
> http://svnweb.freebsd.org/base?view=revision&revision=175337
> Not sure whether it shall apply cleanly.

It does not and there is too much differences in the code for my skills to 
apply manually :-)

> Alternatively try replacing BUS_DMA_ALLOCNOW to 0 in
> bus_dma_tag_create(9).

I'm not sure I undersdand this right... Do you mean this change?
This is the only place in busdma_machdep.c where BUS_DMA_ALLOCNOW is used.

--- busdma_machdep.c.orig   2011-09-16 01:56:52.0 +0700
+++ busdma_machdep.c2011-09-16 01:57:01.0 +0700
@@ -284,7 +284,7 @@
newtag->flags |= BUS_DMA_COULD_BOUNCE;
 
if (((newtag->flags & BUS_DMA_COULD_BOUNCE) != 0) &&
-   (flags & BUS_DMA_ALLOCNOW) != 0) {
+   (flags & 0) != 0) {
struct bounce_zone *bz;
 
/* Must bounce */

I'm going to lose control on this remote box if I apply wrong patch for re(4).

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

Re: busdma MFC broke ipfw fwd for RELENG_6

[Eugene Grosbein]

16.09.2011 02:19, YongHyeon PYUN пишет:
> On Fri, Sep 16, 2011 at 02:02:37AM +0700, Eugene Grosbein wrote:
>> 16.09.2011 01:15, YongHyeon PYUN пишет:
>>
>>> I remember re(4) in 6.x also have a couple of bus_dma(9) bugs. How
>>> about applying the following revision?
>>> http://svnweb.freebsd.org/base?view=revision&revision=175337
>>> Not sure whether it shall apply cleanly.
>>
>> It does not and there is too much differences in the code for my skills to 
>> apply manually :-)
>>
>>> Alternatively try replacing BUS_DMA_ALLOCNOW to 0 in
>>> bus_dma_tag_create(9).
>>
>> I'm not sure I undersdand this right... Do you mean this change?
> 
> No, change BUS_DMA_ALLOCNOW used in re(4).

With clean RELENG_6 sources and the only following patch,
the problem still persists.

--- if_re.c.orig2008-03-02 21:54:48.0 +0700
+++ if_re.c 2011-09-16 11:37:30.0 +0700
@@ -1077,7 +1077,7 @@
nseg = 32;
error = bus_dma_tag_create(sc->rl_parent_tag, ETHER_ALIGN, 0,
BUS_SPACE_MAXADDR_32BIT, BUS_SPACE_MAXADDR, NULL,
-   NULL, MCLBYTES * nseg, nseg, MCLBYTES, BUS_DMA_ALLOCNOW,
+   NULL, MCLBYTES * nseg, nseg, MCLBYTES, 0,
NULL, NULL, &sc->rl_ldata.rl_mtag);
if (error) {
device_printf(dev, "could not allocate dma tag\n");
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: busdma MFC broke ipfw fwd for RELENG_6

[Eugene Grosbein]

17.09.2011 02:13, YongHyeon PYUN пишет:
> On Fri, Sep 16, 2011 at 11:45:25AM +0700, Eugene Grosbein wrote:
>> 16.09.2011 02:19, YongHyeon PYUN пишет:
>>> On Fri, Sep 16, 2011 at 02:02:37AM +0700, Eugene Grosbein wrote:
>>>> 16.09.2011 01:15, YongHyeon PYUN пишет:
>>>>
>>>>> I remember re(4) in 6.x also have a couple of bus_dma(9) bugs. How
>>>>> about applying the following revision?
>>>>> http://svnweb.freebsd.org/base?view=revision&revision=175337
>>>>> Not sure whether it shall apply cleanly.
>>>>
>>>> It does not and there is too much differences in the code for my skills to 
>>>> apply manually :-)
>>>>
>>>>> Alternatively try replacing BUS_DMA_ALLOCNOW to 0 in
>>>>> bus_dma_tag_create(9).
>>>>
>>>> I'm not sure I undersdand this right... Do you mean this change?
>>>
>>> No, change BUS_DMA_ALLOCNOW used in re(4).
>>
>> With clean RELENG_6 sources and the only following patch,
>> the problem still persists.
>>
> 
> :-(
> I have back-ported re(4)/rl(4) for latest 6.x.
> http://people.freebsd.org/~yongari/re/6.x/README.txt
> Just compile tested and not sure whether it fixes the issue.

I confirm that the problem disappears using this driver with clean RELENG_6 
sources
and RealTek 8169/8169S/8169SB(L)/8110S/8110SB(L) Gigabit Ethernet:

re0@pci1:11:0:  class=0x02 card=0x816910ec chip=0x816910ec rev=0x10 hdr=0x00
vendor = 'Realtek Semiconductor'
device = 'RTL8110SB Single-Chip Gigabit LOM Ethernet Controller'
class  = network
subclass   = ethernet
cap 01[dc] = powerspec 2  supports D0 D1 D2 D3  current D0

Also, this machine uses RT8139 (rl0) for its LAN and it works too.

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

Re: busdma MFC broke ipfw fwd for RELENG_6

[Eugene Grosbein]

18.09.2011 03:05, YongHyeon PYUN пишет:

>>> I have back-ported re(4)/rl(4) for latest 6.x.
>>> http://people.freebsd.org/~yongari/re/6.x/README.txt
>>> Just compile tested and not sure whether it fixes the issue.
>>
>> I confirm that the problem disappears using this driver with clean RELENG_6 
>> sources
>> and RealTek 8169/8169S/8169SB(L)/8110S/8110SB(L) Gigabit Ethernet:
>>
>> re0@pci1:11:0:  class=0x02 card=0x816910ec chip=0x816910ec rev=0x10 
>> hdr=0x00
>> vendor = 'Realtek Semiconductor'
>> device = 'RTL8110SB Single-Chip Gigabit LOM Ethernet Controller'
>> class  = network
>> subclass   = ethernet
>> cap 01[dc] = powerspec 2  supports D0 D1 D2 D3  current D0
>>
>> Also, this machine uses RT8139 (rl0) for its LAN and it works too.
> 
> Glad to hear that. Unfortunately I have no plan or time to merge
> all changes made in re(4)/rl(4) to stable/6 so you may have to
> stick to this unofficial driver. As you already know, I overhauled
> these drivers long time ago and there were too many changes.

Well, given that before busdma commit that hardware worked just fine with stock 
driver,
it could be less overhead for me to rollback that one busdma small chunk :-)
Who knows, which drivers got broken then in 2010 in 6.4-STABLE with busdma 
change besides re(4)...
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: busdma MFC broke ipfw fwd for RELENG_6

[Eugene Grosbein]

18.09.2011 03:22, YongHyeon PYUN пишет:

>>> Glad to hear that. Unfortunately I have no plan or time to merge
>>> all changes made in re(4)/rl(4) to stable/6 so you may have to
>>> stick to this unofficial driver. As you already know, I overhauled
>>> these drivers long time ago and there were too many changes.
>>
>> Well, given that before busdma commit that hardware worked just fine with 
>> stock driver,
>> it could be less overhead for me to rollback that one busdma small chunk :-)
>> Who knows, which drivers got broken then in 2010 in 6.4-STABLE with busdma 
>> change besides re(4)...
> 
> I agree but that MFC may have fixed bugs in other driver which
> correctly used bus_dma(9) KPI.

It is possible. With luck, this legacy 6.4-STABLE system runs rock-stable
and have no other problems with drivers/

Thanks for your drivers backport, anyway, now this case is closed :-)
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: Realtek integrated nic problem

[Eugene Grosbein]

> I have a box using this Realtek nic:
> 
> re0@pci0:1:0:0:   class=0x02 card=0x81051019 chip=0x813610ec rev=0x05
> hdr=0x00
> class  = network
> subclass   = ethernet
> 
> but FreeBSD is unable to use it. Its 8.1 pfSense. Newer versions would run
> it ?
> 
> thanks,
> 
> matheus
> 

Your card is supported with re(4) driver in 8.2-RELEASE and later versions.

Eugene Grosbein

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

FreeBSD IPMI driver problem

[Eugene Grosbein]

Hi!

I use several SuperMicro boxes with intergrated IPMI card.
http://www.supermicro.com/products/system/1U/5016/SYS-5016T-MTF.cfm

FreeBSD 8.2 sometimes hang in the past after panics so I use IPMI's watchdog
and generally it works nice with 5 minute timeout. The card is detected as 
following:

ipmi0:  on isa0
ipmi0: KCS mode found at io 0xca2 alignment 0x1 on isa
ipmi0: IPMI device rev. 1, firmware rev. 1.07, version 2.0
ipmi0: Number of channels 2
ipmi0: Attached watchdog

Sometimes ipmi driver issues "KCS errors" to system logs that I ignore
as they seem harmless. However, one of my boxes suddenly rebooted with watchdog
after following errors written to console:

ipmi0: KCS: Reply address mismatch
ipmi0: KCS error: 01
ipmi0: KCS: Reply address mismatch
ipmi0: KCS error: 01
ipmi0: KCS: Command mismatch
ipmi0: KCS error: 01
ipmi0: KCS: Reply address mismatch
ipmi0: KCS error: 01
ipmi0: KCS: Reply address mismatch
ipmi0: KCS error: 01
ipmi0: KCS: Command mismatch
ipmi0: KCS error: 01
ipmi0: KCS: Reply address mismatch
ipmi0: KCS error: 01
ipmi0: KCS: Reply address mismatch
ipmi0: KCS error: 01
ipmi0: KCS: Reply address mismatch
ipmi0: KCS error: 01
ipmi0: Failed to reset watchdog
ipmi0: KCS: Command mismatch
ipmi0: KCS error: 01

It seems, the driver lost ability to contact IPMI watchdog timer and that was 
the reason of reboot.

