Hi to all
We are currently encountering disconnections fromt our VM host to VNX storage. Upon checking fabriclogs on dcx 8510 Director switch,
we are recieving the following logs
06:52:52.599972 SCN LR_PORT(0);g=0x0 LR_OUT D2,P0 D2,P0 92 NA06:52:56.599866 SCN LR_PORT(0);g=0x0 LR_OUT D2,P0 D2,P0 92 NA06:53:00.599674 SCN LR_PORT(0);g=0x0 LR_OUT D2,P0 D2,P0 92 NA06:53:04.599934 SCN LR_PORT(0);g=0x0 LR_OUT D2,P0 D2,P0 92 NA06:55:45.600322 SCN LR_PORT(0);g=0x0 LR_OUT D2,P0 D2,P0 92 NA06:55:49.601108 SCN LR_PORT(0);g=0x0 LR_OUT D2,P0 D2,P0 92 NA
Please enlighten me what is the meaning of the fabric log above.
Below is the SFP show for index 92 which is slot 10 port 12
=============Identifier: 3 SFPConnector: 7 LCTransceiver: 700c406000000000 4,8,16_Gbps M5,M6 sw Inter,Short_distEncoding: 6 64B66BBaud Rate: 140 (units 100 megabaud)Length 9u: 0 (units km)Length 9u: 0 (units 100 meters)Length 50u (OM2): 4 (units 10 meters)Length 50u (OM3): 0 (units 10 meters)Length 62.5u:2 (units 10 meters)Length Cu: 0 (units 1 meter)Vendor Name: BROCADEVendor OUI: 00:05:1eVendor PN: 57-0000088-01Vendor Rev: AWavelength: 850 (units nm)Options: 003a Loss_of_Sig,Tx_Fault,Tx_DisableBR Max: 0BR Min: 0Serial No: HAA215131001NG5Date Code: 150327DD Type: 0x68Enh Options: 0xfaStatus/Ctrl: 0x82Pwr On Time: 3.00 years (26313 hours)E-Wrap Control: 0O-Wrap Control: 0Alarm flags[0,1] = 0x0, 0x40Warn Flags[0,1] = 0x0, 0x40Alarm Warnlow high low highTemperature: 52 Centigrade -5 85 0 75Current: 7.310 mAmps 2.500 12.000 2.000 11.500Voltage: 3309.6 mVolts 3000.0 3600.0 3130.0 3460.0RX Power: -4.6 dBm (347.3uW) 31.6 uW 1258.9 uW 31.6 uW 794.0 uWTX Power: -2.4 dBm (573.1 uW) 126.0 uW 1258.9 uW 251.0 uW 794.0 uW
and the portshow
portIndex: 92portName: VNX5800_VM_SPB3portHealth: HEALTHY
Authentication: NoneportDisableReason: NoneportCFlags: 0x1portFlags: 0x20b03 PRESENT ACTIVE F_PORT G_PORT LOGICAL_ONLINE LOGIN NOELP ACCEPTLocalSwcFlags: 0x0portType: 24.0portState: 1OnlineProtocol: FCportPhys: 6In_Sync portScn: 32F_Portport generation number: 0state transition count: 0
portId: 0b5c00portIfId: 43a2080bportWwn: 20:5c:50:eb:1a:72:4e:00portWwn of device(s) connected:50:06:01:6b:08:60:3a:12Distance: normalportSpeed: N8Gbps
FEC: InactiveCredit Recovery: InactiveAoq: InactiveFAA: InactiveF_Trunk: InactiveLE domain: 0Peer beacon: OffFC Fastwrite: OFFInterrupts: 0 Link_failure: 0 Frjt: 0Unknown: 0 Loss_of_sync: 0 Fbsy: 0Lli: 625 Loss_of_sig: 0Proc_rqrd: 4706 Protocol_err: 0Timed_out: 0 Invalid_word: 0Rx_flushed: 0 Invalid_crc: 0Tx_unavail: 0 Delim_err: 0Free_buffer: 0 Address_err: 0Overrun: 0 Lr_in: 19Suspended: 0 Lr_out: 284Parity_err: 0 Ols_in: 02_parity_err: 0 Ols_out: 0CMI_bus_err: 0
here also is the porterrshow
frames enc crc crc too too bad enc disc link loss loss frjt fbsy c3timeout pcstx rx in err g_eof shrt long eof out c3 fail sync sig tx rx err
92: 2.0g 3.3g 0 0 0 0 0 0 0 34.1k 0 0 0 0 0 34.0k 35 0
It's really difficult to read the porterrshow. Are the two 34K counter c3 disc and c3timeout? If so, this could indicate a slow drain device, not necessarily the device attached to this port. I recommend a SAN Health report to start.
Yes ma'am / sir. it is the c3disc and c3timeout. i already uploaded san health with 5 minutes duration. still waiting for the output.BTW what do you mean not necessarily attached to this port? Is it possible that the cause is the server hba zoned to this port and not directly the storage(vnx) attached to this port? Thanks.
The fabriclog --show events of LR_OUT to port 92 are indicative of the SAN switch port 92 not
receiving a timely receipt of R_Rdy primitives from the host HBA. The switch port, if deprived
long enough of R-Rdy to replenish the port buffers will issue a link reset (LR_OUT) to
re-establish buffers. I agree with earlier comments that you should be investigating the
host HBA and its applications which are driving the i/o.