Legacy HDS Forums

Server hung due to path errors - no errors seen on SAN or USP

Discussion created by Legacy HDS Forums on Mar 14, 2009
Latest reply on Apr 15, 2009 by Legacy HDS Forums

Originally posted by: PP BIJU KRISHNAN



Hi Gurus,

I'm tired of troubleshooting this problem. One of our AIX hosts hangs abruptly freezing the application and reports excessive IO errors at this time. Initially the host had trouble bringing up one path, although IBM sais that errors were from SAN, I insisted and got the HBA and cables replaced. This fixed the path issue, but one day later the errors re-occured and the system freezed.

The USP port that the host connects to has other hosts which are NOT complaining at all. The switch ports show a high number of enc outs but this could be generated at time of connection. I have done a portstats clear and observed the ports of an hour and see no errors.

I'm not sure if the IO board of this LPAR has issues since I'm NOT a p series admin.

Any hints highly appreciated.

Mar 13 13:01:24 host2190 user:err|error syslog: KAPL08019-E The path (0x000A0011) detected an error (0x0000004E). (0x00000000)
Mar 13 13:01:24 host2190 user:err|error syslog: KAPL08019-E The path (0x000A0011) detected an error (0x0000004E). (0x00000000)
Mar 13 13:01:24 host2190 user:err|error syslog: KAPL08019-E The path (0x000A0011) detected an error (0x0000004E). (0x00000000)
Mar 13 13:03:22 host2190 user:err|error syslog: KAPL08022-E A path error occurred. ErrorCode = 0000004E, PathID = 17, PathName = 08.05.000000000043CF00.0008, DNum = 0, HDevName = dlmfdrv8
Mar 13 13:03:22 host2190 user:err|error syslog: KAPL08022-E A path error occurred. ErrorCode = 0000004E, PathID = 17, PathName = 08.05.000000000043CF00.0008, DNum = 0, HDevName = dlmfdrv8
Mar 13 13:03:22 host2190 user:err|error syslog: KAPL08022-E A path error occurred. ErrorCode = 0000004E, PathID = 17, PathName = 08.05.000000000043CF00.0008, DNum = 0, HDevName = dlmfdrv8
Mar 13 13:03:22 host2190 user:err|error syslog: KAPL08019-E The path (0x000A000D) detected an error (0x0000004E). (0x00000000)
Mar 13 13:03:22 host2190 user:err|error syslog: KAPL08019-E The path (0x000A000D) detected an error (0x0000004E). (0x00000000)
Mar 13 13:03:22 host2190 user:err|error syslog: KAPL08019-E The path (0x000A000D) detected an error (0x0000004E). (0x00000000)

The WIO is high during this problem

00:00:00    %usr    %sys    %wio   %idle   physc   %entc

13:15:01      44       5      51       0    0.21    52.7
13:20:01      11       5      79       5    0.07    18.0
13:25:01      55      11      30       4    0.27    67.8
13:30:01      88      11       0       0    0.99   246.9
13:35:01      88      12       0       0    0.99   248.3

Outcomes