02-20-2017 08:31 AM
This problem has been raise on couple of ocassions but never, I think, to the extent I have seen so far, that would fairly easily help to explain what pparticular combination of errros seen on a port would mean.
I have following situation, where port is seen in switchshow, but booting it into ESX effectively render it unusable (its a blade in Blade Center H, one of 14, and only this one is casuing issues, generating high volume of errors on counters).
:admin> porterrshow 12 frames enc crc crc too too bad enc disc link loss loss frjt fbsy c3timeout pcs tx rx in err g_eof shrt long eof out c3 fail sync sig tx rx err 12: 5.8k 3.3k 0 0 0 0 0 0 3.7k 499 8 7 0 0 0 499 0 0
I am not entirely sure how to interpret these.
As far as I can see, this is fairly unlikely to be a problem with fabric itself, and most of those problems would inidicate some sort of physical iussues with transport (i.e. fibre, patch panel, sfp, HBA in a blade, etc.).
Is there any web based error translator, which could help with identyfying problem based on the combination of errors coming up on port?
I tend to resolve my problems myself but am bit desparate and in this particular case, would it make sense to raise it with vendor support at all - in this case switch rather than blade (these are 5470s blade sweitches rebadged by IBM, running v7.4.1e)?
02-20-2017 02:52 PM - edited 02-20-2017 02:53 PM
No this looks more lake a bad physical problem. "Enc out" means encoding errors out of frames. Where is this port 12 located and what is connected to this port?
02-20-2017 10:04 PM
Port is a part of Brocade 5470 blade switch with FOS 7.4.1e.(in Blade Center H chassis), with a IBM HS22 blade connected to it. This is internal port, part of the blade switch plugged into the chassis.
It seems that this issue exist on one fabric only (its OK on the similar switch on the other fabric).
02-21-2017 12:38 AM
02-21-2017 12:48 AM
02-21-2017 01:06 AM
02-21-2017 05:01 AM - edited 02-21-2017 05:01 AM
Two things to look at: