Jace 403 Broke :(
"A temporary shipping hold on all WEB-403 and WEB-545 extended memory controllers is in place effective immediately.
Recently, it was discovered that the NAND FLASH chip used in some WEB-403-AX-E and WEB-545-AX-E units is showing signs that it is losing pre-programmed data and causing trouble during system start-up."
Well, we just received this lovely message this morning and have a had a Jace 403 failure shortly there after, go figure....
This 3.2 Jace was in service for approximately one month, never rebooted, never any problems. We had a tech stop by to replace the existing station, with one that had some updated graphics. No modules, points, devices, no nothing was added, he was running the correct platform daemon and workbench.
He used the station copier to make the switch, selected "all files" and made the switch. The Jace rebooted and seems to have failed during start up. The red heartbeat lights blinks twice and then turns solid. He is unable to access the platform or ping the IP.
We have tried hard rebooting multiple times, with 5 minute power interruptions and no luck, same thing.
My question(s) is this related? is there a fix? what serial numbers/date codes are effected? would an attempt at going in via the shell mode, even be worth it (my guy on-site does not have the cable)? OR is this just a fluke and this thing is toast?
I don't think HW even has a handle on it yet according to the email. Fortunatley I only have 2 extendeds out there.
Intelli-Building = Less Stress, commissioned with diligence!
I just heard about that today, supposedly they have narrowed the effected shipments to November and after. Sounds like you have a winner.
After speaking with some inside guys, the NAND Flash chip in question is with a date code of 720?, 7 for the year and the 20 represents the week. This is effecting all vendors of the Jace 4 series, however Honeywell thus far is the only one to let us know about the issue officially, kudos to them. We deal with HW, Distech, Delta, JCI, etc. and have a ton of these things out there. We have had some unexplained failures and now believe this would have been the cause.
Apparently what happens is when ever a whatever, a program object, a px, a history, a alarm, a anything that accesses that particular bad block of the flash, its curtains. Things could run fine for months, it reboots, the whatever hits that bad block and toast.
Hopefully soon we get some date codes, not that it matters much for the ones already in service; but we do have some getting ready to be commissioned and would like to know ahead of time. I know there working on it and we should know more soon.
Looking for any and all updates,,,,,
Private or otherwise.
I have a ton of these things out there, getting ready for a commish and on the way?
Email address is in profile.
I think I have the information you are looking for. I had an email from Staefa which included a very specific picture. The text of the email is below, I can't figure out how to post the pic so if anyone needs it post your email.
We have recently received word from our supplier (Tridium, Inc.) of the shipping hold for Expanded Memory AX – Network Managers.
As announced on February 21, Tridium elected to place temporary shipping holds on all units listed above. At that time, we believed there might be an issue that could cause some units containing a specific batch of NAND Flash chips to lose their pre-programmed data and trigger trouble during system start-up.
After an in-depth evaluation by manufacturing and engineering teams, we have confirmed that this is indeed the case. We have determined that a limited supply of NAND Flash chips from one of our suppliers are defective and may cause this condition to occur in some units. The supplier of the flash chip has acknowledged the problem and corrected it on all their current and future builds.
We have concluded this is an isolated case and relates only to the flash memory chip that has a date code of “720 and contains the mark B” as a designator on the identification tag. Please use the following diagrams for reference:
Last edited by cacontrolgirl; 03-10-2008 at 04:48 PM.
Reason: used the word email instead of the word "post"
Received that same e-mail. I'll try to post up the images. We had the exact same issues when we tried rebooting our 403 expanded memory JACE. Red LED flashes twice then stays on solid. We left it completely unplugged for about 15 minutes then plugged and unplugged until it came back up, (usually about 5 times). In the meantime, filed an RMA with Staefa and getting it shipped out ASAP. Definitely check any of yours that are having problems. Good luck!
I received one more update with a little more information regarding the production dates, etc. :
Units containing the problematic memory chip were shipped from November, 2007 through February, 2008. The memory contained in this chip is required for the unit to be fully functional because it notifies the unit how to start. Without the memory, the unit is inoperable and will not start past the attempt to load the flash memory. If you had a computer connected to the serial port with the jumper inserted to place the unit in serial mode and attempted to boot; you would see the following message dumped to the serial port:
Press <ctrl-c> to stop autoboot...
Loading image from on-board nand flash