uptime wrong in 2.5.70

uptime wrong in 2.5.70

[Clemens Schwaighofer]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

I a got a test vmware running with a 2.5.70 and I have sligh "overflow"
with my uptime.

gentoo root # uptime
 22:29:47 up 14667 days, 19:08,  3 users,  load average: 0.00, 0.00, 0.00

I think thats a bit too much ;)

Linux gentoo.tequila.intern 2.5.80 #1 Tue May 27 14:42:51 JST 2003 i686
Intel(R) Penitum(R) 4 CPU 1.60GHz GenuineIntel GNU/Linux
running on a
Gentoo System (unstable tree)

- --
Clemens Schwaighofer - IT Engineer & System Administration
==========================================================
Tequila Japan, 6-17-2 Ginza Chuo-ku, Tokyo 104-8167, JAPAN
Tel: +81-(0)3-3545-7703            Fax: +81-(0)3-3545-7343
http://www.tequila.jp
==========================================================
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQE+6ZA9jBz/yQjBxz8RAgnJAJ4yZTZJuP5QJOZv3Lc9Awnr4sblpQCeOHaD
fgjlR74Svry26Jh+1oBjt6g=
=6rUw
-----END PGP SIGNATURE-----

Re: uptime wrong in 2.5.70

[Tim Schmielau]

> I a got a test vmware running with a 2.5.70 and I have sligh "overflow"
> with my uptime.
>
> gentoo root # uptime
>  22:29:47 up 14667 days, 19:08,  3 users,  load average: 0.00, 0.00, 0.00

Doesn't ring any bell yet. Can you cat /proc/uptime and /proc/stat output?
Is this immediately after booting? Reproducable?

Tim

Re: uptime wrong in 2.5.70

[Alex Goddard]

On Fri, 13 Jun 2003, Tim Schmielau wrote:

> > I a got a test vmware running with a 2.5.70 and I have sligh "overflow"
> > with my uptime.
> >
> > gentoo root # uptime
> >  22:29:47 up 14667 days, 19:08,  3 users,  load average: 0.00, 0.00, 0.00
> 
> Doesn't ring any bell yet. Can you cat /proc/uptime and /proc/stat output?
> Is this immediately after booting? Reproducable?

I had this happen yesterday with a 2.5.70-bkWhatever from about three days
ago.

Immediately after boot my uptime was reported as 14664 days, 13 hours.  I 
didn't do a cat of either /proc/uptime or /proc/stat.  If the bug hits 
again, I'll do so and post the results.

-- 
Alex Goddard
agoddard@purdue.edu

Re: uptime wrong in 2.5.70

[george anzinger]

Clemens Schwaighofer wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Hi,
> 
> I a got a test vmware running with a 2.5.70 and I have sligh "overflow"
> with my uptime.
> 
> gentoo root # uptime
>  22:29:47 up 14667 days, 19:08,  3 users,  load average: 0.00, 0.00, 0.00

Uptime currently reports a conversion of jiffies which is currently 
jacked up to a few seconds short of 32 bits worth of jiffies (for 
testing purposes).

I have a patch pending with Andrew to convert uptime to use the POSIX 
monotonic clock which a) will start at 0 at boot time and b) will 
account for NTP clock adjustments.  Should give an uptime real close 
to the best of watches (or even better) :)

-g
> 
> I think thats a bit too much ;)
> 
> Linux gentoo.tequila.intern 2.5.80 #1 Tue May 27 14:42:51 JST 2003 i686
> Intel(R) Penitum(R) 4 CPU 1.60GHz GenuineIntel GNU/Linux
> running on a
> Gentoo System (unstable tree)
> 
> - --
> Clemens Schwaighofer - IT Engineer & System Administration
> ==========================================================
> Tequila Japan, 6-17-2 Ginza Chuo-ku, Tokyo 104-8167, JAPAN
> Tel: +81-(0)3-3545-7703            Fax: +81-(0)3-3545-7343
> http://www.tequila.jp
> ==========================================================
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.2.1 (MingW32)
> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
> 
> iD8DBQE+6ZA9jBz/yQjBxz8RAgnJAJ4yZTZJuP5QJOZv3Lc9Awnr4sblpQCeOHaD
> fgjlR74Svry26Jh+1oBjt6g=
> =6rUw
> -----END PGP SIGNATURE-----
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
> 

-- 
George Anzinger   george@mvista.com
High-res-timers:  http://sourceforge.net/projects/high-res-timers/
Preemption patch: http://www.kernel.org/pub/linux/kernel/people/rml

Re: uptime wrong in 2.5.70

[john stultz]

On Fri, 2003-06-13 at 13:34, george anzinger wrote:
> Clemens Schwaighofer wrote:
> > I a got a test vmware running with a 2.5.70 and I have sligh "overflow"
> > with my uptime.
> > 
> > gentoo root # uptime
> >  22:29:47 up 14667 days, 19:08,  3 users,  load average: 0.00, 0.00, 0.00
> 
> Uptime currently reports a conversion of jiffies which is currently 
> jacked up to a few seconds short of 32 bits worth of jiffies (for 
> testing purposes).

