[DRBD-user] Request for help with DRBD-0.7

Philipp Reisner philipp.reisner at linbit.com
Tue May 11 15:34:27 CEST 2004

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.


On Tuesday 11 May 2004 12:49, Ron O'Hara wrote:
> Ok .. but here's some bad feedback ... I just took the cvs version
> (which is labeled 0.7-pre7) and it has the same problem I mentioned a
> couple of days ago where the 2nd box locks up solid as soon as it
> contacts the primary.  This time I've written down all the trace back
> that I can see on the screen: (without the hex offsets)
>
> in_atomic (): 1  irqs_disabled (): 0
> call trace
>
>    __might_sleep +0xab/0xc9
>    printk
>    profile_exit_task
>    do_exit
>    do_page_fault
>    do_divide_error
>    do_page_fault
>    reschedule_interrupt
>    inet_recevmsg
>    sock_recvmsg
>    do_page_fault
>    error_code
>    __wake_up_common
>    __wake_up
>    _drbd_process_ee
>    default_wake_function
>    drbd_process_ee
>    drbd_asender
>    drbd_thread_setup +0x79
>    drbd_thread_setup +0x00
>    kernel_thread_helper
>
> Interesting pieces in this puzzle (since timing is at the heart of all
> good comedy)
> - Gigabit ethernet connection ... things hapen quickly here
> - bonded ethernet interface (slightly slower ... but not much)
> - linux 2.6.4 stock kernel
> - 1Gb ram
>
> Is this enough to go by?  The machine is still sitting there with the
> offsets on the screen and the fault is 100% reproducible.

Hmm, 

In case you have a serial cable null-modem-cable, you can use this,
a terminal program (=minicom) and the kernel parameter 
"console=tty0 console=ttyS0,57600" to catch OOPSes...

Regarding the OOPS. By now I do not understand what is going on
here... 

there is a call to wake_up() in _drbd_process_ee. But why should 
a page-fault happen within this wake_up() call ???

While looking at the description of your setup, I thought that
you might be the first one who tested DRBD-0.7 on a machine
that actually uses HIGHMEM ? 

Could you try to reproduce this with HIGHMEM disabled ?

-Philipp
-- 
: Dipl-Ing Philipp Reisner                      Tel +43-1-8178292-50 :
: LINBIT Information Technologies GmbH          Fax +43-1-8178292-82 :
: Schönbrunnerstr 244, 1120 Vienna, Austria    http://www.linbit.com :



More information about the drbd-user mailing list