[prev in list] [next in list] [prev in thread] [next in thread] 

List:       gfs-devel
Subject:    Re: PC machine is freezed after GFS mount
From:       Ken Preslan <kpreslan () zarniwoop ! com>
Date:       1999-09-07 7:32:34
[Download RAW message or body]

There is yet another update on the FTP site.  Update 5 will fix this problem.
(ftp://ftp.globalfilesystem.org/pub/GFS/linux/gfs-fixes/19990720/)

[A new release is long overdue.  We'll put one up as soon as the recent changes
to GFS get cleaned up a bit more.]

> 
> 
> Thank you for good helps. 
> 
> I didn't know the existence of gfs-fixes directory. So I used these
> patches on kernel 2.2.10. The result is very nice!
> 
> But for kernel 2.2.11(2.2.12) +
> GFS2-19990720-kernel-patches-update1.patch.gz +
> GFS2-19990720-update[1234].patch.gz,
> when I do "dd if=/dev/pool/share of=/dev/null bs=512 count=100" after
> "passemble", following Oops messages is output.
> 
> Unable to handle kernel paging request at virtual address c95a1dec
> current->tss.cr3 = 059bc000, %cr3 = 059bc000
> *pde = 00000000
> Oops: 0002
> CPU:    0
> EIP:    0010:[<c886233b>]
> EFLAGS: 00010046
> eax: c88622a0   ebx: c95a1ccc   ecx: 00000000   edx: 00000000
> esi: 00000246   edi: 00007901   ebp: c561bce4   esp: c561bcd0
> ds: 0018   es: 0018   ss: 0018
> Process dd (pid: 522, process nr: 51, stackpage=c561b000)
> Stack: c028e1d0 00000000 c0129fe5 0000003f 00000400 00000040 c01996c8 00007901
>        c669a9ce c669a9d0 00000002 00000040 00000000 c561be78 c561be78 c5617901
>        c028e1d0 00000400 c012c6e2 00000000 00000040 c561bd78 c59414a0 4013900a
> Call Trace: [<c0129fe5>] [<c01996c8>] [<c012c6e2>] [<c0127892>] [<c01088fc>]
> Code: f0 0f ba ab 20 01 00 00 00 0f 82 3a 24 00 00 8b 93 28 01 00
> 
> 
> At Thu, 2 Sep 1999 09:45:33 +0200 (METDST), <Aneta.Baran@cern.ch> wrote:
> > On Thu, 2 Sep 1999, Yukawa Tatsuaki wrote:
> > 
> > > But when I issue mount command(mount -t gfs /dev/pool/xxx /yyy) for
> > > first machine(other machines isn't set up GFS yet), that machine is
> > > freezed completely. Before I issue mount command, things go well.
> > If you have not yet installed the bug fixes for current GFS release, try
> > to do it now.  There are four patches at
> > ftp://gfs.lcse.umn.edu/pub/GFS/linux/gfs-fixes/19990720/ . The first one
> > should solve your problem, but of course you should install all of them.
> > I had the same problems before applying the patches, but now everything
> > works OK
> > 
> > Hope that helps.
> > 
> > cheers,
> > -- Aneta.
> 
> At Thu, 2 Sep 1999 02:58:05 -0500 (CDT), <kpreslan@zarniwoop.com> wrote:
> > I don't see anything wrong with anything you included in your email.
> > We'll figure it out, though.  Here are some things to try:
> > 
> > 1)  There are four updates to the 19990720 GFS release.  There is also one
> > update to the 19990720 GFS kernel patches.  You can get the updates at:
> > 
> > ftp://ftp.globalfilesystem.org/pub/GFS/linux/gfs-fixes/19990720/
> > 
> > Try applying these new updates and see if it helps any.  Particularly the new
> > kernel patches should help.  As I remember, 2.2.11 and later don't patch
> > cleanly with the old 19990720 patches.  This may be the cause of your problem.
> > 
> > 2)  After you assemble your pool (using passemble), can you read and write
> > from it?  You should be able to do
> > "dd if=/dev/pool/share of=/dev/null bs=512 count=100" and
> > "dd if=/dev/zero of=/dev/pool/share bs 512 count=100" without generating
> > any errors.  You should also do a "dmesg" command to make sure pool isn't
> > printing errors to the console.
> > 
> > 3)  Can you make and use an Ext2 filesystem on top of the pool? 
> > 
> > 4)  You didn't include exactly what you did with mkfs_gfs.  You did something
> > like "mkfs_gfs /dev/pool/share -p dlock-0.6 -t share", right?
> > 
> > 5)  The easiest way to solve this problem may be to set up a serial console.
> > To do this do the following steps:
> > 
> > 	A)  When you compile your kernel, say "y" to "Support for console on
> > 		serial port" under the "Character devices".
> > 	B)  Add the line:  append "console=ttyS0" to the section for your
> > 		kernel in /etc/lilo.conf, rerun lilo, and reboot.
> > 	C)  Connect a serial cable from the first serial port of the machine
> > 		you're testing on to a serial port on another machine.
> > 
> > You should then be able to use the second machine to look at the error
> > messages GFS is probably printing out right before it crashes.  These messages
> > will help us figure out what's wrong.
> > 
> > 
> > Let me know if you still have problems,
> > 
> > Ken 
> 
> ----
> Tatsuaki Yukawa
> Internet Systems Laboratory
> Information & Network Technologies Laboratories
> Sony Corporation
> 
> To unsubscribe from this list: send the line "unsubscribe gfs-devel" in
> the body of a message to majordomo@lcse.umn.edu
> 


To unsubscribe from this list: send the line "unsubscribe gfs-devel" in
the body of a message to majordomo@lcse.umn.edu

[prev in list] [next in list] [prev in thread] [next in thread]