What can be done to avoid such resets in the future?

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

Re: backup for /var/db/ports

[Eugene Grosbein]

17.10.2011 10:13, Doug Barton пишет:
> On 10/14/2011 17:48, Oleg Ginzburg wrote:
>> Hi
>>
>> With /etc/periodic/daily/220.backup-pkgdb I would also suggest backing up 
>> /var/db/ports dir
> 
> I'm curious as to the reason for doing this. The options are easy to
> recreate, and not particularly dynamic.

How do you recreate them without backup after N years passed? :-)

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

Re: FYI: 9.0-RELEASE announced...

[Eugene Grosbein]

13.01.2012 06:54, Ken Smith пишет:

> If you downloaded the amd64 and/or i386 DVD images before now you
> might want to check the checksums with the ones posted in the
> release announcement.  The fix to make sysinstall(8) happy about
> installing from the DVD images was the *only* change made to the
> updated images.  The "bad" images were never available via
> Bittorrent so if you got the images that way you wouldn't have
> a bad image.

I've just downloaded FreeBSD-9.0-RELEASE-amd64-disc1.iso using
official torrent tracker linked in original announcement.
I use "transmission" torrent client running under FreeBSD 8.2-STABLE.
MD5 of the resulting image does not match:

MD5 (FreeBSD-9.0-RELEASE-amd64-disc1.iso) = b23ef73412bd50ed62ef8613ca1a4199

I've removed the torrent and downloaded file and tried again,
got the same wrong MD5.

I've tried the image with VirtualBox, virtual machine starts and hangs hard
at "Checksum verification" stage.

Is it my local problem or Bittorrent's image is wrong?

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

Re: FYI: 9.0-RELEASE announced...

[Eugene Grosbein]

14.01.2012 01:05, Eugene Grosbein пишет:
> 13.01.2012 06:54, Ken Smith пишет:
> 
>> If you downloaded the amd64 and/or i386 DVD images before now you
>> might want to check the checksums with the ones posted in the
>> release announcement.  The fix to make sysinstall(8) happy about
>> installing from the DVD images was the *only* change made to the
>> updated images.  The "bad" images were never available via
>> Bittorrent so if you got the images that way you wouldn't have
>> a bad image.
> 
> I've just downloaded FreeBSD-9.0-RELEASE-amd64-disc1.iso using
> official torrent tracker linked in original announcement.
> I use "transmission" torrent client running under FreeBSD 8.2-STABLE.
> MD5 of the resulting image does not match:
> 
> MD5 (FreeBSD-9.0-RELEASE-amd64-disc1.iso) = b23ef73412bd50ed62ef8613ca1a4199

Sorry for noise, that's right MD5.

> I've removed the torrent and downloaded file and tried again,
> got the same wrong MD5.
> 
> I've tried the image with VirtualBox, virtual machine starts and hangs hard
> at "Checksum verification" stage.

It seems, FreeBSD 9.0-RELEASE can install as VirtualBox-4.0.10 Guest
using SAS-emulated disk controller only. In IDE or SATA mode it hangs.

Cc'd to Alexander Motin.

Eugene Grosbein

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

bsdinstall and swap partition

[Eugene Grosbein]

Hi!

Is it possible to create swap partition at the beginning of disk
while installing 9.0-RELEASE? For MBR partitioning, it should occupy
first partition of FreeBSD slice.

sysinstall used to ask a type of created partition and
for swap it choosed 'b' partition automatically but bsdinstall uses
'a' partition for first one despite of its type.

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

Re: FYI: 9.0-RELEASE announced...

[Eugene Grosbein]

>> It seems, FreeBSD 9.0-RELEASE can install as VirtualBox-4.0.10 Guest
>> using SAS-emulated disk controller only. In IDE or SATA mode it hangs.
> 
> FWIW, 9.0-RELEASE i386 installed fine on Virtualbox 4.1.8 hosted on a 
> MacBook Pro:

FreeBSD Ports collection has 4.0.14 only.

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

Re: Floppy disks don't work with FreeBSD 9.0

[Eugene Grosbein]

17.03.2012 20:59, Thomas Laus пишет:
> List:
> 
> I keep archival floppy disk copies of EPROM images on CDROM and occasionally 
> need to DD them back to a floppy disk for my programmer to load.  I recently 
> upgraded two of my shop computers from FreeBSD 7 Stable to FreeBSD 9 Stable 
> with clean installs.  I tried the 'fdformat' command prior to 'dd' of the 
> image file yesterday.  All of my floppy disks showed errors on all tracks.  I 
> used a ^C to abort the command and the PC dropped core.  I used the same 
> disks on my laptop with OpenBSD 5 and everything worked as expected.  I was 
> able to format and diskdupe the image to them.  I looked over the Release 
> Notes for FreeBSD 9 and there wasn't any mention of eliminating floppy disk 
> support on either FreeBSD 8 or 9.  This happened on two different PC's, both 
> running FreeBSD 9.  The hardware isn't suspect because these same two PC's 
> were able to format and dupe images when running FreeBSD 7.

You need to supply more details. dmesg lines concerning fdc at least.

Eugene Grosbein

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

Re: RELENG_8 kernel as of Apr 14 does not boot

[Eugene Grosbein]

16.04.2012 11:26, Alexey Dokuchaev пишет:
> Hi,
> 
> Just update my 8.x kernel sources last weekend, and newly built kernel did
> not boot for me:
> 
> link_elf: symbol mem_range_softc undefined
> KLD file acpi.ko - could not finalize loading
> kernel trap 12 with interrupts disabled
> 
> This is stripped down kernel with everything possible loaded from modules.
> Any ideas?  Did not see any warnings in UPDATING...

Try to add 'device mem' to your kernel configuration.

Eugene Grosbein

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

Re: RELENG_8 kernel as of Apr 14 does not boot

[Eugene Grosbein]

>>> Just update my 8.x kernel sources last weekend, and newly built kernel did
>>> not boot for me:
>>>
>>> link_elf: symbol mem_range_softc undefined
>>> KLD file acpi.ko - could not finalize loading
>>> kernel trap 12 with interrupts disabled
>>
>> Try to add 'device mem' to your kernel configuration.
> 
> :-)
> 
> I explicitly have "nodevice mem" and "nodevice io" in my config.  They are
> being loaded from /boot/loader.conf.  This worked fine for quite a while.
> 
> I will try to have it compiled-in, but would still prefer it fixed, or in
> case it cannot be fixed and mem.ko cannot be loaded separately from now on,
> appropriate entry in UPDATING.

It seems John Baldwin brought dependency of acpi.ko on device mem
4 days ago to RELENG_8 with MFC:

http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/i386/acpica/acpi_wakeup.c#rev1.50.2.3

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

Re: RELENG_8 kernel as of Apr 14 does not boot

[Eugene Grosbein]

> On Monday, April 16, 2012 3:59:09 am Eugene Grosbein wrote:
>>>>> Just update my 8.x kernel sources last weekend, and newly built kernel did
>>>>> not boot for me:
>>>>>
>>>>> link_elf: symbol mem_range_softc undefined
>>>>> KLD file acpi.ko - could not finalize loading
>>>>> kernel trap 12 with interrupts disabled
>>>>
>>>> Try to add 'device mem' to your kernel configuration.
>>>
>>> :-)
>>>
>>> I explicitly have "nodevice mem" and "nodevice io" in my config.  They are
>>> being loaded from /boot/loader.conf.  This worked fine for quite a while.
>>>
>>> I will try to have it compiled-in, but would still prefer it fixed, or in
>>> case it cannot be fixed and mem.ko cannot be loaded separately from now on,
>>> appropriate entry in UPDATING.
>>
>> It seems John Baldwin brought dependency of acpi.ko on device mem
>> 4 days ago to RELENG_8 with MFC:
>>
>> http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/i386/acpica/acpi_wakeup.c#rev1.50.2.3
> 
> Hmm, this has been broken for a long time on HEAD and 9 it seems.  However, 
> there
> you get compile breakage (as acpi is no longer supported as a module in 9+) 
> if you
> try to build a kernel with 'nodevice mem'.
> 
> Hmm, mp_machdep.c also breaks.  That is probably true on i386 as well, and has
> been true even on 7.x.  (That is, you can't use 'nodevice mem' and 'SMP' in 
> the
> same kernel.)
> 
> The simplest fix is to just move mem_range_softc out of mem.ko into the base 
> kernel.
> 
> OTOH, what are you trying to gain by putting mem.ko into a module rather than 
> part of
> the base kernel?  Do you just want no /dev/mem file or are you trying to 
> disable all
> of the MTRR support as well?  It may be that we need to rethink what goes 
> into mem.ko
> and have it only exclude /dev/mem but always leave MTRR support enabled.

