i have spoken to EMC and they advised me to do an upgrade from Release 2.19.300.5.019 (currently running) to Release 02.19.300.5.040 as CX300 to deal with ATA drives in a better way.
when it is done, i will inform you for the results.
have you opened a Service Request with EMC for this issue? i have a feeling that running zerodisk on the unbound atas might clear your problem but without having analyzed spcollects i'd not do that. i am sure EMC is interested in finding the root cause here as well.
yes i have done, EMC adviced me to do an upgrade for the FLARE code after sending them the spcollects. still waiting for the customer to specify a date to do that because as you know, i have to the upgrade while the array is online.
what didi you mean by running the zerodisk on the unbound ATA.
it is an engineering level command and it does exactly what it says, writes 0s to all sectors on the disk. the disk has to be unbound, otherwise FLARE rejects the command. i'm afraid it is not allowed to give out the details for this command here but i guess you could ask your EMC CE to run it for you. still, without knowing the root cause of your problem, i'd not use it and follow the CS suggestions.
If you really wanted to use the command you could look in the knowledgebase. There is a support solution that gives the command, but I would warn that the command is DANGEROUS and you must be very careful not to zero disks containing data.
However I believe that unbinding and then rebinding performs the same function. This is explained in emc143057.
The situation I would consider this most useful is for array decommissioning. Some companies, like mine, require that no data walks out the door.
imho, this command is not so dangerous. FLARE rejects it if one tries to zeromark a disk that is part of a raid group even if there are no lun space on it. but i'd like to know why Rami run into this problem in the first place and will the flare upgrade correct it. NDU to latest release is kind of first thing CS typically asks you to do but it is not always the cure...
Rami1
2 Intern
•
131 Posts
0
February 28th, 2007 23:00
when it is done, i will inform you for the results.
Rami1
2 Intern
•
131 Posts
0
March 7th, 2007 00:00
an_hidden_KB
87 Posts
0
March 7th, 2007 03:00
Rami1
2 Intern
•
131 Posts
0
March 7th, 2007 04:00
what didi you mean by running the zerodisk on the unbound ATA.
an_hidden_KB
87 Posts
0
March 7th, 2007 05:00
it is an engineering level command and it does exactly what it says, writes 0s to all sectors on the disk. the disk has to be unbound, otherwise FLARE rejects the command. i'm afraid it is not allowed to give out the details for this command here but i guess you could ask your EMC CE to run it for you. still, without knowing the root cause of your problem, i'd not use it and follow the CS suggestions.
EKellerman
76 Posts
1
March 8th, 2007 06:00
However I believe that unbinding and then rebinding performs the same function. This is explained in emc143057.
The situation I would consider this most useful is for array decommissioning. Some companies, like mine, require that no data walks out the door.
an_hidden_KB
87 Posts
0
March 8th, 2007 07:00