Any access to jiffies should be subtracting INITIAL_JIFFIES, so uptime
should still work correctly. I've been unable to reproduce this problem,
so if anyone else sees it I'd love to get more info. 

> I have a patch pending with Andrew to convert uptime to use the POSIX 
> monotonic clock which a) will start at 0 at boot time and b) will 
> account for NTP clock adjustments.  Should give an uptime real close 
> to the best of watches (or even better) :)

Sounds interesting.

-john

Re: uptime wrong in 2.5.70

[Tim Schmielau]

On Fri, 13 Jun 2003, george anzinger wrote:

> Clemens Schwaighofer wrote:
> >  22:29:47 up 14667 days, 19:08,  3 users,  load average: 0.00, 0.00, 0.00
>
> Uptime currently reports a conversion of jiffies which is currently
> jacked up to a few seconds short of 32 bits worth of jiffies (for
> testing purposes).
>

That would explain being off by 49 days, 17 hours (or 497 days if
HZ=100).
However, the two reported cases are roughly 295 as much, and I have no
clue about that number.

Tim

Re: uptime wrong in 2.5.70

[Clemens Schwaighofer]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Tim Schmielau wrote:

>>I a got a test vmware running with a 2.5.70 and I have sligh "overflow"
>>with my uptime.
>>
>>gentoo root # uptime
>> 22:29:47 up 14667 days, 19:08,  3 users,  load average: 0.00, 0.00, 0.00
>
>
> Doesn't ring any bell yet. Can you cat /proc/uptime and /proc/stat output?
> Is this immediately after booting? Reproducable?

gentoo root # cat /proc/uptime
1267537132.92 278990.27
gentoo root # cat /proc/stat
cpu  1542550 0 3915839 27412781 392624
cpu0 1542550 0 3915839 27412781 392624
intr 333933393 332637964 7735 0 0 0 0 3 0 2 0 0 0 7417 0 0 16 0 622268
657988 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ctxt 15529955
btime 4083180513
processes 659768
procs_running 3
procs_blocked 0
gentoo root # uptime
 16:59:20 up 14670 days, 13:39,  3 users,  load average: 1.22, 1.04, 0.49

I will reboot the box now and see if it happens again. but as I read in
other postings to this thread it seams to happen again.

thought I got anohter mail, from David Schwartz who claims that this is
because <quote> This is due to a known bug in the Penitum(R) processor,
which Linux has never claimed to support. </quote>

- --
Clemens Schwaighofer - IT Engineer & System Administration
==========================================================
Tequila Japan, 6-17-2 Ginza Chuo-ku, Tokyo 104-8167, JAPAN
Tel: +81-(0)3-3545-7703            Fax: +81-(0)3-3545-7343
http://www.tequila.jp
==========================================================
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQE+7RLyjBz/yQjBxz8RAtLYAJ4rFrLGtCT7UcX3m+oHJSAssJOf9gCfflW7
b0odmjhcJ7AyZeFMnTwjbag=
=B6J3
-----END PGP SIGNATURE-----

uptime wrong in 2.5.72

[Christian Kujau]

hi,


i have read the thread "uptime wrong in 2.5.70", but it did not occur to 
me or i've just not noticed. now i habve 2.5.72 on an Alpha machine, and 
my uptime is wrong too:

evil@lila:~$ uptime
20:39:43 up 12223 days, 18:39,  2 users,  load average: 1.00, 1.15, 1.13
evil@lila:~$
evil@lila:~$ date
Fri Jun 20 20:39:47 CEST 2003
evil@lila:~$

someone asked for /proc/stat and /proc/uptime, so i'll put it in here too:

evil@lila:~$ cat /proc/uptime
1056134456.56 119.85
evil@lila:~$
evil@lila:~$ cat /proc/stat
cpu  19532572 231368852 6161047 86525 35469
cpu0 19532572 231368852 6161047 86525 35469
intr 268431707 0 18 0 0 0 0 0 0 257197217 0 0 264591 3 0 0 10969878
ctxt 5528866
btime 1055883310
processes 251305
procs_running 2
procs_blocked 0
evil@lila:~$


i'll check if it's reproduceable.

Thanks,
Christian.

Re: uptime wrong in 2.5.72

[Christian Kujau]

Christian Kujau schrieb:
> i'll check if it's reproduceable.
> 

yes, it is.