No ends in sight

24 03 2007

Firstly, thanks to Sangod for his comments. The morning after I wrote the piece I was showing my new spangly blog to a team mate, saying how I found other storage blogs useful, went to Sangod’s site and there was a reference to my post! It made me feel very special (not “Special Needs” as my co-workers commented!) However my joy was brief.

My long distance relationship is now over. It’s now shorter, thanks to having having both DWDM routes at 15km, which may occasionally diverge if we’re lucky. It’s an accident just waiting to happen. At least all my links are up, even if I did have to travel to the remote site to swap a tx/rx connection around. But those shiny green lights on the ports on the Brocade 48000s are as deceptive as Gordon Brown’s latest budget (don’t get me started on how he’s cut the basic rate of tax yet the lowest paid in our society will be worse off). The links are up, I have 2 x 2gb dynamically trunked paths on each fabric, the switches see each other, but can anyone see their remote storage? Our remote USP array can see the WWN of the remote servers, but any attempt to access that storage times out.

I see the ‘encoding errors outside of frame’ count go up faster than my blood pressure, which indicates to me that the connections are about as crap as the customer service from my ISP. We have DWDM links to another remote site provided by another telecoms company that are mostly reliable and after aeons of up time those ‘enc_out’ errors number are around 3,000, far less than the 300,000+ in a couple of hours seen on the new links. If anyone reads this and has any thoughts on this subject, I’d appreciate the input.

But that’s not the end of it. In another location I have one half of a Brocade 12000 director preventing us from propagating any configuration changes. Even an ‘alishow’ request is ignored with a ‘fabric busy’ message. The other half of the 12000 is fine. This leads me nicely into another area of irritation. Firstly, it irks me that I have to get support from the re-seller rather than direct from Brocade (though I believe if we pay them enough now they’ll help out), especially as our fabrics have built up over time so switches have come from different re-sellers. But then they re-badge them. To me, it’s a Brocade 12000. So to the re-seller, it should be a Brocade 12000, not an HP whatever or IBM thingumyjig (at least HDS supply them vanilla). Anyhoo, I’ve had a call logged for about 3 days now, still no solution, last I heard the problem is relaxing somewhere on the Indian sub-continent. At least if I end up power cycling the director then only fabric is affected and data can be reached on the other, unlike another site that had each half of the 12000 on a different fabric. It certainly hit the fan when that director crashed recently. Roll on hardware refresh time, I like the Brocade switches, but the 12000 is certainly not their best design.

On a personal note, got a letter today saying I could be a match for someone needing a bone marrow transplant, so some more test are needed. If it comes to it, donating marrow or stem cells is a tad more invasive than your average blood donation. I knew this when I signed up, but seeing this now in writing puts it into perspective. A little scary.




Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: