Legacy Product Discussions

Ontap upgrade to v6.3

johnchang
14,792 Views

Hi,

I know this may be a tough question to ask since the filer F740 and the os ONTAP v6.1 are quite old but we still use these in our test environment and I'm having some problems that I can't get support through netapp as we ran out of support.

 

My problem is that the ONTAP OS no longer boots and I was trying to upgrade to v6.3. I have downloaded the v6.3 on the floppy disks and tried to install but does not work, can anyone tell me how I would go about installing the new version?

39 REPLIES 39

johnchang
8,380 Views

OK, based on the drawing that I just saw, I have 2 FC9 shelves on each F740 filer.

So, In order to execute your recommendation, I would basically power down the 2nd filer along with it's 2 shelves.

Power down the shelf 1 on the first F740 filer.

connect shelf 0 to the filer on both slots 6 & 7? using the input pors ? I don't have terminators unfortunately. And then start the OS installation ?

wstroh
8,384 Views

Close. You'll only want one set of cables going to the FC shelves. You can hook up a secondary path (i.e. the shelves would be connected via both slots 6 and 7) however for purposes of simplifying the loop, let's just use one connection for the time being.

As far as powering down things are concerned, I would recommend powering down both heads and all shelves involved before cabling them up. This will allow for a "cleaner" startup.

johnchang
8,308 Views

Ok,

so if I'm using only one shelf (shelf 0) and one filer, what should the cable connection look like?

Do, I take the FC cable put it in slot 7 of the filer and put the db9 end to the input port of shelf 0 and if so, which side as you know FC9 shelfs have two VEMs. Is that all I need to start installing the new os ?

wstroh
8,308 Views

Yeup.

The DB9 should be connected to the left port (the in port) of the leftmost VEM module if you're looking at it from the back of the shelf. Page 27 of the FC9 guide located here:

http://now.netapp.com/NOW/knowledge/docs/hardware/filer/fcshf92.pdf

shows the setup for multiple FC9 shelves. It will be the same for your setup, just with a single shelf. Then you should be able to floppy boot 6.3 and start the install.

Now, the caveat.

If you still experience loop instability I would try the following:

- Power down the system (shelves and filer)

- remove the 'B' module VEM (the VEM on the right side of the shelf) and swap it with the 'A' module

- Power the system back up

This will let you rule out the module.

If you still have instability, power everything down again and try a different cable.

Next, you would try a different adapter.

Finally, as you only need 2 disks for ONTAP to install (a parity and a data), you would remove every disk from the shelf except for the ones in bays 3 and 4 (count from the right of the shelf starting at bay 0). 3 and 4 are the scsi enclosure services bays for the FC9 shelves, so those are required for the loop to work properly.

johnchang
8,309 Views

Ok,

So, I tried installing v6.1R1 using only the shelf 0 and it did not work - see attachment.

Then I tried only disks 3 & 4 with v6.3. This time it let me into the special menu where I selected option 4 and it also complained, see attachment: v6.3 OS install test.text.

Perhaps the disks are bad?

wstroh
8,309 Views

Tue Aug 19 19:55:36 GMT [net_e0:info]: Ethernet e0: Link up.
Tue Aug 19 19:55:44 GMT [ispfc_init_2:warning]: Resetting Fibre Channel adapter 7.
Tue Aug 19 19:55:58 GMT [scsi.cmd.pastTimeToLive:error]: Device 7.4: request failed after retry #0: cdb 0x1b.
ispfc: 11 loop-initialization events seen in 60 seconds on Fibre Channel adapter 7: indicates loop stability problem.
Tue Aug 19 19:56:43 GMT [ses_admin:error]: Disk 7.4 not detected by Enclosure Services.
Tue Aug 19 19:56:57 GMT [ispfc_init_2:warning]: Resetting Fibre Channel adapter 7.
ispfc: 10 loop-initialization events seen in 60 seconds on Fibre Channel adapter 7: indicates loop stability problem.
Tue Aug 19 19:57:17 GMT [ispfc_init_2:warning]: Resetting Fibre Channel adapter 7.
Tue Aug 19 19:57:36 GMT [scsi.cmd.pastTimeToLive:error]: Device 7.4: request failed after retry #0: cdb 0x1b.
ispfc: 9 loop-initialization events seen in 60 seconds on Fibre Channel adapter 7: indicates loop stability problem.
Tue Aug 19 19:58:21 GMT [ses_admin:error]: Disk 7.4 not detected by Enclosure Services.
Tue Aug 19 19:58:21 GMT [ses_admin:error]: No possible actual addresses found.

