Lobotomy. Sometimes I think that is what some folks have had.
Not sure how much of this you will understand but...
Wednesday I caught a case where a customer is running a backup job and it fails cause it causes his remote fibre disks to fail over from one controller to another. It fails over when he ran some info gathering tool that collects stats on the system. First thing I wonder is why the heck he is running info gathering during a backup, but...
Anyhoo there were supposedly traces of this whole thing for me to look at. After 2 days of waiting for traces for a supposedly critical call I get them and check them out on Saturday and confirm that the info tool is indeed causing the failover. It sends the fibre controllers on the storage all willy-nilly. The storage in question is crap - lots of problems - and not responding to the query like it is supposed to. So I write it up and send the analysis. This morning after arguing with some idiots about the analysis and why I am wrong
I find out that
a) traces are not from the customer but from a test lab
b) customer's problem is a backup failing - he is not running the info gathering tool
c) test lab couldn't reproduce the issue with just the backup SW so they added in the info gathering tool. So I spent time on the weekend looking at traces for the wrong problem.
I told them come back to me when you can either reproduce it or get an actual trace from the customer site. The response was we can't - customer won't let us touch the equipment any more. I didn't say well that's cause you were steppin' on the old crank in your golf shoes the past month. I did say that I couldn't help any more until I got traces of the actual problem.
Not sure how much of this you will understand but...
Wednesday I caught a case where a customer is running a backup job and it fails cause it causes his remote fibre disks to fail over from one controller to another. It fails over when he ran some info gathering tool that collects stats on the system. First thing I wonder is why the heck he is running info gathering during a backup, but...
Anyhoo there were supposedly traces of this whole thing for me to look at. After 2 days of waiting for traces for a supposedly critical call I get them and check them out on Saturday and confirm that the info tool is indeed causing the failover. It sends the fibre controllers on the storage all willy-nilly. The storage in question is crap - lots of problems - and not responding to the query like it is supposed to. So I write it up and send the analysis. This morning after arguing with some idiots about the analysis and why I am wrong
I find out that a) traces are not from the customer but from a test lab
b) customer's problem is a backup failing - he is not running the info gathering tool
c) test lab couldn't reproduce the issue with just the backup SW so they added in the info gathering tool. So I spent time on the weekend looking at traces for the wrong problem.
I told them come back to me when you can either reproduce it or get an actual trace from the customer site. The response was we can't - customer won't let us touch the equipment any more. I didn't say well that's cause you were steppin' on the old crank in your golf shoes the past month. I did say that I couldn't help any more until I got traces of the actual problem.



!) to tell them what I thought of them.
Comment