When we performing core blade power cycle like ( poweron and poweroff) some of the suse linux boxes path got dead. Since this is the online activity I want to know is there any parameter we need to check for suse linux before taking power cycle of core blade.
Others servers connected to this SAN switch did not have this problem. Kindly let me know if there is any fixes for same.
core blade activity should not impact the host paths . can you post the `slotshow -m` and confirm which slot you were working with? Also in which slots those Linux servers connected to?
SUSE xen host HBA is conncted to blade 2, Core blade is 5 that we have power cycle
When a core blade is powercycled, all the frames going through it are discarded. This fact may cause that sensitive hosts report errors on some paths. You could check the HBA parameters, since maybe some timeouts are a little bit low and may need to be increased so that actions like rebooting a core blade don't affect them.