7.4 was chatty earlier as well. pull that disk and put it off to the side and let's replace him with someone else, then powercycle everything and try the boot again. If you encounter the same (or close) results, try hooking up a different shelf as shelf0. or you could try the adapter in slot 6. Unfortunately at this point, it's a lot of plug 'n chug to get results that the system is happy with.

johnchang
8,309 Views

Hi, ok so as you know my first 2 attempts didn't work so this time, what I did you use the full shelf loaded with all the disks and used the install floppies for v6.3 and it looked really good:

Got to hte special menu and option 4 worked, to initialize the disks which it did.

Then it gave me the setup menu to set the hostname, ip address etc..... etc..... then at the end it told me to reboot to use the new settings so I did.

But in the end it did work because of this error:


Alpha Open Firmware by FirmWorks
Copyright 1995-2000 FirmWorks, Network Appliance. All Rights Reserved.
Firmware release 2.8_a2

Memory size is 512 MB
Testing SIO
Testing LCD
Probing devices
Testing 512MB
16MB chunks

1 to 16MB

16 to 32MB

32 to 48MB

48 to 64MB

64 to 80MB

80 to 96MB

96 to 112MB

112 to 128MB

128 to 144MB

144 to 160MB

160 to 176MB

176 to 192MB

192 to 208MB

208 to 224MB

224 to 240MB

240 to 256MB

256 to 272MB

272 to 288MB

288 to 304MB

304 to 320MB

320 to 336MB

336 to 352MB

352 to 368MB

368 to 384MB

384 to 400MB

400 to 416MB

416 to 432MB

432 to 448MB

448 to 464MB

464 to 480MB

480 to 496MB

496 to 512MB

Complete
Finding image...
Loading scsi-disk

Can't open boot device


Startup failed

I've attached the logs, not sure what went wrong but this is the furthest I've gotten thus far.

Thank you very much!!!!!!!!!!!!!!!!!!!!!!!!

REALLY REALLY appreciate your help with this.

wstroh
8,309 Views

Hey John!

Alrighty, we're on the home stretch. Ran into the same problem many many times back in the day and it's an easy fix 🙂

What you've done is booted off of floppies. This loads the kernel into memory, but it hasn't been written anywhere. What you'll want to do is once setup has completed, grab the 6.3 system files from the NOW site:

(I'm actually giving you the link to 6.3.3, as that was the best release in the 6.3 family)

http://now.netapp.com/NOW/download/software/ontap/6.3.3/

Scroll to the bottom of that page and click 'continue', then click through the EULA and finally grab the System Files depending on if you're doing the install from Unix (49mb file) or Windows (39mb file)

Follow the install instructions for that here:

http://now.netapp.com/NOW/knowledge/docs/ontap/rel633/html/ontap/upgrade/index.htm

The key point here is you *must* issue a 'download' as this will copy the kernel to the boot blocks, allowing the filer to boot next time without using the boot floppies.

On the home stretch John! 🙂

johnchang
8,310 Views

Will,

Thanks so much for you guidance.

Few more questions before I start the system file installation. I will be using my laptop that is running windows.

1. Do I need to boot of the floppies again to start this procedure as we have no working OS on the filer ?

2. After running the setup off the floppies, will I be able to connect to the filer_ipaddress:/etc from my windows ?

3. So, I just double click the 633_setup_a.exe file and that is it?

My BIG question is how do I do these:

Step

Action

1

Log in to your client as Administrator or an account that has full control on the filer C$ share.

2

Map a drive to the C$ directory of your filer according to the procedure for your version of Windows.

once again thanks soooooooooooooo much Will.....

wstroh
8,663 Views

Hi John!

