Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 17 Mar 2004 08:11:55 +0100
From:      "Poul-Henning Kamp" <phk@phk.freebsd.dk>
To:        Don Bowman <don@sandvine.com>
Cc:        "'current@freebsd.org'" <current@freebsd.org>
Subject:   Re: equivalent of kern.timecounter.method? 
Message-ID:  <4414.1079507515@critter.freebsd.dk>
In-Reply-To: Your message of "Tue, 16 Mar 2004 22:04:32 EST." <FE045D4D9F7AED4CBFF1B3B813C85337045D85C7@mail.sandvine.com> 

next in thread | previous in thread | raw e-mail | index | archive | help

Can you check that both the periodic interrupts actually fire ?

critter phk> vmstat -i
interrupt                          total       rate
irq0: clk                       85613519         99	<<<
irq1: atkbd0                     1222031          1
irq6: fdc0                             2          0
irq7: ppc0                             1          0
irq8: rtc                      109586318        127	<<<
irq9: acpi0                         1638          0


In message <FE045D4D9F7AED4CBFF1B3B813C85337045D85C7@mail.sandvine.com>, Don Bo
wman writes:
>
>in 4.x there is a sysctl kern.timecounter.method.
>If this isn't set to '1', the system doesn't work
>correctly (at least with SMP), you get microuptime()
>backwards, and problems with processes hanging
>while time catches up, etc.
>
>What is the equivalent of this sysctl in 5.x?
>
>kern.timecounter.nbinuptime: 6810682
>kern.timecounter.nnanouptime: 4
>kern.timecounter.nmicrouptime: 2218
>kern.timecounter.nbintime: 7091
>kern.timecounter.nnanotime: 172
>kern.timecounter.nmicrotime: 6919
>kern.timecounter.ngetbinuptime: 0
>kern.timecounter.ngetnanouptime: 490
>kern.timecounter.ngetmicrouptime: 30909
>kern.timecounter.ngetbintime: 0
>kern.timecounter.ngetnanotime: 0
>kern.timecounter.ngetmicrotime: 187117
>kern.timecounter.nsetclock: 4
>kern.timecounter.hardware: ACPI-fast
>kern.timecounter.choice: TSC(-100) ACPI-fast(1000) i8254(0) dummy(-1000000)
>kern.timecounter.tick: 1
>kern.timecounter.smp_tsc: 0
>
>is what i have.
>
>I'm having a problem with this machine sometimes hanging...
>sometimes for 5-10s, sometimes permanently (?). When hung,
>everything (serial console, etc) is locked. This includes
>the key-sequence to drop to db on the serial.
>
>I'm not sure the time and the hang are related, but there's
>a suspicious ntpd message just before the hang,
>kernel time discipline status [2041, 2040, etc]
>
>5.2.1-RELEASE-p1 (RELENG_5_2) is what its running, with the (absolutely
>critical ufs_vnops.c fix of 1.235).
>
>The only thing the machine runs is postgresql.
>
>Suggestions on how to debug this lockup?
>
>_______________________________________________
>freebsd-current@freebsd.org mailing list
>http://lists.freebsd.org/mailman/listinfo/freebsd-current
>To unsubscribe, send any mail to "freebsd-current-unsubscribe@freebsd.org"
>

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4414.1079507515>