Cisco MDS oversubscription

[karimassem]

Hi ,

I have an MDS 9513 switch

I've been having some performance issues and I'd like to rule out port group oversubscription. When I look in device manager and check oversubscription right now it looks fine. I'd like to use fabric manager to run a report that shows how much the  port groups have been utilized in the last 24 hours.

or any other way to check the history of port group oversubscription

in device manager I can right click and how oversubscription and that will give me a real time snapshot, I'm looking for a way to run a report that shows the history .

Thanks

Karim

[dynamoxxx]

are you running Cisco fabric manager server (licensed) ? It records historical information in mySQL database that can be viewed with fabric manager web client.

[Kris Vandecruys]

as Dynamoxxx mentioned, you can find this information here: http://localhost/s323.do

substitute localhost for the relevant hostname of your FMS of course

[mattkauffmann]

Karim,

this may be completely off the mark but we've had two 9513's in the last 6 weeks that we've had to completely replace because of Xbar Super frame errors.  The symptom is odd slowness between hosts and storage.  You may want to run this command on your switches and see if your seeing any SF errors.  A couple here or there might not be a problem but on one switch were getting thousands and on another switch we were getting a few every few minutes but on every module the symptoms are the same either way. One of the swtiches we've had for almost 3 years but it started having this problem in March of 2010.  We didn't find out until this weekend when we started putting storage ports on it so it seems to mainfest itself after you get to some sort of traffic level.

sh logging onboard starttime dd/mm/yy-hh:mm:ss error-stats

this should come back clean - nothing on any moduels. If you see this

------------------------------------------------------------------------------
ERROR STATISTICS INFORMATION FOR DEVICE ID 59 DEVICE Skyline-xbar
------------------------------------------------------------------------------
                                   |                |    Time Stamp   |In|Port
    Error Stat Counter Name        |    Count       |MM/DD/YY HH:MM:SS|st|Rang
                                   |                |                 |Id|e
------------------------------------------------------------------------------
FI1_CNT_BAD_SF                     |2b              |01/20/11 21:55:20|00|1-24
FI1_CNT_BAD_DI_ERR_SF              |28              |01/20/11 21:55:20|00|1-24
FI1_CNT_AR_BAD_SF                  |2b              |01/20/11 21:55:20|00|1-24

You have a signifigant hardware problem with your switch and the only fix we've found so far is wholesale chassis replacement. Modules, chassis, power supplies, the whole thing.  Just a thought, it's easy enough to check.

sho hardware internal packet-dropped-reason - will output this if there are issues

Module: 01       Dropped Packets: YES

+-----------+---------------+-------------------+------------------------------+
|           |               |      DROPS        |                              |
|   PORTS   |  DEVICE NAME  |-------------------|          COUNTER NAME        |
|           |               | Rx(Hex) | Tx(Hex) |                              |
+-----------+---------------+---------+---------+------------------------------+
|1-24       |Skyline-xbar   |    -    |5        |AM_ACC_AR_ERR                 |
|           |               |    -    |5        |TOTAL                         |
+-----------+---------------+---------+---------+------------------------------+
|1-6        |Skyline-ni     |    -    |1        |BM_ERR_QUEUE_DROP_PKT         |
|           |               |    -    |1        |TOTAL                         |
+-----------+---------------+---------+---------+------------------------------+
|1-24       |Skyline-xbar   |    -    |5        |FI1_CNT_BAD_SF                |
|           |               |    -    |5        |FI1_CNT_AR_BAD_SF             |
|           |               |    -    |a        |TOTAL                         |
+-----------+---------------+---------+---------+------------------------------+
|7-12       |Skyline-ni     |    -    |2        |BM_ERR_QUEUE_DROP_PKT         |
|           |               |    -    |2        |TOTAL                         |
+-----------+---------------+---------+---------+------------------------------+
|19-24      |Skyline-ni     |    -    |2        |BM_ERR_QUEUE_DROP_PKT         |
|           |               |    -    |2        |TOTAL                         |
+-----------+---------------+---------+---------+------------------------------+
|20         |Stratosphere   |    -    |1        |FCP_CNTR_LAF_C3_TIMEOUT_FRAMES_DISCARD|
|           |               |    -    |1        |TOTAL                         |