>>> 1. Do I need to boot of the floppies again to start this procedure as we have no working OS on the filer ?

Yes. Right now we have no boot blocks, so you'll need to use them one more time going into this final operation.

>>> 2. After running the setup off the floppies, will I be able to connect to the filer_ipaddress:/etc from my windows ?

Sort of. After setup you'll have to setup cifs. This is very infrastructure specific, so the instructions I'm including below are for setting it up in the most generalized sense. I ran through it on my 960 and it worked, so it's had at least a bit of sanity checking 🙂

>>> 3. So, I just double click the 633_setup_a.exe file and that is it?

A bit more intensive than that I'm afraid,

Once the basic setup has completed (hostname, IP address, etc), you'll need to license CIFS and run 'cifs setup'. The install I ran was under 7.2.4, but it should be relatively similar (it hasn't changed much in the six years I've been using it). In step 2 below, I choose 'workgroup mode' when choosing authentication. This is going to be infrastructure dependent to your site. Choosing 'workgroup' will allow you to use the rest of the flow below to install the system files off of your laptop. For now, I would recommend going that route as you can always change it later by rerunning 'cifs setup'.


1) License cifs.

filer> license cifs=abcdefghi
A cifs license has been installed.
filer> Wed Aug 20 12:43:14 EDT [filer: rc:notice]: cifs licensed
filer>


2) Run cifs setup to enable cifs.

filer> cifs setup
This process will enable CIFS access to the filer from a Windows(R) system.
Use "?" for help at any prompt and Ctrl-C to exit without committing changes.

Your filer does not have WINS configured and is visible only to
clients on the same subnet.
Do you want to make the system visible via WINS? [n]:
This filer is currently configured as a multiprotocol filer.
Would you like to reconfigure this filer to be an NTFS-only filer? [n]:
The default name for this CIFS server is 'filer'.
Would you like to change this name? [n]:
Data ONTAP CIFS services support four styles of user authentication.
Choose the one from the list below that best suits your situation.

(1) Active Directory domain authentication (Active Directory domains only)
(2) Windows NT 4 domain authentication (Windows NT or Active Directory domains)
(3) Windows Workgroup authentication using the filer's local user accounts
(4) /etc/passwd and/or NIS/LDAP authentication

Selection (1-4)? [1]: 3
What is the name of the Workgroup? [WORKGROUP]:
CIFS - Starting SMB protocol...
Welcome to the WORKGROUP Windows(R) workgroup

CIFS local server is running.
filer>


filer> Wed Aug 20 12:46:03 EDT [filer: nbt.nbns.registrationComplete:info]: NBT: All CIFS name registrations have completed for the local server.



3) Verify that a share has been created for C$


filer> cifs shares
Name Mount Point Description
---- ----------- -----------
ETC$ /etc Remote Administration
BUILTIN\Administrators / Full Control
HOME /vol/vol0/home Default Share
everyone / Full Control
C$ / Remote Administration
BUILTIN\Administrators / Full Control
filer>


4) Map to a *drive letter* on your Windows machine using 'administrator' as your username and root's password as the 'password'


5) Verify via the net use command

Y:\>net use
New connections will not be remembered.

Status Local Remote Network
------------------------------------------------------------------------------
OK Y: \\192.168.75.10\C$ Microsoft Windows Network



6) Install system files

Double click on the 6.3.3 setup.exe file

This will bring up a WinZip self-extractor box stating that you'll need to have C$ mapped as administrator.

Click 'ok'

It will prompt you with <drive>:\etc

Replace <drive> with the drive letter you mapped C$ to. In the example above, I replace it with 'Y' so the box will read:

Unzip to folder: Y:\etc

This will unzip the system files to the /etc directory of the filer. Upon completion it may ask you if you wish to run inst_perm.bat. Click 'yes' (this will set the correct permissions on the files in /etc should they have been overwritten.

