Transient aio exhaustion causes multipath errors
Last updated on NOVEMBER 14, 2012
Applies to:
Linux OS - Version Oracle Linux 5.5 and laterLinux x86
Linux x86-64
Symptoms
The error message "io_setup failed" is sometimes visible in the output from "multipath-ll", and the associated paths are listed with state 'faulty' in the topology section, as follows:
pdcnap901_mnhprdb_dta01 (360a98000646743332f4a6b374f437356) dm-32 NETAPP,LUN [size=700G][features=1 queue_if_no_path][hwhandler=0][rw] \_ round-robin 0 [prio=0][active] \_ 3:0:1:2 sdd 8:48 [active][faulty] \_ 5:0:7:2 sdfb 129:208 [active][faulty] \_ 3:0:7:2 sdhh 133:112 [active][faulty] \_ 5:0:4:2 sdu 65:64 [active][faulty] \_ round-robin 0 [prio=2][enabled] \_ 3:0:2:2 sdbc 67:96 [active][ready] \_ 5:0:5:2 sdbv 68:144 [active][ready] \_ 5:0:6:2 sddy 128:0 [active][faulty] \_ 3:0:6:2 sdge 131:160 [active][faulty]
This is especially common on Fibre-connected NetApp storage.
Cause
Sign In with your My Oracle Support account |
|
Don't have a My Oracle Support account? Click to get started |
My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms