ノードがパニック状態になる「giveback or ARL hung in replication_engine while doing_precOMMIT in SK process」が発生して停止する
環境
- ONTAP 9
- FAS / AFF
問題
ノードが停止し、次のソフトウェアパニック文字列とコアダンプが報告されます。
PANIC: giveback or arl hung in replication_engine while doing SENDHOME_DOING_PRECOMMIT in SK process sendhome_hang_detector
on release 9.8P12 (C) on Sun Thu May 12 20:07:27 EDT 2022
compile flags: x86_64.optimize
HA: current time (in sk_msecs) 2403764 (in sk_cycles) 6227546366396
DUMPCORE: START
Dumping to disks: 0d.00.15P2
coredump: Using block 461193217 instead of block 461193216 as the sparecore abort block
...........................................................................................................................
...........................................................................................................................
...........................................................................................................................
...........................................................................................................................
DUMPCORE: END -- coredump written
coredump: compression ms 315347 compressions 209628 bytes in 566686916608 bytes out 53045401980 ratio 1068%
comp BW 1713MB/s
coredump: write ms 51529 write count 15376 write bytes 53047083008 write BW 89MB/s
coredump: total ms 570912 chunks of all zeroes 23715