Hi,
I've been
spying in the group, and thought you could use a bit of expertise here (feel
free to forward the info back into toasters, though).
It's not
specifically a missing file - this happens when the wrong ONTAP Archive was
used for the type of Head you have.
This mistake is often made
when you swap the head for a newer model, or get a P-release and only get a
list of archive files with no description of the head linked to which archive.
When you get a full
release (select your Filer from the download page, then the ONTAP version),
the download table will show you properly which archives belong to which
series.
There are
three types of partially compatible processors in use: a) intel, b) amd, c)
multi-core AMD
You can
run the lesser on the more recent, since the processors are compatible, but
will get that error.
For example, FAS3020 and
FAS3050 use the _e archive (ELF, AMD), while FAS3040, FAS3070
and
the FAS3100 and FAS6000
series use the _q archive (AMD multi-core),
the _i archive (Intel) is
only for old Filers (FAS900 series, F800 series),
the _a archive (older
versions, Alpha F700 series), and
The _m archive for
FAS250/FAS270 (MIPS processor).
The latter two will never run
on other processors, of course.
Fix:
Just get the right .exe
archive into /etc/software (create the directory if you don't already have
it),
then run "software install
<archive_name>", "download" and "reboot" to get the kernel on
the
CF Card refreshed as well and
boot into the right one.
In a cluster you do the stuff
until the reboot on both heads, then do a takeover and giveback
in
both directions (this is
essential for SAN environments). If your clsuter is CIFS only, then
two
reboots under "cf disable"
causes less interruptions, though.
Since you keep the same
ONTAP version (just a different processor kernel), you should have
a
quick reboot (no ESH/AT-FCX
updates).
Kind
regards,
--
Olaf Leimann
Escalation Engineer, EMEA Technical Support Center
Toll Free Technical Support (EMEA): 00800-44-638277
My working hours are: 10:30h to 19:00h CET
Get answers NOW! - NetApp On the Web: http://now.netapp.com
Want
to speed up case resolution?
Download the new Remote Support Diagnostics
Tool.
Learn how at: http://now.netapp.com/NOW/download/tools/rsa/
Everyone at NetApp is committed to the highest level of customer
satisfaction.
If at any time you feel that you are less than very
satisfied with our support,
please contact my manager Tola Soyode (tolas@netapp.com) at
+31-6-46375165.
Sent: maandag 8 juni
2009 20:20
To: Learmonth, Peter
Cc: toasters@mathworks.com
Subject: Re: Unable to open
function name/address mapping file
Thanks all for the replies, however, I had checked my oldest (2
weeks old) snapshot before posting here, I couldn't see that directory in the
snapshot either.
The filer is due for an upgrade in about 2 weeks
time, just want to know whether this(missing file/directory) will affect the
normal functioning of the system, so far nobody complained.
I am the only
one who manages the filers here and I have not made any changes except for
setting "nfs.mountd.trace" to "on" for a brief period last week.
The
system seems to be working fine and the commands ndmpd,ndmpcopy are available,
though I don't use them currently.
On Mon, Jun 8, 2009 at 11:00 PM, Learmonth, Peter
<Peter.Learmonth@netapp.com> wrote:
Hi
LOhit
Sounds
like the filer was zeroed and set up without extracting all the supporting
files, or booted from a new kernel that doesn't match the ONTAP files
installed, or somebody deleted some stuff under /etc. You can check
/etc/.snapshot to see if/when /etc/boot last existed. You need to
refer to the upgrade docs and do a "software install" or use NFS or CIFS to
extract the appropriate setup.exe / tarball.
You may
also notice certain commands are missing, like ndmpd, ndmcopy and a handful
of others (do ? from the command line and see if those commands are
there).
Make
sure the version of ONTAP you extract matches the current running
version. Do a version and version -b and make sure they all
match.
Share
and enjoy!
Peter
Hello All,
I have been getting this warning message on the
console since past 4 days. Have you seen this before by any chance?
I
am running 7.2.3 on a 3020 and no changes were made in the recent past, this
message starting appearing all of a sudden. BTW, there is no directory
called "boot" in "/etc".
Thu Jun
4 23:50:00 GMT [mgr.stack.openFail:warning]: Unable to open function
name/address mapping file /etc/boot/mapfile_7.2.3.L: No such file or
directory--
LOhit
--
LOhit