Once this operation has completed (takes roughly a minute), log into the filer and issue 'download'. This will load the boot blocks on the FCAL disks with the kernel.


filer> download

download: Reminder: upgrade both the nodes in the Cluster
download: You can cancel this operation by hitting Ctrl-C in the next 6 seconds.
download: Depending on system load, it may take many minutes
download: to complete this operation. Until it finishes, you will
download: not be able to use the console.
Wed Aug 20 12:53:47 EDT [filer: download.request:notice]: Operator requested download initiated

download: Downloading boot device
..........................................
download: Downloading boot device (Service Area)
......................................................
download: Downloading disks: .....done
download: boot blocks successfully copied to disk(s)

filer> Wed Aug 20 12:58:40 EDT [filer: download.requestDone:notice]: Operator requested download completed


7) Reboot

Once the above is complete, it is safe to reboot the filer

filer> reboot

The filer should boot back up running 6.3.3.

johnchang
8,664 Views

Will,

Thaks for such detailed procedure but here is one problem that keeps me from moving forward. Just like yesterday, I booted from the floppies and then setup the filer as below:

netapp1toc> setup

NetApp Release 6.3: Wed Aug 7 23:04:57 PDT 2002

System ID: 0016807404 (netapp1toc); partner ID: 0016807718 ()



slot 0: System Board

Processors: 1

Memory Size: 512 MB

slot 0: Fibre Channel Host Adapter 0a

slot 0: SCSI Host Adapter 0b

slot 0: Ethernet Controller e0

MAC Address: 00:a0:98:00:94:c8 (auto-100tx-fd-up)

slot 3: FastEnet-10/100 Quad Ethernet Controller

e3a MAC Address: 00:20:fc:1e:5d:30 (auto-unknown-cfg_down)

e3b MAC Address: 00:20:fc:1e:5d:31 (auto-unknown-cfg_down)

e3c MAC Address: 00:20:fc:1e:5d:32 (auto-unknown-cfg_down)

e3d MAC Address: 00:20:fc:1e:5d:33 (auto-unknown-cfg_down)

slot 6: Fibre Channel Host Adapter 6

7 Disks: 241.5GB

1 shelf with VEM

slot 7: Fibre Channel Host Adapter 7

slot 8: Interconnect adapter

slot 9: NVRAM

Memory Size: 32 MB

Please enter the new hostname [netapp1toc]:

Do you want to configure virtual network interfaces? [n]: y

Number of virtual interfaces to configure? [0] 1

Name of virtual interface #1 []: vif1

Is vif1 a single [s] or multi [m] virtual interface? [m] s

Number of links for vif1? [0] 4

Name of link #1 for vif1 []: e3a

Name of link #2 for vif1 []: e3b

Name of link #3 for vif1 []: e3c

Name of link #4 for vif1 []: e3d

Please enter the IP address for Network Interface vif1 []: 57.4.26.209

Please enter the netmask for Network Interface vif1 [255.0.0.0]: 255.255.255.0

Should virtual interface vif1 take over a partner virtual interface during failover? [n]: n

Please enter media type for vif1 {100tx-fd, tp-fd, 100tx, tp, auto} auto:

Please enter the IP address for Network Interface e0 []: 10.3.26.209

Please enter the netmask for Network Interface e0 [255.0.0.0]: 255.255.255.0

Should interface e0 take over a partner IP address during failover? [n]:

Please enter media type for e0 {100tx-fd, tp-fd, 100tx, tp, auto} auto:

Would you like to continue setup through the web interface? [n]:

Please enter the name or IP address of the default gateway: 57.4.26.1

The administration host is given root access to the filer's

/etc files for system administration. To allow /etc root access

to all NFS clients enter RETURN below.

Please enter the name or IP address of the administration host: 57.4.26.221

Where is the filer located? []: P3_testbed

Do you want to run DNS resolver? [n]:

Do you want to run NIS client? [n]:

The root volume currently contains 2 disks; you may add more

disks to it later using the "vol add" command.

Now type 'reboot' for changes to take effect.

So, I reboot but then I can't access this filer through either the e0 interface or the vif1 interface

Question: Do the ip interfaces become available immediately AFTER the setup has been finished or does the filer need a reboot ?

I think once I can ping/mount /vol/vol0 , I should be able to move on but I'm stuck here.

wstroh
7,479 Views

Hey John!

