12-06-2018 06:25 AM
System X 3100 M5 (5457) loaded with Windows Server 2016 and has been running fine since installation 14 months ago. I RDP'ed into the system yesterday and Windows was prompting for a restart, so I allowed that. It took about 15 minutes for the machine to shutdown and so I assume that it installed outstanding Windows Updates. But the system never restarted. No matter what I do, the console displays (0x 2) IMM STATUS: IMM MEMORY INITIALIZED and does nothing further. The USB keyboard does not respond to CAPS LOCK or NUM LOCK and I can't contact the IMM2.
I spent most of the day trying to recover. I removed add-in card (ServerRAID) and all but one DIMM. I attempted to boot using the backup UEFI firmware and the console confirms the backup is being used, but still hangs at (0x 2) IMM STATUS: IMM MEMORY INITIALIZED. I've reset CMOS several times and still no change. And yes, I've pulled the power cords for at least 30 seconds on every attempt.
There are no error indicators lit anywhere on the system board. The IMM2 power and heartbeat indicators appear normal.
I should be able to access IMM2 as long as the power is connected and I allow enough time for IMM2 to boot, regardless of whether the system is powered up or not. If I could get into IMM2, then I should be able to flash new UEFI firmware.
I attempted to open case #407F9PJ with Technical Support, but they won't even talk to me because the system is beyond the one-year warranty. That's even though our company is a Lenovo System X authorized reseller.
What else should I try?
Thanks in advance for any suggestions...
12-06-2018 06:49 PM
Can the system boot into OS or bios after that error?
If not, suggest to reseat all the DIMMs and boot up with the minimize configurtaion.
If the server boots into OS with the minumize configurtaion, suggest to update all the firmware/uEFI to the latest version.
If the problem persist after updated code, please contact hardware support to replace the system board.
12-10-2018 05:39 AM
Thank you for your reply and the suggestions. Answers to your questions:
"Can the system boot to OS or BIOS after that error?" There is no further activity after that message and so it does not boot into OS or BIOS. Sometimes, if I leave the system setting long enough, the watchdog timer will reset the system and the cycle repeats. But it always ends up displaying " IMM MEMORY INITIALIZED" and goes no further.
"If not, suggest to reseat all the DIMMs and boot up with the minimize configuration." I have tried that and there is no effect. I've even tried starting the system with NO DIMMs installed, with each of the two DIMMs installed seperately and with both DIMMs installed. The result is the same.
"If the server boots into OS with the minumize configuration, suggest to update all the firmware/uEFI to the latest version." It doesn't boot and the IMM2 never starts. That means you can't use IMM2 to load fresh firmware.
"If the problem persist after updated code, please contact hardware support to replace the system board." I did attempt to contact technical support, but was never permitted to speak to a technician even though our company is an authorized Lenovo Partner and System x Reseller.
The system in question was installed in September 2017 and so is 2-3 months beyond the one-year warranty. The ONLY choice for technical upport appears to be for us to pay up-front for an on-site technician (technical support claims they do not provide telephone assistance for out-of-warranty systems, period.). I am not willing to pay for an on-site visit to have the tech inform me the system board must be replaced - I already know that.
As I have researched this further, I see that this is a common problem that many users have experienced across the range of System x models. The "workaround" that Lenovo proposes is "not install Windows Updates". That is not a viable workaround because it is impractical. Further, if I advised a customer to "not install Windows Updates" and the customer then had a breach or failure resulting from that advice, our company could be liable for damages.
In my opinion, the root cause of these problems is poor design. That root cause is compounded by Lenovo's unwillingness to develop a fix to the BIOS and some way to deploy that fix. Both problems are further compounded by Lenovo's policy to not support System X products out of warranty - even if Lenovo understands that these products were defective when shipped and failed due to poor design.
So we have a server that we spent thousands on (including DIMMs, ServeRAID, redundant fans and power supplies, IMM advanced license, drives, etc.), have it arrive defective out-of-the-box, have the hidden defects become apparent 14 months after installation and then have Lenovo refuse to provide any assistance. I would not expect many of those customers to purchase a new System X server as a replacement.
Thanks again for your reply and suggestions.