Loading...
Loading...
Scenario This problem can occur when an SDC sends a zero-write (a write I/O where the data payload is all zeros) to an SDS. The inflight checksum is enabled in the system. Symptoms SDS disconnects from an MDM, then reconnect. The "exp.0" file shows the following panic string: 02/03 14:58:34.740139 Panic in file /data/builds/workspace/ScaleIO-Common-Job@2/src/tgt/ioh/ioh.c, line 2227, function ioh_PrepareDataAndChksmForWrite, PID 2648.Panic Expression !(((((*pReqFlags) & (0x8)) != 0) ? ((BOOL)1) : ((BOOL)0))) PANIC_ID_tgt_1497349764087. /opt/emc/scaleio/sds/bin/sds-3.0.1000.208(mosDbg_PanicPrepare+0x135) [0x8dd885] /opt/emc/scaleio/sds/bin/sds-3.0.1000.208(ioh_PrepareDataAndChksmForWrite+0x22d) [0x5f938d] /opt/emc/scaleio/sds/bin/sds-3.0.1000.208(iohComb_WritePrimary+0x644) [0x5fef84] /opt/emc/scaleio/sds/bin/sds-3.0.1000.208(ioh_Write+0x703) [0x601fe3] /opt/emc/scaleio/sds/bin/sds-3.0.1000.208(ioh_NewRequest+0x42b5) [0x606995] /opt/emc/scaleio/sds/bin/sds-3.0.1000.208(contNet_RecvIORequest+0x28d) [0x609a9d] /opt/emc/scaleio/sds/bin/sds-3.0.1000.208(netRecvGroup_WaitForWork+0x38f) [0x6cec7f] /opt/emc/scaleio/sds/bin/sds-3.0.1000.208(netRecvGroup_WaitForWorkLoop+0x18) [0x6cee08] /opt/emc/scaleio/sds/bin/sds-3.0.1000.208(mosUmt_StartFunc+0x7a) [0x717d5a] /lib64/libc.so.6(+0x48190) [0x7fcb84ece190] Impact Depending on the state of the system, it can temporarily go to either DEGRADED or DATA_FAILED state.
When the inflight checksum is enabled in the system, SDCs calculate and attach a checksum to each write I/O. If a write request happens to be all-zeros, the checksum might not be calculated (invalid) which trigger an SDS panic - it expects a valid checksum for all writes, even an all-zeros write.
Disable the inflight checksum mechanism.
Click on a version to see all relevant bugs
Dell Integration
Learn more about where this data comes from
Bug Scrub Advisor
Streamline upgrades with automated vendor bug scrubs
BugZero Enterprise
Wish you caught this bug sooner? Get proactive today.