System is a CS1000M SG on Rls 6.0, current on patches. There is an Avaya bulletin, released March of 2011, about an issue with these cards on this type of system. If you pull the card, or remove power to the card, the card may or may not reboot upon power being restored, or card being re-inserted. The bulletin is correct, as we did have this issue. Does anyone have any information about this issue. I have heard that it is only cards made by a manufacturer called Norvell, but cards made by Emerson do not have an issue. I have heard it is the FMD CF that is the issue, but I thought that was a different bulletin/issue.
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations derfloh on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
NT4N39AAE5 CPPIV Dark Card
- Thread starter pcat200
- Start date
Firebird Scrambler
Systems Engineer
Hello Pcat200
I think this is what you are after below.
CS1000 Compact Flash (RMD) issue on CPPIV Processors
Background:
Issues have been observed on CS1000M single group, multi-group or CS1000E systems with CPPIV processors, in which problems with communication between the CPPIV processor and a FMD or RMD Compact Flash card can cause the system TTY to lockup, disk cache errors, an inability to access PDT, FMDs to go out of service, lose synchronization and may cause the system to go into split mode, do a graceful switchover or INI. This has been seen on a small number of 6.0 and 5.5 systems. This issue is believed not to be applicable to CS1000E systems with CPPM processors.
Analysis:
The ongoing investigation has shown that the system is most prone to these potential conditions when an RMD has just been inserted or a backup is being performed to a newly inserted RMD.
Recommendations:
The investigation is ongoing, however at this time we can provide the following recommendations:
1. Install preventative patch MPLR30442 (CPPIV only). This patch adds additional logic code to ensure that when an RMD is inserted or removed that the ATA BUS is properly reset and that the device is properly mounted and un-mounted without causing the FMD of the CPPIV to go out of service.
2. Install preventative patch MPLR30371. This patch will run a CHECKDISK on the card before doing a backup to it - if bad blocks are found the backup won't occur, thus reducing the likelihood of the problem being encountered and an INI occurring.
3. Install preventative patch MPLR28935 for release 6.0 systems only. This patch addresses a
situation in which the system will spend too much time writing to the FMD when not properly logged into a security domain. This additional time spent by the writing task can cause the system to slow down and/or INI. (Please Note that this patch has Special Instructions, including a manual INI)
4. Install preventative patch MPLR30410. This patch will optimize data block usage in the FMD.
5. Make sure that RMD cards are not inserted if MPLR30442 is not installed. If the system does not currently have the RMD inserted, modify the backup rules to backup the data using FTP. If the RMD is already inserted into the CPU and no issues are being seen on the system, run a CHECKDISK on the
SWITCHED & IPT TECHNICAL SPECIALIST CENTRES
RMD. If the CHECKDISK passes (see below for an example of a passed CHECKDISK),
then the RMD can be left in the faceplate of the CPU and continue to be used for backup purposes.
pdt> chkdsk /cf2
/cf2/ - disk check in progress ...
total # of clusters: 61,848
# of free clusters: 61,448
# of bad clusters: 0
total free space: 122,896 Kb
max contiguous free space: 124,338,176 bytes
# of files: 0
# of folders: 0
total bytes in files: 0
# of lost chains: 0
total bytes in lost chains: 0
All the best
Firebird Scrambler
Meridian 1 / Succession and BCM / Norstar Programmer in the UK
If it's working, then leave it alone!.
I think this is what you are after below.
CS1000 Compact Flash (RMD) issue on CPPIV Processors
Background:
Issues have been observed on CS1000M single group, multi-group or CS1000E systems with CPPIV processors, in which problems with communication between the CPPIV processor and a FMD or RMD Compact Flash card can cause the system TTY to lockup, disk cache errors, an inability to access PDT, FMDs to go out of service, lose synchronization and may cause the system to go into split mode, do a graceful switchover or INI. This has been seen on a small number of 6.0 and 5.5 systems. This issue is believed not to be applicable to CS1000E systems with CPPM processors.
Analysis:
The ongoing investigation has shown that the system is most prone to these potential conditions when an RMD has just been inserted or a backup is being performed to a newly inserted RMD.
Recommendations:
The investigation is ongoing, however at this time we can provide the following recommendations:
1. Install preventative patch MPLR30442 (CPPIV only). This patch adds additional logic code to ensure that when an RMD is inserted or removed that the ATA BUS is properly reset and that the device is properly mounted and un-mounted without causing the FMD of the CPPIV to go out of service.
2. Install preventative patch MPLR30371. This patch will run a CHECKDISK on the card before doing a backup to it - if bad blocks are found the backup won't occur, thus reducing the likelihood of the problem being encountered and an INI occurring.
3. Install preventative patch MPLR28935 for release 6.0 systems only. This patch addresses a
situation in which the system will spend too much time writing to the FMD when not properly logged into a security domain. This additional time spent by the writing task can cause the system to slow down and/or INI. (Please Note that this patch has Special Instructions, including a manual INI)
4. Install preventative patch MPLR30410. This patch will optimize data block usage in the FMD.
5. Make sure that RMD cards are not inserted if MPLR30442 is not installed. If the system does not currently have the RMD inserted, modify the backup rules to backup the data using FTP. If the RMD is already inserted into the CPU and no issues are being seen on the system, run a CHECKDISK on the
SWITCHED & IPT TECHNICAL SPECIALIST CENTRES
RMD. If the CHECKDISK passes (see below for an example of a passed CHECKDISK),
then the RMD can be left in the faceplate of the CPU and continue to be used for backup purposes.
pdt> chkdsk /cf2
/cf2/ - disk check in progress ...
total # of clusters: 61,848
# of free clusters: 61,448
# of bad clusters: 0
total free space: 122,896 Kb
max contiguous free space: 124,338,176 bytes
# of files: 0
# of folders: 0
total bytes in files: 0
# of lost chains: 0
total bytes in lost chains: 0
All the best
Firebird Scrambler
Meridian 1 / Succession and BCM / Norstar Programmer in the UK
If it's working, then leave it alone!.