I guess, Alexey just tries to make smallest possible kernel just for fun :-)
Or, for PicoBSD case where kernel should be booted from very small media and 
modules from another one.

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

Re: Restricting users from certain privileges

[Eugene Grosbein]

28.04.2012 14:50, Zenny пишет:

>> try sudo from ports, security/sudo
>>
>> cheers,
>>danny
>>
>>
> Thanks Daniel, but sudo gives all (not selective) root privileges to the
> user (admin in my case). So this is not what I am trying to achieve in my
> original post.

Please do study sudo real power :-)
It can give selective privileges per-command,
an d it can also allow one to run some command with some arguments only
and not with others. Or, without any arguments only - as you tune
its sudoers configuration file.

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

Unkillable /sbin/ipfw process

[Eugene Grosbein]

Hi!

I've found easy way to make ipfw(8) to become unkillable even witk kill -9.
It is displayed as "running" and takes all CPU cycles.

Just run the following script with argument 122 for 8.3/i386
or with 121 for 8.3/amd64.

#!/bin/sh

args="add 60001 count ip from any to { "

for i in `jot $1 1`
do
args="${args}127.0.0.$i or "
done
args="${args}127.0.1.1 }";

ipfw delete 60001
echo ipfw $args
ipfw $args
#EOF

After one /sbin/ipfw is stuck in this state,
all others invocations of /sbin/ipfw (including "ipfw show")
add another stuck ipfw process.

See also http://www.freebsd.org/cgi/query-pr.cgi?pr=65961

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

Re: i386 binaries on amd64: ldconfig problems

[Eugene Grosbein]

31.05.2012 16:58, Konstantin Belousov writes:

>> But actually I shouldn't have to use LD_32_LIBRARY_PATH.
>> I mean, it's ldconfig's job to configure the directories for
>> locating the libraries.
>>
>> What is wrong here?
> The library search order is LD_{32}_LIBRARY_PATH, then DT_RPATH from
> the binary, then hints, then /lib:/usr/lib. So if rpath of the binary
> contains /usr/local/lib, you get /usr/local/lib before hints.
> 
> Rtld uses only the search path from the hints file. When a library with
> the matched name found, rtld tries to load it. Regardless of the result
> of the load attempt, further components of the search path list are not
> tried.
> 
> Look at the olvwm binary with readelf and see whether DT_RPATH specifies
> /usr/local/lib.

I've faced exactly same problem. What can be done other to rebuild
of all such 32bit bit binaries to make them work for transition period?
Should libmap32.conf help? It seems it does not.

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

Re: i386 binaries on amd64: ldconfig problems

[Eugene Grosbein]

31.05.2012 18:22, Oliver Fromme пишет:
> Eugene Grosbein wrote:
>  > 31.05.2012 16:58, Konstantin Belousov writes:
>  > > The library search order is LD_{32}_LIBRARY_PATH, then DT_RPATH from
>  > > the binary, then hints, then /lib:/usr/lib. So if rpath of the binary
>  > > contains /usr/local/lib, you get /usr/local/lib before hints.
>  > > 
>  > > Rtld uses only the search path from the hints file. When a library with
>  > > the matched name found, rtld tries to load it. Regardless of the result
>  > > of the load attempt, further components of the search path list are not
>  > > tried.
>  > > 
>  > > Look at the olvwm binary with readelf and see whether DT_RPATH specifies
>  > > /usr/local/lib.
>  > 
>  > I've faced exactly same problem. What can be done other to rebuild
>  > of all such 32bit bit binaries to make them work for transition period?
>  > Should libmap32.conf help? It seems it does not.
> 
> Does LD_32_LIBRARY_PATH not work for you?

I did not know about it. Now I've rebuilt all ports as 64 bit already.
But I will try it for next migration :-)

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

Re: su problem

[Eugene Grosbein]

09.06.2012 19:47, Sami Halabi пишет:
> %su -
> Password:
> load: 0.00  cmd: su 30588 [ttydcd] 0.91r 0.00u 0.00s 0% 2092k

Perpaps, your system had no keyboard attached at boot time;
or for some other reason it booted with /dev/console being serial console
instead of vidconsole. su locks trying to access serial console
that is /dev/ttyd0 by default and has Carrier Detect flag enabled.
Hence, it waits for CD on the first serial port (miserably and hopelessly).

You can check if it's true with "sysctl kern.console" command.
You could ask someone to boot the system with keyboard attached -
no need to type anything, though. The system should detect it
and assingn /dev/ttyv0 as /dev/console instead of /dev/ttyd0.
And "su" won't lock.

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

Re: su problem

[Eugene Grosbein]

10.06.2012 01:45, Sami Halabi пишет:
> Hi,
> 
> %sysctl kern.console
> kern.console: ttyv0,dcons,/dcons,ttyv0,uart,ucom,
> 
> %tail /var/log/messages
> Jun  7 19:54:35 vps16 kernel: Trying to mount root from ufs:/dev/da0s1a
> Jun  7 19:54:36 vps16 kernel: bge0: link state changed to UP
> Jun  7 20:18:04 vps16 kernel: ugen0.2:  at usbus0
> Jun  7 20:18:04 vps16 kernel: ukbd0:  0/0, rev 1.10/2.50, addr 2> on usbus0
> Jun  7 20:18:04 vps16 kernel: kbd2 at ukbd0
> Jun  7 20:18:05 vps16 kernel: uhid0:  0/0, rev 1.10/2.50, addr 2> on usbus0
> Jun  7 20:19:37 vps16 login: ROOT LOGIN (root) ON ttyv1
> Jun  7 20:21:19 vps16 kernel: ugen0.2:  at usbus0
> (disconnected)
> Jun  7 20:21:19 vps16 kernel: ukbd0: at uhub0, port 1, addr 2 (disconnected)
> Jun  7 20:21:19 vps16 kernel: uhid0: at uhub0, port 1, addr 2 (disconnected)
> %
> 
> the system was loaded with keyboard and disconnected later if i understand
> the logs...
> 
> New ideas are appreciated, and thanks in advance,
> Sami

I still believe your problem concerns serial console:
su writes to it, indirectly - it notes root login through syslogd
that writes the message to /dev/console and locks in your case, locking su.

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

Re: su problem

[Eugene Grosbein]

11.06.2012 00:28, Sami Halabi пишет:
> But how do I solve the problem?

First you know for sure if it's true.
Please show output of "grep console /etc/syslog.conf" and "kenv | grep console".
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: su problem

[Eugene Grosbein]

11.06.2012 01:21, Sami Halabi пишет:
> Hi,
> 
> %grep console /etc/syslog.conf
> *.err;kern.warning;auth.notice;mail.crit/dev/console
> # uncomment this to log all writes to /dev/console to /var/log/console.log
> #console.info    
> /var/log/console.log
> %
> 
> %kenv | grep console
> comconsole_speed="9600"
> console="vidconsole"
> %
> 
> does this makes us sure with the problem?

No, this is against my proposition: console="vidconsole"
states that /dev/console is not serial port but VGA output that cannot lock...
Sorry, no other thoughts yet...
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: su problem

[Eugene Grosbein]

10.06.2012 18:57, Sami Halabi пишет:
> Well,in8.1 & 8.2it worked fine for me, in 8.3 i got this problem.

Have you asked someone to power cycle the box?
The problem may go away if it was caused by unusual actions at boot...

Eugene Grosbein

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

Re: FreeBSD 8.3-STABLE #0 r237184 if_tun auto-up upon create.

[Eugene Grosbein]

18.06.2012 09:47, Jason Hellenthal пишет:
> 
> Hoping someone here might be able to clue me in on what I am seeing.
> 
> FreeBSD 8.3-STABLE #0 r237184
> 
> When creating a if_tun(4) device...
> ifconfig tun0 create 
> 
> With these line in rc.conf(5)
> cloned_interfaces="tun0"
> ifconfig_tun0="metric 100 down"
> 
> And sysctl net.link.tun.devfs_cloning=0
> 
> The resulting tun0 device results in state...
> tun0: flags=8011 metric 100 mtu 1500
> 
> 
> After removing the ifconfig_tun0= and running ifconfig create...
> tun0: flags=8010 metric 0 mtu 1500
> 
> 
> Why would the device enter a UP/DOWN state if there is no addressing
> information being fed to it ? and especially with a directive of "down"
> ?

