ask them to give you output from "powermt display dev=all" as well as "lsdev -Cc disk"

What is the multipathing software installed on the host? Is it powerpath?

Can you check if the switch was swapped out?

thank you dynamox. could you tell me what exactly the Q-IOs error mean.

not a problem anymore but it does tell you that there were errors in the past. It could be anything, physical connectivity issues (HBA going bad, cable going bad, patch panel port going bad), to logical configuration issues (somebody removed a zone, disable the wrong port on the switch). At this point you don't have any dead paths so things are ok but keep watching the error count ..if it's increasing you will need to start troubleshooting different components.

lsdev -cc disk output:

# Output Filename : lsdev_-Cc_disk.txt
# Date : Wed Jul  4 17:44:21 GMT 2012
################################################

hdisk0      Available C6-T1-01 EMC INVISTA FCP Disk
hdisk1      Available C6-T1-01 EMC INVISTA FCP Disk
hdisk2      Available C6-T1-01 EMC INVISTA FCP Disk
hdisk3      Available C6-T1-01 EMC INVISTA FCP Disk
hdisk4      Available C6-T1-01 EMC INVISTA FCP Disk
hdisk5      Available C6-T1-01 EMC INVISTA FCP Disk
hdisk6      Available C6-T1-01 EMC INVISTA FCP Disk
hdisk7      Available C6-T1-01 EMC INVISTA FCP Disk
hdisk8      Available C6-T1-01 EMC INVISTA FCP Disk
hdisk9      Available C6-T1-01 EMC INVISTA FCP Disk
hdisk10     Available C6-T1-01 EMC INVISTA FCP Disk
hdisk11     Available C6-T1-01 EMC INVISTA FCP Disk
hdisk12     Available C6-T1-01 EMC INVISTA FCP Disk
hdisk13     Available C6-T1-01 EMC INVISTA FCP Disk
hdisk14     Available C6-T1-01 EMC INVISTA FCP Disk
hdisk15     Available C6-T1-01 EMC INVISTA FCP Disk
hdisk16     Available C6-T1-01 EMC INVISTA FCP Disk
hdisk17     Available C6-T1-01 EMC INVISTA FCP Disk
hdisk18     Available C8-T1-01 EMC INVISTA FCP Disk
hdisk19     Available C8-T1-01 EMC INVISTA FCP Disk
hdisk20     Available C8-T1-01 EMC INVISTA FCP Disk
hdisk21     Available C8-T1-01 EMC INVISTA FCP Disk
hdisk22     Available C8-T1-01 EMC INVISTA FCP Disk
hdisk23     Available C8-T1-01 EMC INVISTA FCP Disk
hdisk24     Available C8-T1-01 EMC INVISTA FCP Disk
hdisk25     Available C8-T1-01 EMC INVISTA FCP Disk
hdisk26     Available C8-T1-01 EMC INVISTA FCP Disk
hdisk27     Available C8-T1-01 EMC INVISTA FCP Disk
hdisk28     Available C8-T1-01 EMC INVISTA FCP Disk
hdisk29     Available C8-T1-01 EMC INVISTA FCP Disk
hdisk30     Available C8-T1-01 EMC INVISTA FCP Disk
hdisk31     Available C8-T1-01 EMC INVISTA FCP Disk
hdisk32     Available C8-T1-01 EMC INVISTA FCP Disk
hdisk33     Available C8-T1-01 EMC INVISTA FCP Disk
hdisk34     Available C8-T1-01 EMC INVISTA FCP Disk
hdisk35     Available C8-T1-01 EMC INVISTA FCP Disk
hdisk36     Available C6-T1-01 EMC INVISTA FCP Disk
hdisk37     Available C6-T1-01 EMC INVISTA FCP Disk
hdisk38     Available C8-T1-01 EMC INVISTA FCP Disk
hdisk39     Available C8-T1-01 EMC INVISTA FCP Disk
hdiskpower0 Available C6-T1-01 PowerPath Device
hdiskpower1 Available C6-T1-01 PowerPath Device
hdiskpower2 Available C6-T1-01 PowerPath Device
hdiskpower3 Available C6-T1-01 PowerPath Device
hdiskpower4 Available C6-T1-01 PowerPath Device
hdiskpower5 Available C6-T1-01 PowerPath Device
hdiskpower6 Available C6-T1-01 PowerPath Device
hdiskpower7 Available C6-T1-01 PowerPath Device
hdiskpower8 Available C6-T1-01 PowerPath Device
hdiskpower9 Available C6-T1-01 PowerPath Device

those are not Q-IOs (queued io), that last column is errors (why your system admin complained in the first place)

they have recieved an error "EMCP_BUS_DEAD" for this server and they suspect it is to do something with the devs.

Jithu9288 wrote: will a powermt restore will reset the errors?

yes

Hi ankit, how should i check it?

post 7.

also pls check the Q-IOs listed in the powermt output and what exactly it mean? for two paths it shows 3 and 5 respectively. so there were a total of 8 IOs outstanding?

will a powermt restore will reset the errors?

• When powermt display hba is used, the Stats Q-IO displays the current total number of outstanding I/Os on this HBA.
• When powermt display dev=all is used, the Stats Q-IO displays the total number of I/O operations under way to this path. This is the total number of I/O requests to this device that have not completed.
• The sum of in-progress I/Os for all paths should equal the number of in-progress I/Os for the PowerPath device.

You can refer to Knowledge Base Article emc252687.

Sorry buddy, lost you there... What do you want to check? (Please, quote which post you are referring to)

sample powerpath output. Q-IOs it showing 3 and 5 is that a problem?

==============================================================================

--------------- Host ---------------   - Stor -   -- I/O Path --  -- Stats ---

###  HW Path               I/O Paths    Interf.   Mode    State   Q-IOs Errors

==============================================================================

   0 fscsi0                   hdisk11   1E        active  alive       0      3

   1 fscsi1                   hdisk20   17        active  alive       0      0

   1 fscsi1                   hdisk29   0B        active  alive       0      0

   0 fscsi0                   hdisk2    02        active  alive       0      5

What is the status of AIX resetdrives parameter? Its Enabled or Disabled?

Can you check  if the devices have dynamic tracking enabled or disabled?  That is, dyntrk=yes. or no?