mcdown75

[Log in to get rid of this advertisement]

First, if this is the wrong place, I apologize. I am still very green at this. Here is the situation.

We have three servers, a PowerEdge 1800 and two PowerEdge 2650's. We also have several virtual machines running on these servers. The four vm's we are concerned with are as follows: Srv01(PE1800)- vmDB01 Srv02(PE2650)-vmDB02 and vmDB04 Srv03(PE2650)-vmDB03. vmDB01 has the original Oracle Database (DB01) that was built. DB02, DB03 and DB04 were cloned from it. We also run a 1Gb network.

That all well and good, so here is the problem. vmDB01 has close to a 1Gb i/o. vmDB02, vmDB03, and vmDB04 all are running at around 25-50 Mb/sec i/o. A marked decrease to say the least.

Our experienced but overworked Linux guru is convinced that there is a driver missing from Srv02 and Srv03. He is however dedicated to other duties and is trying to walk me through figuring it out as best he can. The only change I can find is that the only vm's affected by this slow i/o on Srv02 and Srv03 are the ones running the Oracle DB's. The other vm's on Srv02 and Srv03 have close to 1Gb i/o.

I have combed the internet for the last two days trying to find even a hint of how to diagnose this or fix it to no avail. Can anyone help me?