04-28-2013 09:20 PM
We have Brocade 10-port 8 Gb SAN Switch Module for IBM BladeCenter which is connected to Core Director (M48)
recently we have been receiving a lot of (833 (SQL Application) errors) from the Blades attached to this Switch module
After running a porterrshow I found that all utilized ports on this switch have large number of enc_out errors (700 - 900) along with loss_sync on only one of the Fabric switch
Also found some c3_discards on ISL ports connecting to Core directors. But these errors are not increasing substantially but they do over a weeks time.
We had this problem before but it was only on some Blades, after replacing the HBA cards from 8G to 4G those errors got disappeared but now its appearing on all the ports which is strange.
Any thoughts or suggestions on this would be much appreciated
I have also attached the output of porterrshow
Thanks in advance
04-29-2013 12:14 AM
Thanks for the reply
Sorry I made a typo its a 20 port 8Gb Switch module for IBM Bladecenter
I have already cleared the stats but they do appear over a period of time (like over a week) not very quickly.
04-30-2013 11:04 PM
Yes the Fillword is set to 3 on all 8GB ports
and you are right the ISL ports are set to 8GB
Now I do see some Bottleneck errors on these ISL ports and c3 discards as well
I have cleared them but they are not increasing
05-01-2013 01:30 PM
I would say clear the stats on the ports where the SQL hosts are connected and see the observe for further enc_out errors. are the enc_out errors increasing on "all utilized ports" still?