That's because the word "up" is automatically added by system rc-scripts
for ifconfig_XXX lines, so real ifconfig statement in your case is:

ifconfig tun0 metric 100 down up

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

Re: Seeking 6.4 make source for ports

[Eugene Grosbein]

20.06.2012 16:17, Mike Pumford пишет:
> Ronald Klop wrote:
> 
>> Do you have a problem yourself or did you just found some messages on
>> google?
>> I presume you are running 6.3 yourself. Why don't you upgrade the OS to
>> 6.4 or 7+, but you do upgrade the ports on an unsupported OS?
>>
> It is a little mind boggling. Unless required I wouldn't even recompile 
> my ports when moving a system from 6.3 to 6.4. In fact I've got an 
> internal system here that started life as 5.3 BETA and have been 
> upgraded all the way to 7.4 via 6.4 with pretty much no downtime apart 
> from the installworld part of the source upgrade. All the 5.x compiled 
> ports ran fine on the 7.4 system until I got round to recompiling them.

You are not required to recompile ports when upgrading within major release,
e.g. from 6.x to 6.4. There is no one reason to do that. They all will work
just as usual.

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

Re: [CFT] Need Testers for: sysutils/bsdconfig

[Eugene Grosbein]

22.06.2012 10:50, Ron McDowell пишет:

>> Again, thank you very much for testing this new software.
>> P.S. Due to the large codebase comprising bsdconfig, ample precautions 
>> should be taken. I've not noticed any negative behavior in months of usage, 
>> but just be warned.
>>
>> P.P.S. I don't think on subscribed to -stable@, so include me in your 
>> replies.
> I'm one of the coauthors of this code, and I am here on -stable.  As 
> stated, this port will only run on 9.0-RELEASE and later.
> 
> Please give it a try!  Thanks!

I've tried it on installed 9.0-STABLE system.

1. At the moment, the documentation does not install:

Could not install package en-freebsd-doc (pkg_add: unable to fetch
'ftp://ftp.freebsd.org/pub/FreeBSD/ports/amd64/packages-9-stable/Latest/en-freebsd-doc.tbz'
by URL)

2. For system running single gmirror gm0 consisting of 2 whole disks
ada0 and ada1, 'Label allocated disk partitions' offers
only ada0 and ada1 for disklabel editor and not gm0.
And disklabel editor shows empty list of partitions.

3. Similar for FDISK Partition Editor (ada0/ada1 and not gm0 for choise),
but Partition Editor presents whole disk space as 'unused'

Note that current SATA disk prices and quality offers no choice for servers
other that some kind of RAID.

4. Networking Devices configuration presents lagg0 device as .
It should provide descriptive string like 'Link aggregation/failover'.

5. Same for vlan16. For vlan9 is shows right 'IEEE 802.1Q VLAN network 
interface'.
It should work same way for vlan1-vlan4095 interfaces at least.

6. Same for ipfw0 pseudo-interface.

7. Networking Devices configuration does not allow to configure any interface
while there are mounted NFS volumes. Should present a warning only, not 
disallow the operation.
For example, it should be possible to configure new vlan interface while NFS 
mount
uses another vlan.

8. In DNS Nameserver Configuration, it's not clear that one, in fact,
can remove unneeded DNS server through two-step procedure - first try to edit,
then clear the address. It should be more obvious at first.

That's enough for this time :-)

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

Re: [CFT] Need Testers for: sysutils/bsdconfig

[Eugene Grosbein]

22.06.2012 14:37, Devin Teske пишет:

>> 5. Same for vlan16. For vlan9 is shows right 'IEEE 802.1Q VLAN network 
>> interface'.
>> It should work same way for vlan1-vlan4095 interfaces at least.
>>
> 
> I'd like to know if the sysctl MIB's for describing network interfaces is 
> reliable. Maybe I'll keep the static list as a fallback. But yes, you're 
> absolutely right -- I should have supported up to 5 digits even (ifconfig has 
> internal limits of 16-bit unsigned integer for the interface instance-number).
> 
> 
>> 6. Same for ipfw0 pseudo-interface.
>>
> 
> Curious what sysctl says about it.

I do not know what sysctl subtree do you refer to.

>> 7. Networking Devices configuration does not allow to configure any interface
>> while there are mounted NFS volumes. Should present a warning only, not 
>> disallow the operation.
> 
> Did I completely disallow it?

Yes.

> I'll have to re-check -- I thought that I had made it so that you could 
> view/edit the configuration but that the warning says that changes will not 
> become effective until you either reboot or visit the menu again when no NFS 
> mounts are active.
> 
> 
>> For example, it should be possible to configure new vlan interface while NFS 
>> mount
>> uses another clan.
>>
> 
> Do you know of a handy way of determining which NFS mount is using which 
> network interface? And further, is there a handy way of traversing the route 
> path to determine that one interface isn't required as an intermediary 
> transit device? (meaning: can one truly ever know that making a new 
> configuration active on any interface could not potentially drop your entire 
> machine from the net with hung NFS mounts?)
> 
> Many months of testing in the lab produced no less than 6 edge-cases where -- 
> if a network link or route is modified when NFS mounts are active -- the 
> machine can enter an unstable/unusable state.
> 
> So we decided to err on the side of caution when it came to allowing settings 
> to be made-active when NFS mounts are active.
> 
> I'm not against improving the code, but I'm wondering if it wouldn't be safer 
> to stick to disallowing any/all changes from being made-active (while 
> allowing viewing/editing without making-active) when NFS mounts are active.
> 
> NOTE: There are other safe-guards too. For example, if you're logged in via 
> SSH and using X11 forwarding while passing the "-X" flag (to use Xdialog(1)), 
> you are disallowed from making a new hostname active (you can change the 
> hostname, but not make it active) because that would cause the very next 
> iteration of Xdialog(1) to fail due to a surreptitious X authority revocation 
> based on the hostname-change in mid-session.

I'm sure that bsdconfig should emit warnings only but not disallow root to make 
any needed changes.
NFS may use completly unrelated routes/interfaces, X11 may be user over network 
without ssh -X etc.
It's pretty annoying for administrator to fight with tools thinking they know 
better  what root should do.

>> 8. In DNS Nameserver Configuration, it's not clear that one, in fact,
>> can remove unneeded DNS server through two-step procedure - first try to 
>> edit,
>> then clear the address. It should be more obvious at first.
>>
> 
> Can you have a look at "bsdconfig startup_rcconf" and see if that's a better 
> way to go about the deletion-process?
> 
> Or perhaps you're just advocating a helpful message in the text above the 
> menu list that explains how to delete the item? (least amount of work)

Again, just a message.

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

Re: [CFT] Need Testers for: sysutils/bsdconfig

[Eugene Grosbein]

22.06.2012 15:39, Devin Teske wrote:

 5. Same for vlan16. For vlan9 is shows right 'IEEE 802.1Q VLAN network 
 interface'.
 It should work same way for vlan1-vlan4095 interfaces at least.

>>>
>>> I'd like to know if the sysctl MIB's for describing network interfaces is 
>>> reliable. Maybe I'll keep the static list as a fallback. But yes, you're 
>>> absolutely right -- I should have supported up to 5 digits even (ifconfig 
>>> has internal limits of 16-bit unsigned integer for the interface 
>>> instance-number).
>>>
>>>
 6. Same for ipfw0 pseudo-interface.

>>>
>>> Curious what sysctl says about it.
>>
>> I do not know what sysctl subtree do you refer to.
>>
> 
> If you're using em(4) device, try:
> 
> sysctl dev.em.0.%desc
> 
> Otherwise (for example), if using fxp(4), try:
> 
> sysctl dev.fxp.0.%desc
> 
> Or for your vlan:
> 
> sysctl dev.vlan.16.%desc
> 
> And try for your ipfw(4) interface:
> 
> sysctl dev.ipfw.0.%desc
> 
> Are each of those meaningful?
> 
> NOTE: They aren't available unless you have the hardware -- so I can't (for 
> example) try "sysctl dev.fxp.0.%desc" unless I have an fxp0 device displayed 
> in ifconfig(8).

That's driver-dependent. For example, lagg does not presents %desc nor ipfw0
and I suppose pretty many others do not. You could use %desc if it's present and
fall back to internal static list otherwise.

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

Re: NAT issue in 8.1

[Eugene Grosbein]

On Tue, Aug 10, 2010 at 04:33:02PM +0100, Alan Hicks wrote:

> Having upgraded to Release-8.1, there appears to be an issue with 
> network address translation where a newly booted machine fails to setup 
> nat with the error 'ipfw: getsockopt(IP_FW_ADD): Invalid argument'

That's because your kernel (GENERIC?) does not have ipdivert support enabled.
That's not fatal, see later.

