Re: Linux Dev and Device Relationships

2003-02-21 Thread Gordon Messmer
On Thu, 2003-02-20 at 08:30, Robert Vaughn wrote: > The problem that I'm trying to solve is where SAR -d > lists only the following device names... I don't get a full list from 'sar -d' either. Could be a bug in sadc or something... The web page notes some bugfixes in versions newer than what wa

Re: Linux Dev and Device Relationships

2003-02-20 Thread Robert Vaughn
You're right. The device count is five. I get confused with this whole zero number thing. ...Robert --- Bret Hughes <[EMAIL PROTECTED]> wrote: > On Thu, 2003-02-20 at 10:30, Robert Vaughn wrote: > > Thanks for the response. I'm still a little lost. > > > You're right on the relationship betwe

Re: Linux Dev and Device Relationships

2003-02-20 Thread Bret Hughes
On Thu, 2003-02-20 at 10:30, Robert Vaughn wrote: > Thanks for the response. I'm still a little lost. > You're right on the relationship between major/minor > and device names such as sda, sda1, sdb, etc. > > The system that we've been working on is connected to > an EMC SAN. It uses Redhat AS

Re: Linux Dev and Device Relationships

2003-02-20 Thread Robert Vaughn
Thanks for the response. I'm still a little lost. You're right on the relationship between major/minor and device names such as sda, sda1, sdb, etc. The system that we've been working on is connected to an EMC SAN. It uses Redhat AS 2.1 (2.4 kernel). The problem that I'm trying to solve is whe

Re: Linux Dev and Device Relationships

2003-02-19 Thread Gordon Messmer
On Wed, 2003-02-19 at 14:21, Robert Vaughn wrote: > The distinctive number does not translate 1 to 1 with > the minor number. If this were the case then when I > did the following I would get a different answer... > > ls -l /dev/sdbproduces 8,16 > > Using your theory I would expect 8,1 w

Re: Linux Dev and Device Relationships

2003-02-19 Thread Robert Vaughn
Thanks for the response. However, the major/minor theory still doesn't hold. According to 'man sar', 'm' is the major number of the device [correct, I agree], whereas 'n' is a distinctive number. The distinctive number does not translate 1 to 1 with the minor number. If this were the case then

Re: Linux Dev and Device Relationships

2003-02-18 Thread Gordon Messmer
On Tue, 2003-02-18 at 21:49, Robert Vaughn wrote: > Thanks for the quick response. We had taken a look at > the relationship between major and minor but dismissed > that method based on the following. > > In SAR we were able to identify five devices. > >From SAR... > dev8-0 > dev8-1 > dev8-2 > de

Re: Linux Dev and Device Relationships

2003-02-18 Thread Robert Vaughn
Thanks for the quick response. We had taken a look at the relationship between major and minor but dismissed that method based on the following. In SAR we were able to identify five devices. >From SAR... dev8-0 dev8-1 dev8-2 dev8-3 dev8-4 In IOSTAT five devices are identified, sda, sdb, sdc, sdd

Re: Linux Dev and Device Relationships

2003-02-18 Thread nate
Robert Vaughn said: > We are at a loss in understanding how SAR generated > device IO stats relate to IOSTAT device stats. > Specifically, SAR lists devices in the syntax of > dev8-X while IOSTAT lists devices in the syntax of > sdaX. We have not found a direct method of > correlating dev8-X to sd

Linux Dev and Device Relationships

2003-02-18 Thread Robert Vaughn
We are at a loss in understanding how SAR generated device IO stats relate to IOSTAT device stats. Specifically, SAR lists devices in the syntax of dev8-X while IOSTAT lists devices in the syntax of sdaX. We have not found a direct method of correlating dev8-X to sdaX. I understand that, for exa