- Thread starter
- #4
Thanks for the information. Mine did the same as RickyPone's. They sat for 3 hrs during planned downtime, and only CPU 0 rebooted. If the release 4 cards will work then that is the route I have to go. Even if we get a onsite spare of the rel 5, we would not know if it would boot, and if you remove the original card, chances are it would not reboot.
Here is the bulletin:
NT4N39AAE5 CPPIV Dark Card Issue
Background
Avaya has several customer reports of an issue with the NT4N39AAE5 CPP IV processor cards used on CS1000M
Single Group and Multi Group. The symptom that has been experienced on the sites that have seen the issue is that
once the card is installed and functioning correctly, if the card is pulled out of the system and re-inserted immediately or
after a short period, the card does not power up correctly and appears dead. Most cards do not exhibit this issue, but
cards with the problem may have to wait an hour or more before re-inserting for the card to power up correctly and start
functioning again.
Analysis
Investigation by Avaya has revealed that although the majority of sites have never experienced the issue, sites with the problem can
experience it repeatedly. The Avaya labs have since been able to duplicate the problem using cards from the sites experiencing the
issue. The root cause of the issue has been narrowed down to a component within the on board power up circuitry. It has also been
found that a bad card will exhibit the issue in CS1000M systems only. CS1000E systems using NT4N39 cards do not exhibit problems
even with bad NT4N39AAE5 cards. Note that most CS1000E systems do not use the NT4N39AAE5.
The symptoms of the issue occurs as follows. Upon first insertion, the NT4N39AAE5 card power up with no issues. On removal and
re-insertions the card no longer powers up and the faceplate LED does not light up. If the card is left out and allowed to cool and is
then re-inserted the card will power up correctly and will work with no incident. It may take an hour or more to cool the card sufficiently
for re-insertion. Note that the issue is upon power up. If the card is already running there is no issue.
A fix to the issue has been developed and the manufacturer of the NT4N39AAE5 will shortly be implementing the modification on the
power circuit, to eliminate the problem. Also, Avaya is working with the manufacturer to develop a plan to address the field population
of this card.
Recommendation
For NT4N39AAE5 cards already installed, it is recommended not to remove the cards from the shelf unless it is to replace it with
another card.
Corrective Action
Corrective Actions will be communicated via this bulletin when a plan is in place to address the field population. At that time it will also
be announced which release of card from Avaya has the fix.
Here is the bulletin:
NT4N39AAE5 CPPIV Dark Card Issue
Background
Avaya has several customer reports of an issue with the NT4N39AAE5 CPP IV processor cards used on CS1000M
Single Group and Multi Group. The symptom that has been experienced on the sites that have seen the issue is that
once the card is installed and functioning correctly, if the card is pulled out of the system and re-inserted immediately or
after a short period, the card does not power up correctly and appears dead. Most cards do not exhibit this issue, but
cards with the problem may have to wait an hour or more before re-inserting for the card to power up correctly and start
functioning again.
Analysis
Investigation by Avaya has revealed that although the majority of sites have never experienced the issue, sites with the problem can
experience it repeatedly. The Avaya labs have since been able to duplicate the problem using cards from the sites experiencing the
issue. The root cause of the issue has been narrowed down to a component within the on board power up circuitry. It has also been
found that a bad card will exhibit the issue in CS1000M systems only. CS1000E systems using NT4N39 cards do not exhibit problems
even with bad NT4N39AAE5 cards. Note that most CS1000E systems do not use the NT4N39AAE5.
The symptoms of the issue occurs as follows. Upon first insertion, the NT4N39AAE5 card power up with no issues. On removal and
re-insertions the card no longer powers up and the faceplate LED does not light up. If the card is left out and allowed to cool and is
then re-inserted the card will power up correctly and will work with no incident. It may take an hour or more to cool the card sufficiently
for re-insertion. Note that the issue is upon power up. If the card is already running there is no issue.
A fix to the issue has been developed and the manufacturer of the NT4N39AAE5 will shortly be implementing the modification on the
power circuit, to eliminate the problem. Also, Avaya is working with the manufacturer to develop a plan to address the field population
of this card.
Recommendation
For NT4N39AAE5 cards already installed, it is recommended not to remove the cards from the shelf unless it is to replace it with
another card.
Corrective Action
Corrective Actions will be communicated via this bulletin when a plan is in place to address the field population. At that time it will also
be announced which release of card from Avaya has the fix.
Same bulletin I have read. We have some sites with e5 but when I have trouble I put the release 4 in. I had this trouble a year ago before it was a known issue. Had one hospital completely down so I went to another one 40 minutes away. Split it brought an e5 back to the down site and it wouldn't come up. Took it back to the split switch still wouldn't boot. Got some release fours from our warehouse came up pretty as you please. For a while I thought I had a backplane trouble on the original down switch. Thought it was popping any card I plugged in it.
- Status