> 
> Box has two interfaces em0 and xl0 (Dell PowerEdge 1600SC)
> 
> rc.conf
> ifconfig_xl0="inet 192.168.202.5 netmask 255.255.255.0"
> ifconfig_em0="inet 192.168.0.2 netmask 255.255.255.0"
> defaultrouter="192.168.0.1"
> firewall_enable="YES"
> firewall_type="open"
> firewall_logging="yes"
> natd_enable="YES"
> natd_interface="em0"
> 
> Boot excerpt from console, typed as it does not make it to 
> /var/log/messages so apologies for any typo's
> 
> add net default: gateway 192.168.0.1
> Additional routing options: IP gateway=YES
> Starting devd.
> ipfw2 (+ipv6) initialized, divert loadable, nat loadable, rule-based 
> forwarding disabled, default to deny, logging disabled
> load_dn_sched dn_sched FIFO loaded
> load_dn_sched dn_sched QFQ loaded
> load_dn_sched dn_sched RR loaded
> load_dn_sched dn_sched WF2Q+ loaded
> load_dn_sched dn_sched PRIO loaded
> flushed all rules.
> 00100 allow ip from any to any via lo0
> 00200 deny ip from any to 127.0.0.0/8
> 00300 deny ip from 127.0.0.0/8 to any
> 00400 deny ip from any to ::1
> 00500 deny ip from ::1 to any
> 00600 allow ipv6-icmp from :: to ff02::/16
> 00700 allow ipv6-icmp from fe80::/10 to fe80::/10
> 00800 allow ipv6-icmp from fe80::/10 to ff02::/16
> 00900 allow ipv6-icmp from any to any ip6 icmp6types 1
> 01000 allow ipv6-icmp from any to any ip6 icmp6types 2,135,136
> ipfw: getsockopt(IP_FW_ADD): Invalid argument

At this point, there is still no ipdivert support loaded.

> 65000 allow ip from any to any
> Firewall rules loaded.
> Firewall logging enabled.
> Starting natd.

At this point, kernel module ipdivert.ko is loaded.

> Loading /lib/libalias_cuseeme.so
> Loading /lib/libalias_ftp.so
> Loading /lib/libalias_irc.so
> Loading /lib/libalias_nbt.so
> Loading /lib/libalias_pptp.so
> Loading /lib/libalias_skinny.so
> Loading /lib/libalias_smedia.so
> Aug 10 12:02:53 natd[869]: Aliasing to 192.168.0.2, mtu 1500 bytes
> 
> Although all appears to be ok, machines on the xl0 192.168.202.0/24 
> subnet can't see the internet.  Running /etc/rc.firewall manually fixes 
> the issue.

ipdivert.ko is loaded, hence re-loading firewall rules helps.
 
> 
> The machine was upgraded from 8.0 using unmodified cvs sources using 
> buildkernel, buildworld, installworld, installkernel mergemaster.
> 
> Any help appreciated.

Just load ipdivert.ko earlies by means of loader.conf:

echo 'ipdivert_load="YES"' >> /boot/loader.conf

Alternatively, rebuild kernel with options IPFIREWALL and IPDIVERT

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

Re: /sbin/reboot

[Eugene Grosbein]

On 10.12.2010 11:35, Adam Vande More wrote:

>> Why would you want it to be?   One really shouldn't be running /sbin/reboot
>> directly as part of normal operations.  shutdown does a graceful reboot if
>> and when operators need to perform  reboot.
>>
> 
> AFAIK, the only functional difference between the two is shutdown(8)
> notifies other logged in users of the impending shutdown.  I've used
> reboot(8) for a long time with no ill effects so I'd be interested to hear
> what you meant there.  Since an operator can use shutdown(8) to initiate the
> same shutdown sequence reboot(8) uses, it wouldn't seems to be a security
> based decision.

In short, one should use reboot/halt from single user mode
to avoid calling shutdown command sequences for non-running services.
For multi-user one should use shutdown to not miss these sequences.

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

Re: how to access the SOL (Serial over Lan) port in FreeBSD ?

[Eugene Grosbein]

On 15.12.2010 01:35, Luigi Rizzo wrote:

> i recently bought two motherboards with Intel AMT support,
> a remote management tool which among other things implements
> Serial Over Lan (SOL) -- which seems to be accessible
> via TCP port 16994
> 
> I have enabled the feature in the bios and built a client
> (amtterm, see http://dl.bytesex.org/releases/amtterm/ )
> which does connect to the port, but there is no data.
> 
> Especially, on the FreeBSD side i see no "uart" device
> associated with the SOL port, only uart0 which is the
> physical serial port on board.
> 
> Any idea on how to make the SOL port visible ?
> I am using a stock 8.2-beta kernel for amd64,
> tried to load ipmi but it makes no difference.

Today I successfully tested IPMI 2.0 card integrated within SuperMicro server.
For server side, I've got /dev/ttyu0 as first physical port,
/dev/ttyu1 as second and /dev/ttyu2 for SOL port automatically at boot time.

Open request for /dev/ttyu2 locks waiting for DCD
until I use ipmitool shell at remote workstation and command 'sol activate' in 
the shell.
It gave me clean serial channel and I've successfully got getty login prompt,
logged in, run Midnight Commander (TERM=xterm) without a problem.

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

Re: RFC: Upgrade BIND version in RELENG_7 to BIND 9.6.x

[Eugene Grosbein]

On 18.12.2010 11:41, Doug Barton wrote:

> I am particularly interested in feedback from users with significant DNS
> usage that are still using 9.4, especially if you're using the version
> in the base. I would appreciate it if you could install 9.6 from the
> ports and at minimum run /usr/local/sbin/named-checkconf to see if any
> errors are generated. Of course it would be that much more helpful if
> you could also evaluate BIND 9.6 in operation in your environment.
> 
> Your feedback on the issue of upgrading BIND in RELENG_7 is welcome.
> Sooner is better. :)

I run serveral DNS servers based on FreeBSD+bind.
One of them is FreeBSD 7.3-STABLE and bind-9.4.
It's public primary server for two zones (one forward and one reverse)
and public secondary for several others; plus it's recursive caching DNS server
for users, low-loaded.

I've just updated ports tree and installed dns/bind96.
/usr/local/sbin/named-checkconf shows no messages at all.

I've restarted bind using ports version. All seems OK. I'll write back if 
problems arise.
About upgrading bind version in RELENG_7, I'm for it.

Btw, I run one bind96 server/4.11-STABLE for long time
(and even one bind95/4.11-STABLE) and six 6.11-STABLE servers with stock bind93 
;-)
Never had a problem with them.

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

Re: How to debug a double fault? (Re: Could MSGBUF_SIZE be made a loader tunable?)

[Eugene Grosbein]

On 15.12.2010 16:37, per...@pluto.rain.com wrote:
> Andriy Gapon  wrote:
>> on 06/12/2010 07:20 per...@pluto.rain.com said the following:
>>> Would there be some fundamental problem in changing MSGBUF_SIZE
>>> from a compiled-in constant to a tunable that could be set at the
>>> loader prompt?
>>> I didn't see any obvious downside from examining the 8.1-RELEASE
>>> code ...
>> I also don't immediately see why that wouldn't work.
>> Can you try to come up with a patch?
> 
> I came up with what I think is a start, but it has led to:
> How do I go about debugging a "Fatal double fault"?

You also may checkup the code to not abuse stack.
Kernel threads have not much of free stack space and allocating
increasing length arrays in stack will lead to double fault.

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

Re: Problem with /boot/loader [A new patch]

[Eugene Grosbein]

On Fri, Aug 08, 2008 at 12:49:28PM -0400, John Baldwin wrote:

> My realization this morning is that software interrupts ('int X') in real 
> mode 
> disable interrupts just like hardware interrupts do.  Thus, my patch changes 
> BTX to disable interrupts for both cases 1) and 2) now.  I think this will 
> fix the hangs.  I'm still including the code to explicitly initialize the 
> eflags for user requests to a known-good value.  It still has interrupts 
> enabled which means that case 3) should know always run with interrupts 
> enabled (which is the desired state), but the client can disable interrupts 
> in the eflags in the vm86 structure if desired.
> 
> The updated patch (same URL, new patch) is at 
> http://www.FreeBSD.org/~jhb/patches/btx_hang.patch

Sigh, it does not fix my problem described here:

http://groups.google.ru/group/muc.lists.freebsd.stable/browse_thread/thread/538039f40b469e2a

I've just updated my 7.0-STABLE to latest sources, applied your patch
using "cd /usr/src; patch -p6 < ~/btx_hang.patch", it has applied cleanly.
Then I've rebuilt and reinstalled kernel and world and rebooted.
My problem persists as it was.

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Problem with /boot/loader [A new patch]

[Eugene Grosbein]

On Sat, Aug 09, 2008 at 01:56:42PM +0200, Ulrich Spoerlein wrote:

> > > The updated patch (same URL, new patch) is at 
> > > http://www.FreeBSD.org/~jhb/patches/btx_hang.patch
> > 
> > Sigh, it does not fix my problem described here:
> > 
> > http://groups.google.ru/group/muc.lists.freebsd.stable/browse_thread/thread/538039f40b469e2a
> > 
> > I've just updated my 7.0-STABLE to latest sources, applied your patch
> > using "cd /usr/src; patch -p6 < ~/btx_hang.patch", it has applied cleanly.
> > Then I've rebuilt and reinstalled kernel and world and rebooted.
> > My problem persists as it was.
> 
> I'm not sure about which piece of code you are talking here (boot0,
> boot1, boot2, loader?) But if it's one of the former, you dont need to
> installworld, but install new boot blocks using either fdisk -B or
> bsdlabel -B (or both).

