How to prevemt tresspassing of luns between SP'

[Ravi1987]

HI,

i am using two cluster datacenter which contains 4.0 esx server . second cluster was introduced recently. both cluster having 4 and 5 esx server ( 4.0)

after the creation of second cluster we are facing problem of tress passingt of luns to a single controller. we are using EMC FC based storage.

all was well and balanced and no tresspassing happened when we are with only single cluster. we took support from emc and they said everything is fine from their side and it is a bug from VMware with ref of KB http://kb.vmware.com/kb/1021814. can any body tell me is their any solution of this problem . or can tresspassing create any mess for environemnt.

[Borja_Mari]

Hi,

maybe this helps maybe not :smileyhappy:

If in the esx host side you are using the round robbin multipathing PSP, then maybe changing it to MRU helps.

But if you are using round robbin and you really want to continue using it, then maybe the ALUA mode must be (correctly) configured in the SAN side.

Regards/Saludos,
Pablo

Please consider awarding

any helpful or corrrect answer. Thanks!! -

Por favor considera premiar

cualquier respuesta útil o correcta. ¡¡Muchas gracias!!

Virtually noob blog

[Ravi1987]

Thanks for your suggestion.

Please do me a favour also if i change the algorithm by gui can it hamper production envirnment.

second if i continue with same settings than what will the possible concerns.

[Borja_Mari]

Hi,

i really don't know your actual configuration.

Anyway changing the PSP shouldn't affect the san connectivity. Modifying the SAN configuration (ALUA) can affect the san connectivity.

Then, IMHO i will make the SAN reconfiguration out from production hours.

Hope this helps :smileyhappy:

Regards/Saludos,
Pablo

Please consider awarding

any helpful or corrrect answer. Thanks!! -

Por favor considera premiar

cualquier respuesta útil o correcta. ¡¡Muchas gracias!!

Virtually noob blog