Yeah, you don't want to reboot yet. Those changes have taken effect, but only in memory. You should be able to ping/mount at this point in the game, and be able to carry out the rest of the procedure.

johnchang
7,479 Views

I see,ok then I guess I'll have to boot from the floppis again. any hoots, it looks like using the linux nfs is going to be easier.

Did my setup look correct? As I was not able to ping the filer.

johnchang
7,479 Views

I have tried to enable CIFS not it looks like it didn't work. How can I enable NFS so i can mount /vol/vol0 to the local unix server and then execute the script that way? I can telnet into the system from the admin host and my exportfs looks like this:

/vol/vol0/home -root=10.3.26.230
/vol/vol0 -access=10.3.26.230,root=10.3.26.230

This is what the options look like:

auditlog.enable on

auditlog.max_file_size 10000000

autologout.console.enable on

autologout.console.timeout 60

autologout.telnet.enable on

autologout.telnet.timeout 60

autosupport.cifs.verbose off

autosupport.content complete

autosupport.dafs.verbose off

autosupport.doit DONT

autosupport.enable on

autosupport.from postmaster

autosupport.mailhost mailhost

autosupport.noteto

autosupport.retry.count 15

autosupport.retry.interval 4m

autosupport.support.enable on

autosupport.support.proxy

autosupport.support.to autosupport@netapp.com

autosupport.support.transport http

autosupport.support.url support.netapp.com/asupprod/post/1.0/postAsup

autosupport.throttle on

autosupport.to

backup.log.enable on

cf.takeover.on_failure on

cf.takeover.on_panic off

cifs.audit.enable off

cifs.audit.file_access_events.enable on

cifs.audit.logon_events.enable on

cifs.audit.logsize 524288

cifs.audit.saveas /etc/log/adtlog.evt

cifs.bypass_traverse_checking on

cifs.comment

cifs.guest_account

cifs.home_dir

cifs.home_dir_namestyle

cifs.idle_timeout 1800

cifs.max_mpx 50

cifs.netbios_aliases

cifs.netbios_over_tcp.enable on

cifs.nfs_root_ignore_acl off

cifs.oplocks.enable on

cifs.oplocks.opendelta 8

cifs.per_client_stats.enable off

cifs.perm_check_ro_del_ok off

cifs.perm_check_use_gid on

cifs.restrict_anonymous.enable off

cifs.save_case on

cifs.scopeid

cifs.search_domains

cifs.show_snapshot off

cifs.shutdown_msg_level 2

cifs.sidcache.enable on

cifs.sidcache.lifetime 1440

cifs.snapshot_file_folding.enable off

cifs.symlinks.cycleguard on

cifs.symlinks.enable on

cifs.trace_login off

cksum_offload.gbeII off

console.encoding nfs

coredump.cf.dump_partner_nvram off

dafs.auth_none.enable off

dafs.checksums.enable off

dafs.default_gid 100

dafs.default_uid 100

dafs.enable off

dafs.idle_timeout 7200

dafs.max_disconnected_sessions 32

dafs.max_pending_requests_server 512

dafs.max_request_size 69632

dafs.max_requests 50

dafs.max_requests_server 1024

dafs.max_response_size 69632

dafs.num_requestd 2

dns.domainname

dns.enable off

fcp.enable off

httpd.access legacy

httpd.admin.access legacy

httpd.admin.enable on

httpd.autoindex.enable on

httpd.enable off

httpd.log.format common

httpd.log.max_file_size 2147483647

httpd.rootdir XXX

httpd.timeout 300

httpd.timewait.enable off

ip.fastpath.enable on

ip.match_any_ifaddr on

ip.path_mtu_discovery.enable on

ip.tcp.sack.enable on

iscsi.enable off

locking.grace_lease_seconds 45

ndmpd.preferred_interface disable

nfs.mount_rootonly on

nfs.per_client_stats.enable off

nfs.require_valid_mapped_uid off

nfs.tcp.enable on

nfs.udp.xfersize 32768

nfs.v2.df_2gb_lim off

nfs.v3.enable on

nfs.webnfs.enable off

nfs.webnfs.rootdir XXX

nfs.webnfs.rootdir.set off

nis.domainname

nis.enable off

nis.group_update.enable off

nis.group_update_schedule 24

nis.servers *

pcnfsd.enable off

pcnfsd.umask 22

raid.reconstruct.perf_impact medium

