Unsolved
This post is more than 5 years old
1 Rookie
•
14 Posts
0
2784
June 13th, 2008 08:00
AX150 has an unmanageable storage processor.
I inherited four AX150's from my predecessor. I managed to log into two of them. When Navisphere Express launches, the first thing I see is 2 big orange warnings.
The first one reads: "Navisphere Express can no longer manage the other processor (peer) in this storage system because of a software failure."
The second reads: "The write cache is currently disabled. The write cache is disabled when faulty components are present in the system. Verify that all components in this storage system are working properly."
The paragraph goes on to say that I should reboot the SP. In this case, the SP that doesn't work is SPA. I can reboot it, but the fault returns.
Both of the SPs have normal LED activity while booting. Once booted, neither SP's has a fault light. In Navisphere, SPA displays the status "Unmanaged."
Since the error message said "faulted components" I did some hardware troubleshooting. I swapped the memory and the IO modules, pulled both cmos batteries, and performed a configuration reset. The system came back up in the same condition.
I'm starting to believe this might be normal behavior for a brand new system? What I'm seeing is a configuration problem right? Please tell me I'm making a dopey mistake!
The first one reads: "Navisphere Express can no longer manage the other processor (peer) in this storage system because of a software failure."
The second reads: "The write cache is currently disabled. The write cache is disabled when faulty components are present in the system. Verify that all components in this storage system are working properly."
The paragraph goes on to say that I should reboot the SP. In this case, the SP that doesn't work is SPA. I can reboot it, but the fault returns.
Both of the SPs have normal LED activity while booting. Once booted, neither SP's has a fault light. In Navisphere, SPA displays the status "Unmanaged."
Since the error message said "faulted components" I did some hardware troubleshooting. I swapped the memory and the IO modules, pulled both cmos batteries, and performed a configuration reset. The system came back up in the same condition.
I'm starting to believe this might be normal behavior for a brand new system? What I'm seeing is a configuration problem right? Please tell me I'm making a dopey mistake!
0 events found
No Events found!


brianjohnson1
1 Rookie
•
14 Posts
0
June 14th, 2008 01:00
It appears it has never been used. I utilized the storage system initialization wizard to find the thing, and it had an IP address of 1.2.1.1, with no user name or password assigned. It has no raid groups or luns assigned. It's not connected to any network. The only connection is the ethernet connection between SPB and my laptop.
If I had to guess, I would say the guy who purchased these arrays and installed them couldn't figure out this error, and then abandoned them. We can't send them back to EMC because they've touched classified data, and EMC doesn't have field service out here. (Maybe I should hit up EMC for a job)
Do you have any advice on how I could collect a boot log? If not, do you have any other ideas?
GearoidG
251 Posts
0
June 14th, 2008 01:00
It may be necessary for a CE to go onsite to gather boot logs of the SP that is going unmanaged.
Does the SP thats going unmanaged own any luns and if so do your hosts have access to them?
Gearoid
DGM3
238 Posts
0
June 14th, 2008 21:00
We'll have someone follow-up with you.
Regards,
DGM
brianjohnson1
1 Rookie
•
14 Posts
0
June 16th, 2008 01:00
RRR
6 Operator
•
5.7K Posts
0
June 16th, 2008 02:00
bertog
61 Posts
0
June 16th, 2008 07:00
Both SPA and SPB need to have their Management ports plugged into a network so that they will show up as "Managed" in Navisphere Express. Also, when a system is in an "Uninitialized" state, the IP address of SPA and SPB will be set to 1.1.1.1 and 1.2.1.1, and there will be no username or password. This can all be configured using the Storage System Initialization Wizard, which needs to be run from a Windows client with IP connectivity and on the same physical subnet as the management ports of both SPA and SPB.
brianjohnson1
1 Rookie
•
14 Posts
0
June 16th, 2008 08:00
When you plug both SPA and SPB into a switch, the error clears. Evidently the processors considered themselves "unmanaged" until this happens, even when you are managing the SP with a direct connection.
What threw me off were the error messages. They said "software failure" and "faulty components." If the error message said "check your network connections." I would have been okay.
EMC tells me they may modify the error message. What a great company!
I'm currently working on some major SAN projects, so I'm sure to regale you all with additional follies in the days to come. Thanks everyone for your support.
RRR
6 Operator
•
5.7K Posts
0
June 16th, 2008 23:00
But indeed: an active ethernet connection between both SP's is mandatory for the Clariion.
I knew that.... Really.