As Subject: says, I'm talking about /boot/loader and its command prompt.

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Problem with /boot/loader [A new patch]

[Eugene Grosbein]

On Sat, Aug 09, 2008 at 03:57:21PM -0400, John Baldwin wrote:

> As you are getting BTX faults that is likely a separate issue.  You will need 
> to get a copy of the BTX fault message somehow.

I've got a message from BTX only once and cannot reproduce it.
However, I get loader hangs almost always while trying to type something
in its loader prompt after a character or two entered.
This machine is my home desktop and I can do any debugging needed, just ask.
I can also use serial console, if needed.

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Problem with /boot/loader [A new patch]

[Eugene Grosbein]

On Sat, Aug 09, 2008 at 05:17:31PM -0400, John Baldwin wrote:

> In addition to my earlier message, it would probably be good to narrow down 
> what breaks the loader for you.  For example, does it work ok over serial and 
> only break on vidconsole?  Also, if you just backout sys/boot/i386/btx to 
> 7.0-release and leave the rest of the sys/boot tree at 7.0-stable, do you get 
> a working loader?

I've just rolled sys/boot/i386/btx back to RELENG_7_0_0_RELEASE
leaving rest of src at 7.0-STABLE (plus your patch) and yes, I've got
working loader!

I'll try to establish serial console and retest with RELENG_7 sources.

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Problem with /boot/loader [A new patch]

[Eugene Grosbein]

On Sat, Aug 09, 2008 at 05:17:31PM -0400, John Baldwin wrote:

> > Sigh, it does not fix my problem described here:
> > http://groups.google.ru/group/muc.lists.freebsd.stable/browse_thread/thread
> >/538039f40b469e2a
> > I've just updated my 7.0-STABLE to latest sources, applied your patch
> > using "cd /usr/src; patch -p6 < ~/btx_hang.patch", it has applied cleanly.
> > Then I've rebuilt and reinstalled kernel and world and rebooted.
> > My problem persists as it was.
> 
> In addition to my earlier message, it would probably be good to narrow down 
> what breaks the loader for you.  For example, does it work ok over serial and 
> only break on vidconsole?

I've established serial console, switched back to 7.0-STABLE sources
plus your patch and found that while vidconsole hangs,
serial console is not affected and command prompt works without a problem
with it.

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Problem with /boot/loader [A new patch]

[Eugene Grosbein]

On Mon, Aug 11, 2008 at 11:31:33AM -0400, John Baldwin wrote:

> > I've just rolled sys/boot/i386/btx back to RELENG_7_0_0_RELEASE
> > leaving rest of src at 7.0-STABLE (plus your patch) and yes, I've got
> > working loader!
> 
> Err, my patch should have failed (well, the btx.S part) if you had a 
> 7.0-RELEASE sys/boot/i386/btx.

I've applied patch first, then replaced sys/boot/i386/btx with 7.0-RELEASE
version.

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Problem with /boot/loader [A new patch]

[Eugene Grosbein]

John Baldwin wrote:

> Ok, I'm at a loss for why the new BTX doesn't work for you.  Unfortunately,
> this sort of thing isn't easy to debug.  If you have firewire (and another
> machine with firewire) then I have some debugging code I used with qemu to
> save a summary of the last request made by the loader to BTX.  That can at
> least indicate which BIOS call is hanging.  From there you can dissassemble
> your BIOS to try to determine if there are any spin loops and see what it is
> waiting on.

Thank you very much for the effort. I have firewire here but not another
machine with it, but I'll try to find one. I'll contact you again
when I'll find it.

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: lagg(4) and failover

[Eugene Grosbein]

On Tue, Aug 12, 2008 at 12:37:15PM +0200, Marian Hettwer wrote:

> I'm using lagg(4) on some of our servers and I'm just wondering how the
> failover is implemented.
> The manpage isn't quite clear:
> 
>  failover Sends and receives traffic only through the master port. 
> If
>   the master port becomes unavailable, the next active port
> is
>   used.  The first interface added is the master port; any
>   interfaces added after that are used as failover devices.
> 
> What is meant by "becomes unavailable"? Is it just the physical link which
> needs to become unavailable to trigger a failover?

Yes. It seems you need lacp protocol described later in the manual.

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Process size.

[Eugene Grosbein]

On Thu, Aug 14, 2008 at 09:30:00AM +0300, Sergey Chumakov wrote:

> $limits
> Resource limits (current):
> ...
>   datasize  1048576 kB
>   stacksize  131072 kB
> 
> How and why is it possible for process to grow up to 1493M and even
> more? I suppose, it will be able to eat all available system memory (was
> killed).

Try to eslablish 'vmemoryuse' limit also.

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Upcoming Releases Schedule...

[Eugene Grosbein]

On Sun, Aug 24, 2008 at 01:45:08AM +0200, Svein Skogen wrote:

> In 25 words or less, what are the major changes in 7.0->7.1 and 6.3->6.4
>  for us end users?

In more words, but pretty interesting:

http://people.freebsd.org/~bmah/relnotes/6-STABLE/relnotes-i386.html
http://people.freebsd.org/~bmah/relnotes/7-STABLE/relnotes.html

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Long delays for USB realbtx boot

[Eugene Grosbein]

On Thu, Sep 11, 2008 at 01:55:01PM +0100, Bruce M Simpson wrote:

> I have hacked NanoBSD locally to deal with generation of images for 
> booting off USB keys. I am using the RELENG_7 branch as real-mode BTX 
> support is necessary to support USB boot.
> 
> During testing I noticed that whilst the images eventually boot, there 
> is an extremely long delay before doing so, between when the BIOS reads 
> the boot sector and when the BTX loader messages appear.
> 
> Test system   Time
> 
> Thinkpad T43  1m 40s
> ASUS Vintage AH-1 1m  7s
> 
> Any ideas why this long delay is occuring?

For me, it helps to include these knobs to Nano config file:

CONF_WORLD='
BOOT_MBR_FLAGS=0x0  
         
BOOT_BOOT1_FLAGS=0x0
...
'

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Long delays for USB realbtx boot

[Eugene Grosbein]

Bruce M Simpson wrote:
> 
> Eugene Grosbein wrote:
> > For me, it helps to include these knobs to Nano config file:
> >
> > CONF_WORLD='
> > BOOT_MBR_FLAGS=0x0
> > BOOT_BOOT1_FLAGS=0x0
> > ...
> > '
> >
> 
> I added this to the CONF_WORLD in my config file. Unfortunately this
> seems to break USB boot completely for me.

Sorry, I missed that you are using USB flash. I use IDE flash only.

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Long delays for USB realbtx boot

[Eugene Grosbein]

On Fri, Sep 12, 2008 at 05:53:39PM +0100, Bruce M. Simpson wrote:

>So it appears the "active" flag was not being set for the first 
> partition, it seems NanoBSD's fdisk script didn't set it; that should 
> probably get patched, as this was the root cause of the USB booting delay.

I also always use boot0 for my NanoBSDs as I need safe way for remote
upgrades, so have two code slices. And I always use these knobs:

NANO_BOOTLOADER="boot/boot0"
NANO_BOOT0CFG="-o packet -s 1 -m 3 -t 36"

May it make a difference?

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

RELENG_7: buildworld failed with MODULES_WITH_WORLD=

[Eugene Grosbein]

Hi!

I've just tried to build NanoBSD from 7.0-STABLE sources
with MODULES_WITH_WORLD knob enabled and it failed.
Note that NanoBSD uses make -j3 by default and I have dualcore system.

===> sys/modules/nfslockd (depend)
@ -> /usr/local/src/sys
machine -> /usr/local/src/sys/i386/include
echo "#define INET6 1" > opt_inet6.h
awk -f @/tools/vnode_if.awk @/kern/vnode_if.src -p
awk -f @/tools/vnode_if.awk @/kern/vnode_if.src -q
awk -f @/tools/vnode_if.awk @/kern/vnode_if.src -h
rm -f .depend
mkdep -f .depend -a   -nostdinc -D_KERNEL -DKLD_MODULE -I. -I@
-I@/contrib/altq /usr/local/src/sys/modules/nfslockd/../../nlm/nlm_advlock.c
/usr/local/src/sys/modules/nfslockd/../../nlm/nlm_prot_clnt.c
/usr/local/src/sys/modules/nfslockd/../../nlm/nlm_prot_impl.c
/usr/local/src/sys/modules/nfslockd/../../nlm/nlm_prot_server.c
/usr/local/src/sys/modules/nfslockd/../../nlm/nlm_prot_svc.c
/usr/local/src/sys/modules/nfslockd/../../nlm/nlm_prot_xdr.c
/usr/local/src/sys/modules/nfslockd/../../nlm/sm_inter_xdr.c
In file included from
/usr/local/src/sys/modules/nfslockd/../../nlm/nlm_advlock.c:48:
@/nfsclient/nfs.h:40:21: error: opt_nfs.h: No such file or directory
In file included from
/usr/local/src/sys/modules/nfslockd/../../nlm/nlm_prot_impl.c:56:
@/nfsclient/nfs.h:40:21: error: opt_nfs.h: No such file or directory
mkdep: compile failed
*** Error code 1