raid.resync.perf_impact medium

raid.scrub.duration 360

raid.scrub.enable on

raid.scrub.perf_impact low

raid.timeout 24

raid.verify.perf_impact low

rmc.setup off

rsh.access legacy

rsh.enable on

security.passwd.rules.enable on

snapmirror.access legacy

snapmirror.checkip.enable off

snapmirror.delayed_acks.enable on

snapmirror.enable off

snapmirror.log.enable on

snapvault.access none

snapvault.enable off

snmp.access legacy

snmp.enable off

tape.persistent_reservations off

telnet.access legacy

telnet.enable on

tftpd.enable off

tftpd.logging off

tftpd.rootdir /etc/tftpboot

timed.enable off

timed.log off

timed.max_skew 30m

timed.min_skew 50

timed.proto ntp

timed.sched hourly

timed.servers

trusted.hosts *

vol.copy.throttle 10

wafl.default_nt_user

wafl.default_security_style unix

wafl.default_unix_user pcuser

wafl.nt_admin_priv_map_to_root on

wafl.root_only_chown on

wafl.wcc_minutes_valid 20

Do I need to enable NFS and if so where or how can I get the license?

netapp1toc> license nfs=ABCDEFGHI

invalid service -- nfs not licensed

We are close, please get me over the hump.....

wstroh
7,479 Views

You can do it over NFS as well (you'd need to download the 6.3.3 tarball instead) if you'd like. 'ABCDEFGHI' is not a valid license code - it's simply one I'm using for purposes of example. You'd need to use whatever license codes you have for the 740.

johnchang
7,479 Views

So I can just use the NFS license from my production Fas270 for installation purposes? Wouldn't I need this license for later when I want to use it to exports the filesystems ?

I actually got the opportunity to talk to a tech from netapp and he told me I can use ftp to transfer the file but when I went to do "options ftp" I didn't see this option.

Any thoughts? oh I did see tftp though....

johnchang
7,480 Views

Will,

It looks like the filer finally works now, thank you for all your hard work but I still have some questions left over.

1. The web filerview tells me I'm only using 2 disks, whata do I need to do to get the filer to see ALL the disks on the shelf.

2. How do I add the second shelf ?

Once again, you have been a life saver.....

wstroh
7,480 Views

Hi John -

Glad to hear you're up and running. To add the rest of the disks (I assume they are spares) you would simply issue a 'vol add <volname> <number of disks>'. For example, if you have five other disks on the shelf, if you issue 'vol add vol0 5' it will add the other five disks.

You can see the state of the disks (spares or otherwise) by issuing a 'vol status -r'.

While you can add the other shelf hot, I would recommend powering down the system, cabling up the second shelf (using the out port of the VEM on shelf 1 cabled to the in port on the VEM of shelf2), set the shelf id of the second shelf, and then power on both shelves, wait approx 30 seconds, and power on the head.

johnchang
7,480 Views

Ok, when I got in this morning, I noticed that I could no longer access the filer. So, when I went to physically see the filer,it said " A disk has failed" and netapp did not fully boot.

I took the laptop and captured some info:



RAID Disk Device HA SHELF BAY CHAN Used (MB/blks) Phys (MB/blks)

--------- ------------ --------------- ---- -------------- --------------

parity FAILED N/A 34500/70656000

data FAILED N/A 34500/70656000

data FAILED N/A 34500/70656000

data FAILED N/A 34500/70656000

data 6.6 6 0 6 FC:A 34500/70656000 35003/71687368

data FAILED N/A 34500/70656000

data FAILED N/A 34500/70656000

data FAILED N/A 34500/70656000

data FAILED N/A 34500/70656000

data FAILED N/A 34500/70656000

data FAILED N/A 34500/70656000

data FAILED N/A 34500/70656000

this doesn't look good.

I've also attached rest of the log.

Public