Hi,
I have another lustre error I'm trying to debug. Again the system is a
2.6.15 kernel with Lustre 1.4.6.
When I run tar -cvf /local/foo.tar /lustre/user/job0815/ it works for
a while and then I get:
OST Errors:
--------------------
[3458604.789746] LustreError: 4193:0:(filter_io.c:373:filter_preprw_read()) io
error -5
[3458606.258634] LustreError: 4213:0:(filter_io.c:373:filter_preprw_read()) io
error -5
[3458606.267275] LustreError: 4213:0:(filter_io.c:373:filter_preprw_read())
previously skipped 2 similar messages
[3458609.444835] LustreError: 4193:0:(filter_io.c:373:filter_preprw_read()) io
error -5
[3458609.453534] LustreError: 4193:0:(filter_io.c:373:filter_preprw_read())
previously skipped 5 similar messages
[3458616.005135] LustreError: 4189:0:(filter_io.c:373:filter_preprw_read()) io
error -5
[3458616.013779] LustreError: 4189:0:(filter_io.c:373:filter_preprw_read())
previously skipped 7 similar messages
Client Syslog:
-----------------------------
Nov 28 17:28:53 ni-01-01 kernel: [533212.314689] LustreError:
10462:0:(client.c:577:ptlrpc_check_status()) @@@ type == PTL_RPC_MSG_ERR, err
== -5 [EMAIL PROTECTED] x462629/t0 o3->[EMAIL PROTECTED]:28 lens 328/280 ref 2
fl Rpc:R/0/0 rc 0/-5
Nov 28 17:28:54 ni-01-01 kernel: [533213.109922] LustreError:
10462:0:(client.c:577:ptlrpc_check_status()) @@@ type == PTL_RPC_MSG_ERR, err
== -5 [EMAIL PROTECTED] x463031/t0 o3->[EMAIL PROTECTED]:28 lens 328/280 ref 2
fl Rpc:R/0/0 rc 0/-5
Nov 28 17:28:54 ni-01-01 kernel: [533213.176032] LustreError:
10462:0:(client.c:577:ptlrpc_check_status()) @@@ type == PTL_RPC_MSG_ERR, err
== -5 [EMAIL PROTECTED] x463059/t0 o3->[EMAIL PROTECTED]:28 lens 328/280 ref 2
fl Rpc:R/0/0 rc 0/-5
Nov 28 17:28:55 ni-01-01 kernel: [533213.795154] LustreError:
10462:0:(client.c:577:ptlrpc_check_status()) @@@ type == PTL_RPC_MSG_ERR, err
== -5 [EMAIL PROTECTED] x463220/t0 o3->[EMAIL PROTECTED]:28 lens 328/280 ref 2
fl Rpc:R/0/0 rc 0/-5
Nov 28 17:28:55 ni-01-01 kernel: [533214.209407] LustreError:
10462:0:(client.c:577:ptlrpc_check_status()) @@@ type == PTL_RPC_MSG_ERR, err
== -5 [EMAIL PROTECTED] x463335/t0 o3->[EMAIL PROTECTED]:28 lens 328/280 ref 2
fl Rpc:R/0/0 rc 0/-5
Nov 28 17:28:55 ni-01-01 kernel: [533214.465867] LustreError:
10462:0:(client.c:577:ptlrpc_check_status()) @@@ type == PTL_RPC_MSG_ERR, err
== -5 [EMAIL PROTECTED] x463449/t0 o3->[EMAIL PROTECTED]:28 lens 328/280 ref 2
fl Rpc:R/0/0 rc 0/-5
Nov 28 17:28:55 ni-01-01 kernel: [533214.633638] LustreError:
10462:0:(client.c:577:ptlrpc_check_status()) @@@ type == PTL_RPC_MSG_ERR, err
== -5 [EMAIL PROTECTED] x463503/t0 o3->[EMAIL PROTECTED]:28 lens 328/280 ref 2
fl Rpc:R/0/0 rc 0/-5
Nov 28 17:28:56 ni-01-01 kernel: [533214.978952] LustreError:
10462:0:(client.c:577:ptlrpc_check_status()) @@@ type == PTL_RPC_MSG_ERR, err
== -5 [EMAIL PROTECTED] x463679/t0 o3->[EMAIL PROTECTED]:28 lens 328/280 ref 2
fl Rpc:R/0/0 rc 0/-5
Nov 28 17:28:58 ni-01-01 kernel: [533216.982179] LustreError:
10462:0:(client.c:577:ptlrpc_check_status()) @@@ type == PTL_RPC_MSG_ERR, err
== -5 [EMAIL PROTECTED] x464427/t0 o3->[EMAIL PROTECTED]:28 lens 328/280 ref 2
fl Rpc:R/0/0 rc 0/-5
Nov 28 17:28:58 ni-01-01 kernel: [533217.005021] LustreError:
10462:0:(client.c:577:ptlrpc_check_status()) previously skipped 1 similar
messages
Nov 28 17:28:59 ni-01-01 kernel: [533218.202467] LustreError:
10462:0:(client.c:577:ptlrpc_check_status()) @@@ type == PTL_RPC_MSG_ERR, err
== -5 [EMAIL PROTECTED] x465117/t0 o3->[EMAIL PROTECTED]:28 lens 328/280 ref 2
fl Rpc:R/0/0 rc 0/-5
Nov 28 17:28:59 ni-01-01 kernel: [533218.225331] LustreError:
10462:0:(client.c:577:ptlrpc_check_status()) previously skipped 1 similar
messages
Nov 28 17:29:02 ni-01-01 kernel: [533221.130829] LustreError:
10462:0:(client.c:577:ptlrpc_check_status()) @@@ type == PTL_RPC_MSG_ERR, err
== -5 [EMAIL PROTECTED] x466182/t0 o3->[EMAIL PROTECTED]:28 lens 328/280 ref 2
fl Rpc:R/0/0 rc 0/-5
Nov 28 17:29:02 ni-01-01 kernel: [533221.153689] LustreError:
10462:0:(client.c:577:ptlrpc_check_status()) previously skipped 2 similar
messages
Stopping tar and running "cat file_with_error >/dev/null" on the same
host works fine after the error. So I assume it is not a physical
drive giving an error or filesystem error on the medium or something.
Any tips?
MfG
Goswin
_______________________________________________
Lustre-discuss mailing list
[email protected]
https://mail.clusterfs.com/mailman/listinfo/lustre-discuss