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: RELENG_7: buildworld failed with MODULES_WITH_WORLD=

[Eugene Grosbein]

On Wed, Sep 24, 2008 at 01:27:18AM +0800, Eugene Grosbein wrote:

> I've just tried to build NanoBSD from 7.0-STABLE sources
> with MODULES_WITH_WORLD knob enabled and it failed.
> Note that NanoBSD uses make -j3 by default and I have dualcore system.
> 
> ===> sys/modules/nfslockd (depend)
> @ -> /usr/local/src/sys
> machine -> /usr/local/src/sys/i386/include
> echo "#define INET6 1" > opt_inet6.h
> awk -f @/tools/vnode_if.awk @/kern/vnode_if.src -p
> awk -f @/tools/vnode_if.awk @/kern/vnode_if.src -q
> awk -f @/tools/vnode_if.awk @/kern/vnode_if.src -h
> rm -f .depend
> mkdep -f .depend -a   -nostdinc -D_KERNEL -DKLD_MODULE -I. -I@
> -I@/contrib/altq /usr/local/src/sys/modules/nfslockd/../../nlm/nlm_advlock.c
> /usr/local/src/sys/modules/nfslockd/../../nlm/nlm_prot_clnt.c
> /usr/local/src/sys/modules/nfslockd/../../nlm/nlm_prot_impl.c
> /usr/local/src/sys/modules/nfslockd/../../nlm/nlm_prot_server.c
> /usr/local/src/sys/modules/nfslockd/../../nlm/nlm_prot_svc.c
> /usr/local/src/sys/modules/nfslockd/../../nlm/nlm_prot_xdr.c
> /usr/local/src/sys/modules/nfslockd/../../nlm/sm_inter_xdr.c
> In file included from
> /usr/local/src/sys/modules/nfslockd/../../nlm/nlm_advlock.c:48:
> @/nfsclient/nfs.h:40:21: error: opt_nfs.h: No such file or directory
> In file included from
> /usr/local/src/sys/modules/nfslockd/../../nlm/nlm_prot_impl.c:56:
> @/nfsclient/nfs.h:40:21: error: opt_nfs.h: No such file or directory
> mkdep: compile failed
> *** Error code 1

This is easily repeatable without involving NanoBSD stuff,
just with 'cd /usr/src; make -j3 MODULES_WITH_WORLD=yes buildworld'.

I think this patch should be applied, at least it works for me:

--- sys/modules/nfslockd/Makefile.orig  2008-08-09 16:07:45.0 +0800
+++ sys/modules/nfslockd/Makefile   2008-09-24 02:02:23.0 +0800
@@ -10,7 +10,7 @@
nlm_prot_svc.c \
nlm_prot_xdr.c \
sm_inter_xdr.c
-SRCS+= opt_inet6.h
+SRCS+= opt_inet6.h opt_nfs.h
 
 .if !defined(KERNBUILDDIR)
 NFS_INET6?=1   # 0/1 - requires INET6 to be configured in kernel
@@ -19,6 +19,9 @@
 opt_inet6.h:
echo "#define INET6 1" > ${.TARGET}
 .endif
+
+opt_nfs.h:
+   echo -n > ${.TARGET}
 .endif
 
 .include 
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: RELENG_7: buildworld failed with MODULES_WITH_WORLD=

[Eugene Grosbein]

On Wed, Sep 24, 2008 at 01:27:18AM +0800, Eugene Grosbein wrote:

> I've just tried to build NanoBSD from 7.0-STABLE sources
> with MODULES_WITH_WORLD knob enabled and it failed.
> Note that NanoBSD uses make -j3 by default and I have dualcore system.
> 
> ===> sys/modules/nfslockd (depend)
> @ -> /usr/local/src/sys
> machine -> /usr/local/src/sys/i386/include
> echo "#define INET6 1" > opt_inet6.h
> awk -f @/tools/vnode_if.awk @/kern/vnode_if.src -p
> awk -f @/tools/vnode_if.awk @/kern/vnode_if.src -q
> awk -f @/tools/vnode_if.awk @/kern/vnode_if.src -h
> rm -f .depend
> mkdep -f .depend -a   -nostdinc -D_KERNEL -DKLD_MODULE -I. -I@
> -I@/contrib/altq /usr/local/src/sys/modules/nfslockd/../../nlm/nlm_advlock.c
> /usr/local/src/sys/modules/nfslockd/../../nlm/nlm_prot_clnt.c
> /usr/local/src/sys/modules/nfslockd/../../nlm/nlm_prot_impl.c
> /usr/local/src/sys/modules/nfslockd/../../nlm/nlm_prot_server.c
> /usr/local/src/sys/modules/nfslockd/../../nlm/nlm_prot_svc.c
> /usr/local/src/sys/modules/nfslockd/../../nlm/nlm_prot_xdr.c
> /usr/local/src/sys/modules/nfslockd/../../nlm/sm_inter_xdr.c
> In file included from
> /usr/local/src/sys/modules/nfslockd/../../nlm/nlm_advlock.c:48:
> @/nfsclient/nfs.h:40:21: error: opt_nfs.h: No such file or directory
> In file included from
> /usr/local/src/sys/modules/nfslockd/../../nlm/nlm_prot_impl.c:56:
> @/nfsclient/nfs.h:40:21: error: opt_nfs.h: No such file or directory
> mkdep: compile failed
> *** Error code 1

I see this problem was fixed in HEAD nearly month ago, but not in RELENG_7.
Please perform MFC.

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: UNEXPECTED SOFT UPDATE INCONSISTENCY; RUN fsck MANUALLY

[Eugene Grosbein]

On Mon, Sep 29, 2008 at 03:31:35PM +0200, Miroslav Lachman wrote:

> >Having been bitten by problems in this area more than once, I now always
> >disable background fsck. Having it disabled by default has my vote too.
> Is there any possibility to selectively disable / enable background fsck 
> on specified mount points?
> 
> I can imagine system, where root, /usr, /var and /tmp will be checked by 
> fsck in foreground, but waiting to foreground fsck on data partitions of 
> about 500GB or more (it can take up tens of minutes or "hours") is scary.
> I need server with ssh running up "quickly" after the crash, so I can 
> investigate what the problem was and not just sit and wait tens of 
> minutes "if" machine gets online again or not... answering phone calls 
> of clients in the meantime.

I solve this problem this way: size of /usr is 300Mb or less and it is
mounted read-only, so does not have any problem with fsck (it will
always skip it as clean); write activity on root fs is minimized by moving
objects being modified out of root (/tmp is symlink or other fs,
ntp drift is located on /var, not /etc etc.), background fsck is enabled.

And you always may use /etc/rc.early to force foreground check for /var
even if fsck runs in background later:

#!/bin/sh
/sbin/fsck -p /var || /sbin/fsck -y /var || exit 1

So, if /var was clean then fsck -p skips it, else it checks it in foreground
and marks clean if possible. For serious errors, check runs again trying
its best to clean /var and continue.

Later background fsck skips clean (possibly fixed already) filesystems.

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: du and df don't agree

[Eugene Grosbein]

On Mon, Nov 10, 2008 at 11:01:00AM -0500, Stephen Clark wrote:

> Why would du show 630k used by /tmp while df show 161M used
> by /tmp?
> 
> I have run fstat /tmp and can't find any files that are using
> the space that df is claiming as being used.

You need lsof +aL1 /tmp to see an answer.

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: find -L . -type l -delete

[Eugene Grosbein]

On Thu, Nov 13, 2008 at 03:38:51PM +0200, Andriy Gapon wrote:

> Am I stupid or is our 'find' is seriously broken in one subtle feature?
> 
> $ find -L . -type l
> find all broken symlinks (target doesn't exists)
> 
> $ find -L . -type l -delete
> removes all symlinks!!!

Yes.

> FreeBSD 7.1-PRERELEASE amd64

This is pretty old and known problem:
http://www.freebsd.org/cgi/query-pr.cgi?pr=bin/90687

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: find -L . -type l -delete

[Eugene Grosbein]

On Thu, Nov 13, 2008 at 04:33:22PM +0200, Andriy Gapon wrote:

> >> $ find -L . -type l -delete
> >> removes all symlinks!!!
> >> FreeBSD 7.1-PRERELEASE amd64
> > This is pretty old and known problem:
> > http://www.freebsd.org/cgi/query-pr.cgi?pr=bin/90687
> Wow, that's pretty tough.
> But now I will have to re-tune myself to xargs.

Care to submit a followup to the PR with a note
that it's still the problem for 7.1?

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: ipfw2.c,v 1.76.2.17

[Eugene Grosbein]

On Tue, Dec 02, 2008 at 06:02:16PM +0300, pluknet wrote:

> Since this revision (appeared in 6.3) I think ipfw violates POLA.
> I mean "ipfw table N list" shows values of table in Internet '.' notation.
> 
> A friend of mine was surprised to found Internet representation
> of this "optional 32-bit unsigned value".
> 
> For example security/bruteblock stores unix timestamps here
> and AFAICS there is no possibility to come back to the previous
> output format (other than reverting this revision).

