NetApp System Manager 1.1 - Problem: after upgrade from 1.0.1 to 1.1 and password change

[zinovik_igor]

Hello.

I'm having problems with NSM 1.1.  I administer a cluster pair of FAS 3020c.

First i tried to install NSM 1.0.1 and it installed and was functioning correctly.

It was a bit burden to type root password each time NSM starts, so i typed

root password and saved credentials.  After some time i uninstalled NSM 1.0.1,

changed my root password on filer and installed NSM 1.1.  Right now i'm not

able to login to my filer.  NSM even do not shows me a dialog box where i can

type username and password.  I unselected checkbox

"Enable saving storage system credentials for future sessions" and

this does not helped.  FilerView is enabled and functions normally.

In NSM documentation i read that it stores credentials in encrypted file

on disk.  Maybe i should just find that file and delete it?

Here is how my filer is configured:

fas1> snmp
contact:

location:
        room 142
authtrap:
        0
init:
        0
traphosts:
community:
        ro public

fas1> options httpd
httpd.access                 legacy
httpd.admin.access           legacy
httpd.admin.enable           on
httpd.admin.hostsequiv.enable off
httpd.admin.max_connections  512
httpd.admin.ssl.enable       on
httpd.admin.top-page.authentication on
httpd.autoindex.enable       off
httpd.bypass_traverse_checking off
httpd.enable                 on
httpd.ipv6.enable            off
httpd.log.format             common     (value might be overwritten in takeover)
httpd.method.trace.enable    off
httpd.rootdir                XXX
httpd.timeout                300        (value might be overwritten in takeover)
httpd.timewait.enable        off        (value might be overwritten in takeover)

fas1> secureadmin status
ssh2    - active
ssh1    - inactive
ssl     - active

=== CLIENT ===
Version:      1.1
Build:        1.1.0.459
User:         GK150-04\zinovik
Host:         gk150-04 (172.20.20.216)
Date:         30.03.2010 14:04
OS Version:   5.1
OS Name:      Microsoft Windows XP
Service Pack: Service Pack 3
.NET Version: v2.0.50727
Memory Used:  33 208,00 KB
Memory Total: 524 288,00 KB

=== ERROR ===
NetApp.Library.UserControls.Error.DefaultErrorDetails

=== MESSAGE ===
API invoke failed.  Remote server returned error: (503) Server unavailable.

=== DETAILS ===
API invoke failed.  Remote server returned error: (503)  Server unavailable.

=== CORRECTIVE ACTION ===
No corrective action found

=== STACK TRACE ===
   в NetApp.Ontapi.NaServer.send(Object req)
   в NetApp.Ontapi.NaServer.Invoke(Object req, Type resultType)
   в NetApp.Ontapi.NaApi`1.Invoke(INaServer server)
   в NetApp.SSM.StorageSystem.StorageSystem.InvokeOntapApi[T](NaApi`1 api, Boolean isAsync)
   в NetApp.SSM.StorageSystem.StorageSystem.InvokeONTAPIReal[T](NaApi`1 api, Boolean isAsync, Boolean showLoginDialog)
   в NetApp.SSM.StorageSystem.StorageSystem.VerifyCredentials(Boolean showLoginDialogIfNecessary)
   в NetApp.SSM.StorageSystem.StorageSystem.VerifyCredentials()

=== INNER STACK TRACE ===
   в System.Net.HttpWebRequest.GetRequestStream()
   в NetApp.Ontapi.NaServer.send(Object req)

[rle]

I see that you have a stack trace.  What did you do to get the stack trace?

I assume that after NSM 1.1 was installed that NSM 1.1 found your cluster and is in the left side navigation frame.  What happens when you click on the cluster?

   - Rick -

[zinovik_igor]

If i click on cluster a message box appers with message with two buttons:

"API invoke failed.  Remote server returned error (503).  Server unavailable"

[nagendrk]

Have you tried the following:

1. Remove the storage system from System Manager

2. Close System Manager

3. Restart System Manager and add the storage system

Does this help?

[zinovik_igor]

No

[computersupport]

Check your options trusted.hosts on the filer itself. I had this EXACT same problem and found that I was accessing it from a machine not in the list of trusted hosts, thus throwing the 503 "host not available" error.

[rle]

Did modifying trusted.hosts work?  When httpd.admin.access is set to "legacy", options trusted.hosts is used.

   - Rick -

[zinovik_igor]

fas1> options httpd.admin
httpd.admin.access           legacy
httpd.admin.enable           on
httpd.admin.hostsequiv.enable off
httpd.admin.max_connections  512
httpd.admin.ssl.enable       on
httpd.admin.top-page.authentication on
fas1> options trusted
trusted.hosts                172.20.21.17,172.20.20.216 (same value required in local+partner)

I work on computer with IP address 172.20.20.216

I can successfully access FilerView from my computer.

[vital_bsatu_by · Accepted solution]

I had the same problem. The cause was in HTTP proxy server configuration on the management computer. The exclusion list contained *.localdomainname but not the IP addresses of the filer. So if you open the filer web interface in a browser, you type the filer's name and it connects directly. On the other hand SM1.1 connects by IP address regardless of what you have typed in the dialog box when you added your filer to the SM. So if you enter the filer's name it still resolves it's IP address then tries to establish the SSL connection. As the IP address is not listed in the proxy exception list, it connects via your default proxy server using HTTP CONNECT method. Because your proxy server is obviously not in the trusted.hosts list you get rejected.

So you need to add the IP adresses of the filer to the exception list at Control Panel->Internet Options->Connections->LAN Settings->Advanced->Exceptions.

[zinovik_igor]

Yes.  It fixes problem.  I just disabled proxy in MSIE and NSM work again.  Thanks a lot!