That was fixed 8 months ago. Just upgrade to 6.4.

Eugene Grosbein
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

New snd_hda import and very low mixer volume

[Eugene Grosbein]

Hi!

I've just upgraded from 7.1-PRERELEASE to 7.1-STABLE via source upgrade
and divcovered that my sound stopped working. I was aware about
recent HDA driver update and tried to switch hw.snd.default_unit
back and forth but that did not help.

Finally I've realised that's just mixer values changed their meaning:
now I can't hear anything at levels 50:50 of both 'mixer pcm'
and 'mixer vol' and have to increase values significantly
to make sound: upto 85:85 of both.

Isn't it a bug?
I've Intel D975XBX motherboard with onboard Azalia HDA,
default unit is 0:

$ cat /dev/sndstat
FreeBSD Audio Driver (newpcm: 32bit 2007061600/i386)
Installed devices:
pcm0:  at cad 2 nid 1 on hdac0
[MPSAFE] (1p:2v/1r:1v channels duplex default)
pcm1:  at cad 2 nid 1 on hdac0
[MPSAFE] (1p:1v/0r:0v channels)

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

sysctl lock in RELENG_6

[Eugene Grosbein]

Hi!

I've RELENG_6 system controlling local PBX through RS-232 port, sio(4).
It also runs syslogd, cron, sshd, bsnmpd and sendmail for outgoing reports.

It locks very often: it answers to pings but PBX controlling software stops
responding, local and remote login attempts hang due to 'login' process
stuck in 'sysctl lock' state. Local consoles do switch with 'Alt-Fn'
and DDB works. It shows that sendmail is in 'sysctl lock' state too.

This is NanoBSD installation running from IDE flash, it's swapless
but I think I could manage to obtain crashdump if there is an interest of it.

I've digged commit logs a bit and found this change MFC'd to RELENG_7
but not RELENG_6:

http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/kern/kern_sysctl.c#rev1.177.6.2

It seems RELENG_6 needs this too, doesn't it?
I'm going to merge the change to RELENG_6 and give it a try.

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

Re: sysctl lock in RELENG_6

[Eugene Grosbein]

On Mon, Feb 09, 2009 at 11:48:22PM +0200, Kostik Belousov wrote:

> > I've digged commit logs a bit and found this change MFC'd to RELENG_7
> > but not RELENG_6:
> > 
> > http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/kern/kern_sysctl.c#rev1.177.6.2
> > 
> > It seems RELENG_6 needs this too, doesn't it?
> > I'm going to merge the change to RELENG_6 and give it a try.
> 
> Yes, please give it a try. In fact, it was quite specific situation
> that I observed and produced a fix for. You need execing process that
> needs to grab Giant, e.g. due to image being located on !MPSAFE fs, and
> simultaneous sysctl issued that inspects this process.

Well, my 6.3-STABLE locked very often (sometimes every hour).
Yesterday I've upgraded it to 6.4-STABLE, applied 1.177.2.1 of the file
to sources, rebuilt NanoBSD image and ran it. It has not locked yet.

Btw, root filesystem here is UFS2 without softupdates mounted read-only
but /etc and /var are md(4) (devfs is mounted too).

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

Re: sysctl lock in RELENG_6

[Eugene Grosbein]

> > Well, my 6.3-STABLE locked very often (sometimes every hour).
> > Yesterday I've upgraded it to 6.4-STABLE, applied 1.177.2.1 of the file
> > to sources, rebuilt NanoBSD image and ran it. It has not locked yet.
> Can you be slightly more scientific in your tests ?
> Try RELENG_6 without my patch to see whether it is needed at all.

It locked again today running recent 6.4-STABLE without your patch.
There is a difference - it took more than a week to lock,
6.3-STABLE did this very often.

Now I'll run it with your patch applied.

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

Re: HEADS UP: major CAM ATA MFC

[Eugene Grosbein]

Alexander Motin wrote:

> Feedbacks are welcome as always.

Today's RELENG_8 build is broken (my /usr/src is symlink to /usr/local/src):

# make -j3 MODULES_WITH_WORLD=yes buildworld
[skip]
===> sys/modules/ahci (all)
cc -O2 -pipe -march=prescott -fno-strict-aliasing -Werror -D_KERNEL -DKLD_MODULE
 -nostdinc   -I. -I@ -I@/contrib/altq -finline-limit=8000 --param inline-unit-gr
owth=100 --param large-function-growth=1000 -fno-common  -mno-align-long-strings
 -mpreferred-stack-boundary=2  -mno-mmx -mno-3dnow -mno-sse -mno-sse2 -mno-sse3
-ffreestanding -fstack-protector -std=iso9899:1999 -fstack-protector -Wall -Wred
undant-decls -Wnested-externs -Wstrict-prototypes  -Wmissing-prototypes -Wpointe
r-arith -Winline -Wcast-qual  -Wundef -Wno-pointer-sign -fformat-extensions -c /
usr/local/src/sys/modules/ahci/../../dev/ahci/ahci.c
/usr/local/src/sys/modules/ahci/../../dev/ahci/ahci.c: In function 'ahci_timeout
':
/usr/local/src/sys/modules/ahci/../../dev/ahci/ahci.c:1286: error: 'struct ahci_
channel' has no member named 'fatalerr'
/usr/local/src/sys/modules/ahci/../../dev/ahci/ahci.c: In function 'ahci_end_tra
nsaction':
/usr/local/src/sys/modules/ahci/../../dev/ahci/ahci.c:1368: error: 'struct ahci_
channel' has no member named 'fatalerr'
/usr/local/src/sys/modules/ahci/../../dev/ahci/ahci.c:1384: error: 'struct ahci_
channel' has no member named 'fatalerr'
/usr/local/src/sys/modules/ahci/../../dev/ahci/ahci.c:1396: error: 'struct ahci_
channel' has no member named 'fatalerr'
/usr/local/src/sys/modules/ahci/../../dev/ahci/ahci.c:1405: error: 'struct ahci_
channel' has no member named 'fatalerr'
/usr/local/src/sys/modules/ahci/../../dev/ahci/ahci.c:1447: error: 'struct ahci_
channel' has no member named 'fatalerr'
/usr/local/src/sys/modules/ahci/../../dev/ahci/ahci.c: In function 'ahci_reset':
/usr/local/src/sys/modules/ahci/../../dev/ahci/ahci.c:1703: error: 'struct ahci_
channel' has no member named 'fatalerr'
*** Error code 1

Stop in /usr/local/src/sys/modules/ahci.
*** Error code 1

Stop in /usr/local/src/sys/modules.
*** Error code 1

Stop in /usr/local/src/sys.
*** Error code 1

Stop in /usr/local/src.
*** Error code 1

Stop in /usr/local/src.
*** Error code 1

Stop in /usr/local/src.
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: HEADS UP: major CAM ATA MFC

[Eugene Grosbein]

Jeremy Chadwick wrote:
> I didn't have this problem.  System has AHCI in use, and the kernel is
> built to make use of modular atacore.  Specifically:

[skip]

> I'd recommend you re-csup with a different mirror and see if there are
> any changes.

I'll try. By the way, have you used MODULES_WITH_WORLD while building world?
RELENG_8 still has a couple of other problems with this knob, needed patches 
are:

-- sys/modules/dtrace/lockstat/Makefile.orig2009-09-16 23:05:25.0 
+0800
+++ sys/modules/dtrace/lockstat/Makefile2009-09-16 23:05:45.0 
+0800
@@ -5,7 +5,7 @@
 KMOD=  lockstat
 SRCS=  lockstat.c

-SRCS+= vnode_if.h
+SRCS+= vnode_if.h opt_kdtrace.h

 CFLAGS+=   -I${.CURDIR}/../../../cddl/compat/opensolaris \
-I${.CURDIR}/../../../cddl/contrib/opensolaris/uts/common \

--- sys/modules/sysvipc/sysvmsg/Makefile.orig   2009-09-16 23:04:00.0 
+0800
+++ sys/modules/sysvipc/sysvmsg/Makefile2009-09-16 23:04:12.0 
+0800
@@ -3,6 +3,6 @@
 .PATH: ${.CURDIR}/../../../kern

 KMOD=  sysvmsg
-SRCS=  sysv_msg.c opt_sysvipc.h
+SRCS=  sysv_msg.c opt_sysvipc.h opt_compat.h

 .include 
--- sys/modules/sysvipc/sysvsem/Makefile.orig   2009-09-16 23:02:02.0 
+0800
+++ sys/modules/sysvipc/sysvsem/Makefile2009-09-16 23:01:51.0 
+0800
@@ -3,6 +3,6 @@
 .PATH: ${.CURDIR}/../../../kern

 KMOD=  sysvsem
-SRCS=  sysv_sem.c opt_sysvipc.h
+SRCS=  sysv_sem.c opt_sysvipc.h opt_compat.h

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